Principle and performance analysis of full-bridge IGBT pulse laser power supply

In recent years, high-power Nd:YAG solid-state lasers have been widely used in industrial processing and medical instrument fields, such as material processing, laser ranging, laser marking, laser medical treatment, laser nuclear fusion, etc. Compared with gas lasers or other lasers (such as chemical lasers, free electron lasers, etc.), solid-state lasers have the advantages of compact structure, firmness and durability, and have various operating modes, and can operate in pulse, continuous, Q-switched and mode-locked modes.

2 Principle Block Diagram

The laser power supply introduced in this article provides power for solid-state lasers that work in repetitive pulse mode. The laser uses a xenon lamp as the pump light source. Among the inert gas lamps, xenon has the highest total conversion efficiency. The laser is used for laser marking, and the operating frequency is 60 times per second. The power supply system uses an IGBT tube full-bridge inverter mode, the operating frequency is 20kHz, and the control circuit uses a PWM mode.

Figure 1 Schematic diagram

Figure 1 shows the power supply principle block diagram. The entire circuit can be divided into two parts: the main circuit (power conversion circuit) and the control circuit. The 380V AC voltage from the power grid is rectified and filtered to obtain a DC voltage of about 520V, which is added to the bridge inverter. The main power switch of the inverter uses Mitsubishi's CT60 IGBT tube. The PWM circuit generates a pair of pulse voltages with a phase difference of 180° to control the four power tubes of the inverter bridge, converting the DC voltage into a high-frequency square wave voltage, and then obtaining a high-voltage DC (about 1400V) through a high-frequency high-voltage rectifier bridge to charge the energy storage capacitor Co. After the voltage on capacitor Co is charged to the predetermined value (1000V), the control circuit sends a signal to trigger the discharge thyristor to turn on, and the voltage on Co is quickly released to the load xenon lamp, and the laser works normally.

The pre-ignition trigger circuit is designed for the characteristics of the load xenon lamp. This type of laser requires a high-voltage pulse of nearly 20,000 volts to break down the inside, and then maintain a low continuous current (about 100-200 mA) so that the laser can work normally under the intermittent discharge state of the capacitor Co. Therefore, the working steps of the power supply should be: power on - pre-ignition trigger - capacitor discharge.

3 Working principle and simulation waveform

Figure 2 shows the main circuit of the power supply. V1-V4 form a bridge inverter, with RCD absorption branches connected in parallel at both ends. L is a current limiting inductor, Co is an energy storage capacitor, and Lo is used to limit the discharge current of Co to the load xenon lamp to protect the xenon lamp.

Figure 2 Main circuit diagram

Different from the laser power supply described in Reference 1, the current limiting inductor L is placed on the primary side of the transformer here. This can not only realize the zero voltage turn-on of the power tube, for example, after V1 and V4 are turned off, due to the freewheeling effect of L, D2 and D3 are turned on, and V2 and V3 can realize zero voltage turn-on; it can also share the voltage drop on the primary winding of the transformer, reduce the number of transformer turns, and thus reduce the transformer core.

Figure 3 is the result of simulating Figure 2 using PSPICE software. The upper part of the figure is the transformer primary winding current i1, and the lower part is the output voltage Uo.

Figure 3 Simulation waveform

Simulation parameters: switching frequency f=20kHz,

Dead time t=2μs,

L=100μH, transformer ratio N1:N2=24:60

C0=100μF.

4 Design points

In the power supply system, the design of inductor L and high-frequency transformer T is the key. As shown in Figure 2, the load of the inverter circuit is only the inductor L and the primary side of the transformer T. When the power tube is turned on and the DC voltage Ui is applied to the position shown in Figure 4, the voltage on the inductor is Ui-UT1, then the primary current i1 of the transformer = (Ui-UT1)t/L

Figure 4 Calculation circuit

In the formula, UT1 is the value of the output voltage Uo (UT2 when the high-voltage rectifier bridge is turned on) converted to the primary side of the transformer, UT1=U0/n, n is the transformer secondary and primary turns ratio, and △Uo=（1/C0）,

i2=i1/n

Considering that the output voltage is gradually increasing and the amplitude of the current i1 is constantly decreasing, the calculation process should be an iterative process.

i1(m)=[Ui-U0(m-1)/n]t/L

U0(m)=U0(m-1)+△U0(m)=U0(m-1)+(1/C0)(i1(m)/n)dt

=U0（m－1）+Uit2/2CLn－U0（m－1）t2/2CLn2

Where, t is half a cycle, when the switching frequency f is 20kHz, t=25μs; C0=100μF; L is the inductance of the current limiting inductor.

According to the load characteristics, the maximum operating frequency is 60Hz, that is, the energy storage capacitor discharges 60 times in 1 second, and the cycle is 16.7ms. Considering the discharge time, the charging time is only 11ms at most. Therefore, the maximum value of m in the above formula is 11ms/25μs=440

MATLAB is used to calculate the above formula, and the rising curve of capacitor voltage U0 is plotted under different L values, different n values, and different DC voltage Ui (Ui has an allowable range of variation). The best solution is selected and the parameters are finally determined as follows:

L=100μH

n=N2：N1=60：24

In FIG5 , the horizontal axis represents the number of turns N2 of the secondary winding of the transformer, and the vertical axis represents the output voltage U0. When L=100μH, Ui=520V, f=20kHz, and N2=60, the output voltage U0 is the peak value.

Figure 5 MATLAB calculates the optimal transformation ratio

5 Test results

Figure 6 is the measured waveform, and (b) shows the first few cycles of (a). The upper part of the figure is the waveform of the transformer primary power supply i1, and the lower part is the voltage waveform of the energy storage capacitor U0, which corresponds to the simulation waveform in Figure 3. It can be seen from the figure that after i1 is zero, U0 does not drop immediately, but remains for a period of time. This is because the charging and discharging time is fixed. When U0 reaches the predetermined value, it remains until the charging time ends.

Figure 6 Experimental waveform