I'm trying to get an h bridge circuit working using PWM and a dsPIC33F. This is an existing design, which I am trying to get working using a different approach.

In the schematic, you can see PWM1H and PWM2L are driving the top half bridge. PWM2H and PWM1L are driving the bottom half.

I modified this circuit so that PWM1H and PWM1L are driving the top half and PWM2H and PWM2L are driving the bottom half. I am trying to use a complementary PWM signal with some dead time.

I'm using a dead time of around 1.7us. When I test the circuit with a low frequency of 20Hz, it seems to work, and I can get a voltage of 12V on the output with 50% duty cycle. However, I want to use a higher frequency so the output is not as choppy. I tried using 20kHz, but my current limit on my power supply keeps triggering, so I think something is shorting (current limit set to 250mA).

I tried increasing the dead time even further but it still did not help. I also noticed that the voltage at TP13 (gate drive voltage) was quite distorted at high frequencies, but normal at low frequencies.

I can't seem to figure out what's going on. Does anyone have advice for this?


Thanks.

 

I also noticed that the voltage at TP13 (gate drive voltage) was quite distorted at high frequencies, but normal at low frequencies.

1. Are you sure about that? When measuring with the scope did you maintain the same time scale to compare the slopes? Distorted how?
2. What's the load?
2. Please include the PWM signal sequence.

 

1. Are you sure about that? When measuring with the scope did you maintain the same time scale to compare the slopes? Distorted how?
2. What's the load?
2. Please include the PWM signal sequence.

Hi praondevou,

Thanks for your response. Here are some pictures depicting the strange things I'm seeing. I do not have any load connected right now, although it would be a peltier cooler in the end. The only load right now I guess is the LED.

Note that I am trying to get +12V on TOUT+ and GND on TOUT-. To get GND on the bottom half, I'm inputting a constant 0V for PWM2H and 3.3V for PWM2L.

Also, I removed the NAND gates from the circuit. I've named the pictures attached accordingly...please let me know if you need more details on anything. I can also post the code snippet if you want.

Thanks!