Working Principle and Application Analysis of High Power High Frequency Hard Switching PWM Converter

Inverter welding technology has experienced nearly a decade of development, gradually replacing the backward power frequency thyristor rectification technology and entering the era of high-frequency conversion. In the process of high frequency conversion technology, it has gone through its primary stage, namely hard switching PWM stage, and entered its second stage, namely soft switching PWM stage in recent years.

The topology of hard switching PWM converter is simple. Mature technology, suitable for mass production. Its main chips such as TL494 and UC3525 are relatively stable and reliable, which is the main reason why the hard switching PWM power converter of inverter welding machine is still widely used.

The so-called hard switch PWM (pulse width modulation) refers to that the electronic switch is in the working condition of high current or high voltage at the moment of opening and closing in the process of power conversion, so its workpiece has poor reliability, low efficiency and serious electromagnetic interference.

The so-called soft switching technology refers to the power conversion technology, which is the technology to achieve zero voltage or current at both ends at the moment when the main switching device is turned off and on. That is, ZVS (zero voltage switch) and ZCS (zero current switch) switching technology often used in terminology. Soft switching PWM power converter technology is a revolutionary development compared with hard switching PWM technology. It does improve the reliability, efficiency and electromagnetic interference (EMI) of power products to a considerable extent. At present, most of the high-power switching power supplies developed by domestic peers use hard switching PWM control mode, and only a few use soft switching PWM. Most of the soft switching PWM adopts phase-shifting control mode. Control chips such as UC3875, UC3879 and ucc3895 are used to reduce the switching stress and switching loss of power devices Thus, the efficiency of the whole machine is improved. However, this soft switch also has many shortcomings and regrets, such as:

(1) This soft switching mode of medium and high power phase-shifting control is not full range;

(2) Due to the existence of circulating current, the conduction loss of the switch is large, and the light load aging rate is low, especially when the duty cycle is small, the loss is more serious;

(3) Parasitic oscillation exists in the output rectifier diode;

(4) In order to realize the ZVS of the lagging bridge arm, the inductance must be connected in series in the circuit, which leads to the loss of duty cycle, reduces the output capacity and increases the primary side current rating.

Moreover, the phase-shifting control itself has an insurmountable disadvantage, that is, the dead time is not easy to adjust. When the load is heavy, due to the large circulating current, the capacitor connected in parallel on the leading bridge arm tube discharges rapidly, so it is easy to realize zero voltage conduction. However, when the load is light, the capacitor connected in parallel on the leading bridge arm switch tube discharges very slowly, and the switching tube of the leading bridge arm must delay for a long time to realize ZVS conduction.

Therefore, we absorb the advantages of traditional hard switching PWM power converter, such as simple topology, less integration points, stability and reliability; At the same time, it absorbs the advantages of phase-shift control soft switching PWM power converter, which is easy to realize ZVS and ZCS; A new high-power full bridge soft switching (FB-ZVZCS) technology is introduced to realize ZVS for super forearm and ZCS for lag arm. Thus, the soft switching (FB-ZVZCS) beyond the range of forearm and lag arm umbrella is realized, which greatly improves the reliability, efficiency and electromagnetic interference (EMI) of big power switching power supply products. The realization proves that this control method is very excellent, which can be said to be a revolution to the traditional hard switching high-power switching power supply.

1 full bridge soft switch (FB-ZVZCS) inverter welding machine

The outline diagram of the pulse lemon board of the high-power inverter welder is shown in Figure L. A total of 18 pins are led out of the circuit board. The circuit board is a combination of analog and digital circuits independently developed. This circuit can form the driving pulse required for soft switching.

1.1 inverter welding machine control board

The internal diagram of full bridge soft switch (FB-ZVZCS) control board is shown in Figure 2. The functions of each pin are described below.

Pin 1 is connected to the working power supply (UDD = 12V or 15V);

Pin 2 is connected to the ground of the working power supply;

Pin 3 is the reference power supply (UreF = 5V);

Pin 4 is the reverse input of the voltage error amplifier;

Pin 5 is the same direction input of the voltage error amplifier;

Pin 6 and pin ll are connected to timing capacitor (CT1 = CT2);

Pin 7 and pin 12 are connected to timing resistor (RT1

2 working principle and waveform of inverter welding machine

The waveform diagram is shown in Figure 4. The left arm is the leading bridge arm, the excitation signals of the upper and lower switches are constant frequency and width adjusted pulses, the right arm is the lagging bridge arm, and the excitation signals of the upper and lower switches are constant frequency and width pulses. Next, we will briefly analyze the process of realizing soft switching.

2.1 primary state (T1, T2)

S1 and S4 are on. At this time, the converter outputs energy to the secondary load. At this time, the working state is the same as that of our usual hard switching PWM.

2.2 status 2 (T2, T3)

S1 is off and S4 remains on, because there are capacitors (C1 and C3) on S1 and S3 Therefore, when S1 is turned off, the circuit current is not cut off at the same time, but C1 is charged and C3 is discharged through S4, l and t. at this time, the converter continues to output energy to the secondary load. The current at point a reaches point B through resonant inductor L and transformer t. if the energy of inductor L has not been released, the current continues to flow through the body secondary tube of S3, that is, the voltage at both ends of S3 is zero, providing zero voltage turn-on for S3 When S4 is off, the current on S4 has been approximately zero, so S4 is zero current off at this time.

2.3 status 3 (T3, T4)

At this time, S1, S2 and S4 are in the cut-off state. Due to the leakage inductance LS (very small leakage inductance) of the transformer, there is still some energy in the circuit, causing damping oscillation. Its frequency has nothing to do with the load, but only with the distributed capacitance (C1 and C3) of L, S2 and S4 Concerning, since C1 and C3 are much larger than the distributed capacitance of S2 and S4, this oscillation is only observed at both ends of drain source of S2 and S4, and there is no oscillation at both ends of drain source of S1 and S3. This oscillation will increase the loss of S2 and S4 and has no effect on S1 and S3. In order to reduce the loss on S2 and S4 and meet the opening of S2 and S4 in quasi zero voltage state, only the following conditions need to be met: T3 = T4-T3 = t / 2, t is the oscillation period. If t is too small, the inductance L can be increased. In order to ensure the safe operation of S2 and S4, T3 should be increased appropriately. At this time, l can be increased according to different conditions. C1 and C3 should be smaller when T2 ≥ RC is met. When MOSFET is used for power transistor, C1 and C3 generally obtain 1000 4700pf, and C1 and C3 are generally larger (10 20nf) when IGBT is used for power transistor.

After the above three states, the converter completes half a cycle, and the second half cycle is the same.

2.4 time setting of status 2 and status 3

Whether the design is reasonable is the key to realize soft switching and meet the maximum duty cycle. From the previous work process analysis, it can be seen that if state 2 is set too large, the duty cycle will decrease, the peak current of power tube will increase, and the reverse withstand voltage of secondary rectifier diode will increase, which will increase the loss of power tube and diode and high-frequency noise. Therefore, the duty cycle should be increased as much as possible However, if the design of state 2 is small and C1 and C3 cannot be fully charged and discharged, S1 and S3 cannot realize zero voltage switching, and their loss will increase, which is not allowed. The optimal value of state 3 time is critical. State 3 time is long. Due to high-frequency oscillation, the loss of S2 and S4 will increase, and state 3 time is short, which is easy to cause instantaneous short circuit of S2 and S4. When MOSFET is used for power transistor, state 3 time will increase Generally, it is about 300ns. When the power device adopts IGBT, it is generally larger (300 600ns).

3. Setting of driving waveform dead zone and front and rear edges of inverter welding machine

The dead band setting of S1 and S3 and S2 and S4 driving waveforms, and the setting of the relative positions of the front and rear edges of S1 and S4 or S3 and S2 waveforms are shown in Fig. 5.

4 Conclusion

The experimental results show that the designed high-power soft switching arc welding inverter not only has small volume, light weight and low production cost, but also has high efficiency and high reliability, and the switching loss of ICBT is greatly reduced. The processability, manufacturability and maintainability of the welding machine have reached a high level.