High-precision data acquisition system based on MPX2100 sensor

Abstract: In the measurement and analysis of pressure, the pressure value usually changes slowly, but the measurement accuracy is required to be very high. This paper introduces a high-precision pressure data acquisition system based on the MPX2100 pressure sensor, and gives the performance characteristics of the X-type silicon pressure sensor MPX2100 and the A/D conversion core component ICL7135 and their related interface circuits; it provides a system block diagram, circuit diagrams of each main part, and A/D conversion software design flow chart.

Keywords: high precision; pressure sensor; data acquisition; A/D conversion; LCD display; single-chip microcomputer

1. Introduction

In industries and scientific research such as petroleum, chemical industry, metallurgy, electricity, textile, light industry, water conservancy, etc., relevant pressure detection and analysis must be carried out. Usually, the pressure value changes slowly, but in the process of measuring the pressure value and converting it from non-electrical quantity to electrical quantity, the accuracy is very high. This paper introduces a general high-precision pressure data acquisition system. The pressure sensor of the system uses Motorola's high-precision X-type silicon pressure sensor MPX2100, which has high conversion accuracy, high sensitivity and excellent linearity. Under the control of the high-performance single-chip AT89S52, the amplified and conditioned analog electric quantity is converted into A/D through the high-precision and high-performance chip ICL7135, which can ensure that the system has high data acquisition accuracy and strong anti-interference ability, and long service life. The system adopts LCD display and PS/2 keyboard interface to achieve good human-computer exchange. The application of PLD technology saves the cost of hardware circuit.

2. Hardware composition and working principle of the system

The block diagram of the high-precision pressure data acquisition system is shown in Figure 1. The analog signal output by the pressure sensor is amplified and conditioned, and then converted into a digital quantity by the analog/digital conversion module, and transmitted to the single-chip microcomputer. After calibration, calculation and zero point compensation, it is displayed on the LCD display module. At the same time, it can be transmitted to the host computer through the serial interface to achieve good human-computer exchange. The keyboard provides a means of human-computer interaction.

1. Pressure data acquisition and signal conditioning circuit

The pressure sensor is a device that converts pressure into current/voltage, which can be used to measure physical quantities such as pressure and displacement. There are many types of pressure sensors, among which silicon semiconductor sensors are widely used due to their small size, light weight, low cost, good performance, and easy integration. Silicon piezoresistive sensor is one of them. It uses diffusion or ion implantation to form four resistor strips with equal resistance on a silicon wafer and connects them into a Wheatstone bridge. When there is no external pressure, the bridge is in a balanced state and the bridge output is zero. When there is external pressure, the bridge loses balance and generates an output voltage. The voltage is related to the pressure. By detecting the voltage, the corresponding pressure value can be obtained. However, this sensor causes measurement errors due to the incomplete matching of the four bridge arm resistors, and the zero point offset is large and difficult to adjust.

The X-type silicon pressure sensor produced by Motorola can overcome the above shortcomings. As shown in Figure 2, unlike the Wheatstone bridge, Motorola's patented technology uses a single X-type resistor element instead of a bridge structure. Its piezoresistive element is X-shaped, so it is called an X-type pressure sensor. The X-type resistor is photoetched on the silicon diaphragm using an ion implantation process, and uses computer-controlled laser correction technology and temperature compensation technology to make the Motorola silicon MPX series pressure sensor very accurate. Its analog output voltage is proportional to the input pressure value and the power supply bias voltage, has excellent linearity, high sensitivity, and good long-term repeatability. The MPX2100DP pressure sensor in this series is a high-precision silicon pressure sensor. This system uses MPX2100DP as a pressure sensor, which can well meet the requirements of the system. It has the following characteristics:

① Due to the use of laser fine-tuning technology, the bridge zero drift output is very small, generally less than ±1mV;

② The sensor has a high sensitivity of 40mV±1.5mV;

③ The sensor consists of a thermistor temperature compensation network, which has a good temperature compensation effect in the range of -40℃~+125℃, thereby improving the accuracy of the sensor;

④ It has excellent linearity (±0.25%FS);

⑤ It has a wide operating temperature range (-40℃~+125℃);

⑥ It allows a large overload (400%).

[page]

(2) Signal conditioning circuit

The circuit diagram is shown in Figure 3 (only one sensor output signal is used as an example in the figure, and the multi-way switch CD4051 is not drawn).

The signal is converted into an electrical signal by the high-precision pressure sensor MPX2100DP, and after being selected by the CMOS type 8-to-1 multi-way switch CD4051, it is sent to the amplifier circuit, conditioned and output to the A/D module ICL7135 for high-precision analog-to-digital conversion. The power supply voltage of MPX2100DP is 8V, and its full-scale output x is determined by the following formula: x=40mV×8V/10V=32mV (1)

For the voltage output logic level set to 5V, it is obvious that the 32mV voltage needs to be properly amplified.

The amplifier uses the MC33274 quad operational amplifier, which forms an instrument amplifier circuit with high differential mode gain and high common mode rejection ratio, high input impedance, and adjustable bias circuit. The differential mode amplification is mainly completed by U1A, and U1B is a voltage follower to prevent the feedback current of the operational amplifier from flowing into the negative end of the sensor. At zero pressure, the voltage difference between terminals 2 and 4 of the sensor is zero. Assuming the common mode voltages at terminals 2 and 4 are 4V (half of the sensor power supply voltage), the terminal voltage of U1A is also 4V, which is passed through the U1C and U1D circuits to make its output zero. The zero pressure bias at the output is introduced by R4 and RP1. The value of R7 is selected so that the impedance seen from terminal 13 is about 1kW, and the gain of the amplifier is:

(2)

Selecting a gain of 125 allows the sensor's full-scale output swing of 32mV to be amplified to 125×0.032=4V (half of the power supply voltage).

2. Microcontroller and its peripheral circuits

The microcontroller peripheral circuits are shown in Figure 4.

(1) Introduction to AT89S52

The microcontroller used is the newly launched AT89S52 by ATMEL. This chip has the characteristics of low power consumption and high performance. It is an 8-bit microcontroller using CMOS technology. The main features of AT89S52 are as follows:

① It adopts ATMEL's high-density, non-volatile memory (NV-SRAM) technology;

② It has 256 bytes of RAM and 8KB of in-circuit programmable (ISP) FLASH memory on the chip;

③ There are two low-power and power-saving working modes: idle mode and power-down mode;

④ The chip contains a watchdog timer (WDT). The WDT contains a 14-bit counter and a watchdog timer reset register (WDTRST). As long as WDTRST is written to 01EH first and then 0E1H in sequence, the WDT will start. When the CPU is disturbed and the program falls into an infinite loop or "runaway" state, the WDT can effectively reset the system, improving the system's anti-interference performance. [page]

(2) Peripheral circuit part

The chip select signal of the interface circuit is generated by 74LS138 decoding the high address lines P2.0 (A8), P2.1 (A9), and P2.2 (A10). The main chip select signals include 8155 programmable interface circuit chip select signal CS_8155 (Y0), keyboard interface chip select signal CS_KEY (Y1), and LCD module chip select signal CS_LCD (Y2). The LCD uses the character dot matrix LCD display module GD1602S produced by OCULAR. This module can display 20×2 5×7 dot matrix characters, has powerful functions, and is easy to interface with 8-bit MCU; the keyboard interface is designed as a universal PC interface (PS/2) because the universal PC keyboard has the advantages of low price, high reliability, good versatility, and easy operation, and is easy to maintain. Due to space limitations, this article does not introduce it. A13, A14, and A15 form the coding address line of the multi-channel analog switch 4051. Vxi is the analog signal selected by 4051 according to the address input by the keyboard. It will be sent to the signal conditioning module for amplification and conditioning. In harsh industrial working environments, serial communication interface chips are likely to be damaged by static electricity. Especially in the case where the transmission line is installed outdoors, the interface chip and even the entire system may be struck by lightning. The RS-485 interface chip SN75LBC184 selected in this paper can not only resist lightning impact but also withstand electrostatic discharge impact of up to 8kV. The maximum transmission distance is about 1219m and the maximum transmission rate is 10Mb/s. It can ensure stable and reliable communication with the host computer.

3. Dual-integral analog/digital conversion interface circuit

(1) Introduction to chip ICL7135

ICL7135 is a dual-integral A/D conversion integrated circuit produced by MAXIN Corporation of the United States. The chip has strong anti-interference ability, high resolution and low price. Its resolution is equivalent to 14-bit binary number, the conversion error is ±1LSB, and the conversion output is 0~19999; when the measurement range is 0kN~2000kN, such accuracy makes the instrument's resolution reach 0.1kN; the analog input can ensure the long-term stability of the 0 point at room temperature. Since the conversion result of the 7135 output is a dynamically scanned BCD code, the conventional design is generally connected to the microcontroller through a parallel interface to save the hardware overhead of the microcontroller. At the same time, the timer in the 8155 can also meet the clock needs of the 7135.

(2) A/D conversion circuit and its principle

The circuit schematic is shown in Figure 5.

The connection between 7135 and 8155 is realized through 4-bit 2-to-1 data multiplexer 74LS157. Its selection signal is controlled by the D5 output of 7135. When the A/D conversion is completed, D5 outputs a high level, 74LS157 selects the Class B channel, and the microcontroller reads the ten thousand digit B1, status bits POL (polarity), OVR (overrange) and UR (underrange) through PA0~PA3; when the D5 output is completed, it becomes a low level (this process includes the D4~D1 data output cycle), 74LS157 selects the Class A channel, and the microcontroller reads the 8421 code values ​​B8, B4, B2 and B1, that is, the low-order BCD code, through PA0~PA3, and forms the BCD codes of ten thousand, thousand, hundred, ten and individual digits (that is, the conversion result) in sequence. The interrupt request line PC0 of 8155A port is inverted to form the external interrupt 0 trigger signal of the microcontroller. When 7135 completes one conversion, it generates five data strobe pulses, and sends each BCD code and bit mark to port A respectively; after port A receives a data, the interrupt line PC0 becomes high level, starts the interrupt service program of the microcontroller, and reads the data result of the A/D conversion. The conversion start of 7135 is controlled by P14, and the high level conversion starts and the low level is maintained.

The conditioned 0V~2V analog signal is input from IN+ and IN- after passing through the RC low-pass filter. The result after A/D conversion includes two parts: polarity and range, which reflect the result of the conversion nature. The converted digital signal is D1, D2, D3, D41 and D5, of which D5 (ten thousand digits) can only be 1 or 0, and the other digits are BCD codes of 0~9. The reference voltage required by 7135 is 1/2 of the range. If the range is 2V, the reference voltage is 1V. In order to ensure the temperature stability and accuracy of the conversion, a high-precision reference power supply MC1403 is used, which is adjusted by a metal film multi-turn imported potentiometer. The CBB capacitors C46 (0.1mF) and C47 (1mF) connected in parallel to the reference voltage are mainly used to ensure the stability of the reference voltage.

The integral capacitor is a key component that determines the conversion accuracy. According to the application characteristics of 7135, the integral capacitor C8 and the integral resistor R3 are related to the range, etc., and the following requirements must be met when selecting:

R3 = full-scale voltage/20mA (3)

C8 = 1000×20mA×T/integrator output swing (4)

In the formula, T=1/fclk.

fclk—the clock frequency of 7135, which can generally be selected from 250kHz, 166kHz, 125kHz, and 100kHz, with a typical value of 125kHz. At this time, the conversion rate of 7135 is 3 times/s.

In a ±5V system, if the analog ground is 0V, the swing is ±4V. At this time, the range is -C2V~+2V, then: R3=100kW, C8=0.47mF. [page]

(3) Application of PLD technology

In order to save the cost of hardware circuits and reduce the electromagnetic interference generated by hardware circuits, PLD technology is applied to some circuits of the system. Its programmable logic circuit (see Figure 5) is composed of programmable logic display chip GAL16V8, which mainly completes the logic conversion of the clock signal and conversion end selection signal required by the A/D conversion module to generate the interrupt signal of the single-chip microcomputer external interrupt 0. Its logic equation is as follows:

P16=+(NAND gate) (5)

P14=P7+P8 (OR gate) (6)

P13=(NOT gate) (7)

In the formula, P2, P3, and P7 are the ALE signals of the single-chip microcomputer respectively;

P8—NAND gate output; P9—8155 A port interrupt signal; P13—generated single-chip microcomputer external interrupt trigger signal.

The PLD file generated according to the above logical relationship is compiled by FM software to generate a fuse file *.LED, which can then be written into GAL16V8 through the programmer.

3. System software design

The system software design adopts a modular structure and is programmed in assembly language. The entire program consists of subroutine modules such as the main program, display, keyboard scanning, and A/D conversion processing. Due to space limitations, only the A/D conversion processing subroutine flow chart is listed here, as shown in Figure 6.

4. Correction of nonlinear error

Sensors, amplifiers, and A/D converters always have nonlinear errors. Due to the existence of the above nonlinear relationship, the accuracy is reduced. In order to ensure that the accuracy requirements are met within the entire range, the measured value should be corrected according to the control requirements in practical applications. The correction is generally achieved through software calibration. The specific correction method should be determined according to the working section and quality requirements of the signal. The calculation and control capabilities of the microcontroller are used to find the correction algorithm for the nonlinear relationship, and after repeated testing and adjustment, it meets the design requirements. This system uses piecewise linear interpolation to correct the error of the measured value curve. The method is: divide 0~XMAX into several working sections, and each curve is replaced by a corresponding broken line. VQ can be calculated for each broken line:

Nt:VQ=ai×Nt+bi (8)

In the formula, I is the serial number of a broken line; ai is the slope of the broken line; bi is the intercept of the broken line.

Their correction program flow chart is shown in Figure 7, and the processing relationship schematic curve is shown in Figure 8.

The value of each segment is stored in the single-chip microcomputer in advance. In different working sections, the single-chip microcomputer automatically calls out the corresponding value of each segment for calculation processing. (Due to space limitations, quantitative analysis is not discussed)

V. Conclusion

This article describes a universal high-precision pressure data acquisition system, which has many advantages. It can work normally in various harsh environments, and has strong anti-interference ability, long service life and high resolution. It uses LCD display and PS/2 keyboard interface to achieve good human-computer exchange. It can be widely used in pressure data acquisition, detection and analysis in industries and scientific research such as petroleum, chemical industry, metallurgy, electricity, textile, light industry, water conservancy, etc.