A constant current source design based on power amplifier

In the circuit breaker reliability test equipment, the stability and accuracy of the test power supply are the basis for ensuring the reliability of the test. Otherwise, whether in the circuit breaker factory test or type test, the fluctuation of the test power supply will make it possible for the qualified products to be judged as unqualified and the unqualified products to be judged as qualified. The traditional constant current source production is to use the characteristics of diodes, transistors, and integrated voltage stabilizers to make parameter current stabilizers, series feedback adjustment type current stabilizers, switch current stabilizers, etc., but they often have the disadvantages of small output current range, low current stabilization accuracy, low efficiency, poor reliability, and large output ripple. We designed a constant current source control system based on AT89C51, with current output 0~100A, current accuracy ≤2%, and voltage output 15V. It can achieve fast, high-precision, flexible, and multi-functional control requirements, and provide a stable and accurate test power supply in the circuit breaker reliability test.

Composition of the main circuit
The main circuit is composed of several circuits, including voltage and current regulation circuit, current boost transformer, current detection feedback circuit, input control and display. All the above modules are controlled by AT89C51. The overall structure is shown in Figure 1.

Figure 1 System structure diagram

1 Voltage and current
regulation circuit The voltage regulation module is mainly composed of a transformer and a DS1267 digital potentiometer. The maximum adjustable precision of a single DS1267 is 16 bits. It can be seen that the minimum single change is 1/512. For 220V voltage, it can be basically considered a linear relationship, which meets the voltage regulation accuracy of the constant current source. The current regulation module is mainly composed of the TDA2030 chip and high-power transistors 2SA1302 and 2SC3281. Among them, 2SA1302 and 2SC3281 form a push-pull power amplifier structure. In order to increase the output current, two parallel circuits with the same structure are used. The circuit is shown in Figure 2.

Figure 2 Push-pull power amplifier circuit

In Figure 2, when the voltage signal is input, since the dynamic resistance of the two diodes of IN4001 is very small and the resistance of R2 is small, it can be considered that the change of the base potential of the 2SA1302 tube is approximately equal to the change of the base potential of the 2SC3281 tube, and the potential of the two bases changes the same with the input voltage uin. When it is in the positive half cycle of the input signal and uin gradually increases, the base current of the 2SA1302 tube increases accordingly, and the emitter current must also increase, and the load resistor (i.e., the boost transformer) RL obtains a positive current; when uin decreases to a certain value, the 2SC3281 tube is cut off. Therefore, the positive half cycle of the input signal is mainly driven by the emitter of the 2SA1302 tube. For the same reason, the negative half cycle is mainly driven by the emitter of the 2SC3281 tube.

2 Current boost transformer
This test requires the generation of a large current of 0 to 100A. Considering that this current source is used for online detection of the circuit breaker, the contact resistance of the circuit breaker is 15mΩ, so the power consumed by the load should be P=I2R=1002×0.015=150W. The load consumes 150W of power. Considering the transformer efficiency and power margin, we choose a current boost transformer with a rated capacity of 500VA.

The core area S and the power P of the current-boosting transformer satisfy the following empirical formula: ln(S)=0.498×ln(P)+0.22. With the power P=500VA, the core cross-sectional area S=53.144cm2 can be calculated. According to the calculation results, S=54cm2 is taken, and the middle tongue size of the silicon steel sheet a=6cm and the stack thickness b=9cm are selected.
According to the core cross-sectional area S and the magnetic flux density B of the core, the number of turns per volt N of the primary coil can be determined by the following formula: ln(N)=-0.494×ln(P)－0.317×ln(B)＋6.439. Using high-quality silicon steel sheets, the core B value is 11000 Gauss, and the number of turns per volt N=0.831 is calculated. The primary voltage is 220V, the primary turns N1=220×0.831=183; the secondary voltage is 7V, the secondary turns N2=7×0.831=6.

The number of primary and secondary turns and the maximum secondary current are 100A, and the secondary current is: I1=I2×N2/N1=3.4A. According to experience, 0.3mm2 of wire cross-sectional area is allocated for each ampere of current. In this way, the cross-sectional area of ​​the primary wire is 1.02mm2, and the primary wire can be 135mm2 flat copper wire. The cross-sectional area of ​​the secondary wire is 30mm2, and the secondary wire can be 240mm2 flat copper plate.

3 Current detection feedback circuit control display module
The current detection feedback module consists of a current transformer, a precision absolute value circuit, an active low-pass filter, and an A/D conversion chip. According to the output current, we choose the DHKYZ-500 model current transformer as the current sampling sensor. The full-scale current of the sensor is 500A, and the full-scale secondary output current is 100mA, in order to meet the requirements of the A/D converter input range (0~5V). A/D conversion requires a DC signal, so the AC signal needs to be conditioned. The precision rectifier circuit used in this design is shown in Figure 3. The circuit is mainly composed of two dual operational amplifiers TL062 and related components. The input voltage uin of the circuit is the current sensed output by the current transformer.

Figure 3 Precision rectifier circuit

As shown in Figure 3, when ui>0, Dl is turned on and D2 is reverse blocked. It can be calculated that u11=-ui/2, u12=-u11=ui/2>0; when ui<0, Dl is reverse blocked and D2 is turned on. For the first operational amplifier TL062, u11=-ui/3 can be obtained, so u12=-ui/2>0 can be calculated, u21=-2u12, and finally uo=-u21=2u12, so the output is a full-wave rectified waveform.

Since the signal output by the precision rectifier circuit is a pulsating DC signal, it cannot be directly used as the input signal for AD sampling. Therefore, it must first pass through a low-pass filter to filter out the AC component and take out the DC component before inputting it into the A/D converter.

4 Control display module
At present, the LED display drive circuit in industrial control generally adopts a timing or interrupt control method, which takes up a part of the CPU time, and the dynamic display often has the characteristics of insufficient brightness and flickering, while the static display has defects such as complex hardware circuits. This system uses the OD-DM12864 LCD module, which can be directly connected to the microcomputer serial port, completely solving many shortcomings of LED display. Users only need to program the bit and control registers to select decoding mode, display brightness, shutdown and other functions.

Control algorithm and program design ideas
1. Selection of control algorithm
Constant current source component detection is carried out through a multi-parameter mutually coupled time-varying nonlinear system. There are many factors that affect the accuracy of current detection, and there is a lot of randomness. It is difficult to describe it with an accurate mathematical model. Even if a simple mathematical model of the object is established after simplifying the system by some means, the control effect is not very good. In addition, since the current changes with the changes in component parameters, the control algorithm is required to be highly real-time and the control process is relatively complicated. Therefore, weighing the advantages and disadvantages of various control methods, we use PID to achieve real-time control.

2 Programming ideas
According to the system requirements, a modular programming method is adopted here. The program is decomposed into modules according to the hardware function modules, and then the functions and interface definitions of each module are defined. The main program flow chart is shown in Figure 4.

Figure 4 Main program flow chart

Experimental debugging
The input of this experiment is AC 220V, and the output is a DC voltage of about 15V. The accuracy of the current transformer used for measurement is 0.5 level and the accuracy of the clamp meter is 1.0%rdg±10dgt. The current test value is recorded in the experiment, as shown in Table 1.

As can be seen from Table 1, there is a certain deviation between the set input value and the value detected by the current transformer, but it can be controlled at about 1%, meeting the design requirements. The clamp meter value sometimes has a large deviation, but it is also within the error range. Therefore, the result meets the actual measurement accuracy requirements.