I'm building a class D amplifier with a dsPIC as PWM modulator.

The part of the schematic that I want to talk about is the interaction between the microcontroller output (pwm signal) and the gate of the MOSFET. I use TC4420 and TC4429 as gate drivers.

I know that there could be some overshooting (I think it's called "ringing") every time the output of the driver switches (for the resonant circuit created from gate capacitance and parasitic inductance), but I can see a lot of overshooting even between the output of the microcontroller and the input of the driver that has high impedance!!! Is it normal or am I missing something?? If it's normal, is it true that that overshooting can be lowered only with a series resistor?

Here is the waveform between dsPIC and driver: enter image description here

And this is the output of the gate driver: enter image description here

Talking about switching in the same pcb of a class D amp, is it true that is recomended to have two separate grounds for logic signals and for power? I searched the internet but I don't get this point: how can i have two separate grounds if the power cables has only one ground wire?

  • \$\begingroup\$ output of gate driver looks normal \$\endgroup\$ – Marla Aug 1 '16 at 18:21
  • \$\begingroup\$ Yes, you could drop a resistor between the PIC and the gate driver, but it is probably not necessary. Some processors also have a gate drive or slew rate setting that you can turn down. Alternatively, make the trace between the PIC and the driver wider and shorter (lower inductance). But all this, again, seems unnecessary. \$\endgroup\$ – Daniel Aug 1 '16 at 18:29
  • 4
    \$\begingroup\$ First question : what is your measurement setup? There is no point trying to fix ringing in your circuit if it's actually caused by your scope probe ground lead. \$\endgroup\$ – Brian Drummond Aug 1 '16 at 18:32
  • \$\begingroup\$ @Daniel, thank you for the advices. I also have another problem related to switches: I see a lot of noise in the power supply at the same frequency of the pwm signal. Inserting a lot of bypass capacitors is the only way to reduce this noise? \$\endgroup\$ – Dukenukem Aug 2 '16 at 17:14
  • 1
    \$\begingroup\$ You should address @BrianDrummond 's question and make sure that is real noise. Generally, though, yeah, you'd reduce the voltage ripple with more capacitance. \$\endgroup\$ – Daniel Aug 3 '16 at 2:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.