Design of high-precision temperature controller based on HART protocol

1 Introduction

With the rapid development of control technology, computer technology, communication technology and the development needs of CIMS (Computer Integrated Manufacturing System), the fieldbus technology with fieldbus network as the overall symbol has emerged and developed rapidly, pushing the control system structure and automation instruments to the generation of fieldbus control system and bus-based instruments. This paper introduces microprocessor technology and HART protocol communication technology into the temperature controller, realizing a high-precision, multi-functional programmable two-wire HART protocol high-precision temperature controller.

2 Overview of the hardware principle of the temperature controller

The circuit principle block diagram of the high-precision temperature controller based on the HART protocol is shown in Figure 1. The two wires connecting the HART protocol temperature controller are both power lines and 4-20mA output (control) signal lines and HART signal lines. The current supplying power to the HART protocol temperature controller is limited to less than 4mA (3.6mA in alarm conditions), and the part above 4mA is the signal.

The AD421 produced by AD company is used to realize the A/D and V/I functions. AD421 is a low-power multifunctional chip with 16-bit A/D, V/I and voltage adjustment circuits integrated inside. DC/DC is a power isolator designed based on the principle of switching power supply. The signal input and A/D circuit of the temperature controller based on the HART protocol are powered by DC/DC. The photoelectric isolator with low driving current is used for signal isolation to realize the isolation of the small signal input and output signal of the HART protocol temperature controller sensor and the external power supply, and at the same time realize the low current consumption requirement of the temperature controller.

The working principle of the circuit is: the temperature sensor signal is filtered and sent to AD7714 for amplification and conversion into a corresponding digital signal, and then sent to the CPU after isolation by the optocoupler HP4731. After linearization and correction by the CPU, it is sent to AD421 for conversion into a corresponding 4-20mA standard current output. In addition, the digital communication signal on the loop is filtered and sent to the demodulator HT2012, and the demodulated signal is sent to the CPU through the serial port. Then the CPU sends the corresponding response signal to HT2012 to modulate it into a HART digital signal, which is converted into a corresponding digital signal by the control V/I conversion circuit after shaping and superimposed on the 4-20mA DC signal.

3 Hardware Circuit Detailed Design

3.1 Signal isolation and determination of optocoupler peripheral parameters

The information exchange between A/D converter and CPU is isolated by optical coupling.

1) Determination of the optocoupler peripheral parameters is shown in Figure 2. Taking the calculation of R18 and R20 as an example, since the minimum drive current IF of the optocoupler is 40μA, the solution is obtained according to equations (1) and (2).

Considering the worst case and margin, R18=62KΩ, and similarly, R17=62KΩ.

The calculation of R15, R16, R21, and R22 is similar and will not be described here.

2) The pulse width stretching circuit is shown in Figure 3.

Figure 3 Pulse width stretching circuit and waveform

After the AD7714 data is updated, its DRDY line needs to send a trigger high pulse with a width of 203.5μs, and the minimum response pulse width of the optocoupler is 500μs, so a pulse width expansion circuit must be set. Principle analysis, assuming that in the original state, points a, c and d in the figure are logic low levels; b and e are high levels. When the VI pulse arrives, point b jumps to 2VCC and then discharges to VCC, and points c, d, and e remain unchanged. When the negative jump edge of VI arrives, points b, c, d, and e flip, and then C21 starts to discharge. When the potential at point d drops to the inverter conversion voltage value, the inverter state changes, point e becomes a high level, and charges C13. When the potential at point b rises to the inverter conversion voltage value, the inverter state changes, point C becomes a low level, and a steady-state pulse is drawn from point C.

Since the DRDY high pulse width of AD7714 is 500×tCLKIN=203.5μs, T3 should be less than 203.5μs, T3=-R14C13ln0.5, R14=200KΩ, C13<1.5nf, T2=-R14C13ln0.33=113μs T1=-R13C21ln0.33=TW-T2, since the minimum response pulse width of the optocoupler is 500μs, TW=2.3ms is taken into account for the margin, R13 is 200KΩ, C21=0.00978μf is obtained, and C21=0.01μf is taken.

3.2 Signal input circuit design

The signal input part is composed of single-pair, double-pair, double-resistance, two-wire thermal resistance, three-wire thermal resistance, and four-wire thermal resistance. It is composed of high-frequency bypass through-hole capacitor, constant current source, low-pass filter network (R1-R6, C3-C8), etc. The through-hole capacitor is used to attenuate the high-frequency signal in series with the signal line. Considering the insulation strength of the signal line to the ground, a 100P/500V through-hole capacitor is selected. The low-pass filter network adopts RC filtering. Considering the input impedance requirements of AD7714, R is selected as 1.1K and C is selected as 0.1μf. R9 and R10 are used for sensor disconnection detection, and the resistance value is 22MΩ. The input circuit also considers the lead resistance compensation calculation when detecting the resistance signal.

3.3 Communication signal filtering circuit

The communication signal filter circuit is shown in Figure 4. The filter circuit is composed of a high-pass and low-pass filter to form a bandpass filter. Considering that the HART communication signal contains upper and lower half waves, in order not to cause signal distortion, the DC operating point is set at Vcc/2. C33, C34, R24, R27 and D2C form a second-order high-pass filter. Assuming C33=C34=C, R24=R27=RZE, its transfer function is (3).

Taking the denominator of (4) equal to v2, the cutoff frequency can be obtained as 0.37/2πRC. According to the characteristics of HART signals, the high-pass cutoff frequency is generally 600Hz. A first-order low-pass filter is composed of R22, R23, C30 and D2B. Assuming C22=R, C30=C, its cutoff frequency is 1/2πRC, which is generally 2500Hz in HART communication.

According to the technical index requirements, the temperature controller based on the HART protocol should be intrinsically safe, so all large energy storage components in the circuit should be equipped with voltage limiting or current limiting protection components to meet the intrinsic safety index requirements. In terms of reliability design, the derating design technology is mainly used to make the actual working stress of most components ≤0.1.

4 Software Implementation of Temperature Controller Based on HART Protocol

The development of the temperature controller software is based on the research of the HART protocol data link layer and application layer, combined with the functional requirements of the temperature controller itself and the connection of the hardware circuit. In addition to hardware design, software design is another important aspect of realizing the temperature controller. According to the functional requirements and technical indicators of the temperature controller, the main contents of the software design are:

a. According to the HART protocol specification, complete the HART communication link connection, link arbitration, signal reception, identification, response and transmission from the device data link layer, application layer and inter-layer interface program.

b. Consider signal acquisition compensation, linearization processing and output.

c. Complete the interface program according to the device interface.

d. Realize fault identification and alarm, local debugging and display.

The working principle of the software is shown in Figure 5. It is mainly divided into three parts. One part is the signal acquisition and output; one part is the arrival of the HART communication signal, entering the interrupt, identifying and responding to the communication signal, including some programs such as the data link layer, application layer and inter-layer interface; the other part is the local parameter configuration when a key is pressed.

The author's innovation points:

Compared with traditional field instruments or intelligent field instruments, high-precision temperature controllers based on the HART protocol have made a leap in functionality, namely, the addition of a two-way digital communication function. This gives the temperature controller, with its many advantages, a good application prospect in industrial process control systems, instrument management systems, and networks where these two systems coexist.