0
\$\begingroup\$

enter image description here

I made a schematic for an h-bridge motor driver that will be connected to a MCU.

Is this schematic correct? I'm not sure if the flyback diodes are working correctly since I added that NPN transistor.

A and B are used to control the motor's spinning direction.

\$\endgroup\$
5
\$\begingroup\$

There are several problems:

  1. I didn't look up to see what kind of FET exactly a IRF540 is, but in any case you can't just flip it upside down and expect it to work in a complementary way.

  2. Even with the right FETs oriented correctly, it doesn't make sense to tie the gates of opposite corners together. The motor voltage will be between the two FETs, so the same gate voltage won't apply the same way to both of them.

  3. I can't see a point to the top NPN transistor. You can already shut off the motor by switching combinations of FETs on and off, so all this additional switch does is waste power.

  4. Even if you wanted a master switch transistor, a bipolar in emitter follower doesn't make much sense. To saturate it, you'd have to drive it from a supply higher than Vcc.

\$\endgroup\$
  • \$\begingroup\$ I used the NPN because I wanted a square signal to power the motor, either clockwise or counterclockwise. \$\endgroup\$ – Cristi Nov 1 '12 at 21:13
  • \$\begingroup\$ But now I see I did many mistakes \$\endgroup\$ – Cristi Nov 1 '12 at 21:15
1
\$\begingroup\$

Search for "high side gate drive" ICs/hybrids. That is what you likely want to ensure you saturate Q1 and Q3 if you want to stick with N-channel parts on the high side. A high side gate driver will handle the DC/DC conversion and level shifting for you so that you can apply logic-level inputs while still pulling the gate above Vcc.

There are many h-bridge gate drive ICs available that offer capabilities like higher gate drive currents for the low side FETs and--something very valuable for those experimenting with H-bridges--cross-conduction prevention logic. Cross-conduction will happen if you turn on both FETs on the same side, or if one FET turns on before its partner on the same side finishes turning off. It is almost always destructive to both FETs on that side (since you're basically shorting the supply rails through the two FETs), and can also take out PCB traces, microcontrollers, etc.

As Olin suggests, omit T1 and simply apply your PWM waveform to the FET gates themselves.

\$\endgroup\$
0
\$\begingroup\$

In addition to what Olin said, one of the answers to a similar question suggests that you should add a small resistor (100 ohms or less) on each MOSFET gate to prevent ringing between them.

\$\endgroup\$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.