Buck switching power supply MOS tube gate has pwm signal, but cannot conduct normally

taoanting2022

Buck switching power supply MOS tube gate has pwm signal, but cannot conduct normally [Copy link]

 

 

I am a power supply novice. Recently, my teacher asked me to make a digital buck circuit. I used a signal generator to generate the pwm signal. After I finished, I found that the gate of the pmos tube had a normal pwm signal, but the source and drain could not conduct normally. I thought it was a problem with the mos tube, but I took the mos tube out and built a circuit and found that the mos tube was good. Then I removed the buck part and connected the mos tube directly to the resistor. I found that the voltage waveform of the resistor was also normal, so I thought it was a problem with the buck circuit parameters. But I couldn't find the specific reason after searching for a long time, so I came to ask the experts in the forum~~ This is the buck circuit I used. This is the output voltage. No matter what the duty cycle is, the output voltage is always equal to the power supply voltage. This is the signal of the gate, which feels normal. This is the waveform between the source and the gate. This is the waveform between the source and the drain. This waveform is very strange. I have been working on it for two days and still can't find the reason. I hope the masters can help me solve it!

taoanting2022

Figure 1 is the schematic diagram, Figure 2 is the output voltage, no matter how the duty cycle changes, it is always equal to the power supply voltage, Figure 3 is the gate signal, Figure 4 is the waveform between the source and the gate, and Figure 5 is the source and drain waveform. This is my first post, so I am not very good at typesetting, please forgive me~

呜呼哀哉

This has nothing to do with your circuit. I think you have some misunderstanding about the power supply circuit. Let me ask a simple question. The duty cycle of the buck is D=Vo/VIn. If the circuit parameters remain unchanged, the load changes from 1 ohm to 1M ohm. Does D change? Does the current waveform on the inductor change?

maychang

taoanting2022 Published on 2022-8-20 16:44 Figure 1 is the schematic diagram, Figure 2 is the output voltage, no matter how the duty cycle changes, it is always equal to the power supply voltage, Figure 3 is the gate signal, and Figure 4 is the source and gate voltage...

"But the source and drain cannot conduct normally."

On what phenomenon do you base your statement that the source and drain cannot be normally conductive?

maychang

taoanting2022 Published on 2022-8-20 16:44 Figure 1 is the schematic diagram, Figure 2 is the output voltage, no matter how the duty cycle changes, it is always equal to the power supply voltage, Figure 3 is the gate signal, and Figure 4 is the source and gate voltage...

"Figure 2 is the output voltage. No matter how the duty cycle changes, it is always equal to the power supply voltage."

If the MOS tube is not turned on, how can there be voltage at the output end of the circuit? It can be seen that the MOS tube is turned on, at least part of the time in one PWM cycle.

taoanting2022

Alas, published on 2022-8-20 17:21 This has nothing to do with your circuit. I think you have some misunderstanding about the power supply circuit? Ask a simple question, the duty cycle of the buck is D=Vo/VIn, the circuit...

Sorry for the late reply. I really don't understand much about power supply. Regarding your question, I don't think the duty cycle will be affected, and I don't know about the inductor current.

taoanting2022

maychang posted on 2022-8-20 19:24 『But the source and drain cannot conduct normally.』 On what basis do you say that the source and drain cannot conduct normally? ...

Sorry for the late reply. I saw that the waveform of the source and drain on the oscilloscope is a decaying oscillation wave. I think if it is normally turned on, it should also be a PWM wave.

maychang

taoanting2022 posted on 2022-8-20 19:41 Sorry for the late reply. I saw that the waveform of the source and drain on the oscilloscope is a decaying oscillation wave. I think if it is normally turned on, it should also be a PWM wave...

The attenuated oscillation period is the period when the MOS tube is turned off, and the horizontal straight line period is the period when the MOS tube is turned on.

taoanting2022

maychang published on 2022-8-20 19:26 『Figure 2 is the output voltage. No matter how the duty cycle changes, it is always equal to the power supply voltage』 If the MOS tube is not turned on, how can there be a voltage at the output end of the circuit...

I adjusted the duty cycle, but the output voltage remained unchanged, which was very strange. In a normal buck, does Vo increase as D increases?

taoanting2022

maychang published on 2022-8-20 19:44 The decay oscillation period is the period when the MOS tube is turned off, and the horizontal straight line period is the period when the MOS tube is turned on.

I really don't understand this. I will learn it carefully in the future. This buck is the first power supply I made. It was assigned by the teacher in school. So can I ask why the output voltage cannot be adjusted? Is it related to the parameters of the inductor and capacitor?

taoanting2022

maychang published on 2022-8-20 19:44 The decay oscillation period is the period when the MOS tube is turned off, and the horizontal straight line period is the period when the MOS tube is turned on.

During the debugging process, I removed the capacitors and inductors in the second half, and directly connected a resistor in parallel with the diode. After measuring, I found that the waveform was very good, but it didn't work when I connected everything.

maychang

taoanting2022 posted on 2022-8-20 19:41 Sorry for the late reply. I saw that the waveform of the source and drain on the oscilloscope is a decaying oscillation wave. I think if it is normally turned on, it should also be a PWM wave...

Your PWM signal is input by the signal generator on the left side of R11, and the duty cycle does not change during a measurement. And your load resistance value is quite large, that is, the output current is quite small (calculated based on the output voltage of 12V, about 9mA).

The time constant of the 470uF capacitor and the 1.33 kilo-ohm resistor is about 0.6 seconds, which is much longer than the PWM cycle (at least tens of thousands of times). In such a short PWM cycle, the 470uF capacitor is discharged very little and the voltage hardly drops. In each PWM cycle, the MOS tube is turned on once to charge the capacitor to the power supply voltage.

maychang

taoanting2022 posted on 2022-8-20 19:41 Sorry for the late reply. I saw that the waveform of the source and drain on the oscilloscope is a decaying oscillation wave. I think if it is normally turned on, it should also be a PWM wave...

The question I asked you on the 3rd floor is to illustrate that your Buck circuit is currently operating in this "uncontrollable" state.

maychang

taoanting2022 posted on 2022-8-20 19:41 Sorry for the late reply. I saw that the waveform of the source and drain on the oscilloscope is a decaying oscillation wave. I think if it is normally turned on, it should also be a PWM wave...

So, your circuit is working perfectly fine. If you need the output voltage to vary with the PWM duty cycle, just reduce the value of the load resistor and make the current (average value) in the inductor larger.

However, 10uH inductor is always too small. You might as well use a larger one.

taoanting2022

maychang published on 2022-8-20 19:53 Your PWM signal is input by the signal generator on the left side of R11, and the duty cycle does not change during a measurement. And your load resistance value is quite large, ...

Oh, I got it. Master, should I lower the time constant RC? Should I recalculate the parameters of the inductor? I originally planned to make the switching frequency 100k.

taoanting2022

maychang posted on 2022-8-20 20:03 So, your circuit works perfectly. If you need the output voltage to change with the PWM duty cycle, just reduce the value of the load resistance and make the current in the inductor (flat...

Oh, I'll try. Forgive me for having so many questions and being stupid, haha.

maychang

taoanting2022 posted on 2022-8-20 19:51 During the debugging process, I also removed the capacitors and inductors in the second half, and then directly connected a resistor in parallel with the diode. After measuring, I found that the waveform was very...

Yes. If the inductor and capacitor are removed, the load resistor will of course have a rectangular wave. The appearance of a rectangular wave indicates that your MOS tube is conducting and shutting down normally.

maychang

taoanting2022 posted on 2022-8-20 20:03 Oh, I got it. Master, should I lower the time constant RC now? Should I recalculate the parameters of the inductor? I originally planned to make a switch...

"So should I lower the time constant RC now?"

The point is not to reduce the RC time constant, but to make the current in the inductor continuous, or at least close to continuous.

maychang

taoanting2022 posted on 2022-8-20 20:03 Oh, I got it. Master, should I lower the time constant RC now? Should I recalculate the parameters of the inductor? I originally planned to make a switch...

The first post said that "I have been working on it for two days" and still can't find the cause. In fact, the circuit is working normally, so of course I can't find the cause. Instead of spending two days looking for the cause, it is better to spend two hours to read the Buck circuit textbook (since the teacher asked you to make a Buck circuit, there must be a textbook) and understand how the various parts of the Buck circuit work. This will get twice the result with half the effort.

taoanting2022

maychang posted on 2022-8-20 20:07 『So should I reduce the time constant RC now? 』 It is not to reduce the RC time constant, but to make the current in the inductor continuous, at least close to...

I just went to debug. Master, I reduced the resistance as you said, and only used a 90 ohm one, but it still doesn't work, the voltage is still the power supply voltage, and I set the duty cycle to 0.5. The capacitor on the schematic uses a 470uF, which I think is too big, so I changed it to a 47uF. Should I change it back?