0
\$\begingroup\$

I designed this MOSFET driver for a 24V motor with parts I have in my junkbox:

enter image description here

The idea is basically Q2 and Q3 make a push-pull driver to drive the MOSFET gate to 0 or 24V. This allows the use of a low value resistor for the gate so it can be charged and discharged very quickly.

Q2 and Q3 bases are held up by R2, or shorted to ground by Q1. This allows the use of a 5V PWM signal to drive the high voltage switch. R4 just keeps the base tied to GND so it's not floating.

Is this a good design for this application?

Are the resistor values reasonable?

I think R3 may be a little larger, maybe up to 100 ohms to prevent destroying the gate, but at the same time, avoid negating the use of a push-pull driver. R2 provides a base current of 2.4mA to the transistors which is within the acceptable (<100mA) base current for BC8x7. The collector current of Q2/Q3 is hard to estimate and depends on the real gate charge of the MOSFET

Finally, R1 provides around 1mA of base current for Q1 which is more than enough for the collector current of 2.4mA

\$\endgroup\$
1
  • 1
    \$\begingroup\$ You need to ensure that the MOSFET gate is protected from having a voltage that is typically greater in magnitude than 15 volts below the source voltage. This pretty much applies to all MOSFETs although there may be the odd exception. Read the data sheet of a typical MOSFET to see why. \$\endgroup\$
    – Andy aka
    Commented Dec 3, 2020 at 14:41

1 Answer 1

1
\$\begingroup\$

24V is probably above the max Vgs for your FET so it will smoke. You need to limit Vgs to a safe value. For example you can add a resistor from Q2 base to ground to form a voltage divider with R2.

When Q1 turns on, Q3 base current is not limited, and it will exceed the maximum ratings for BC857. Q3 would need a base resistor.

There is also the issue of hFe and bipolar transistor current: when swinging the FET gate towards Vcc, Q2's base voltage comes from the 10k resistor which is quite a high value. Its base current will be quite small, and BC847 can only output a couple tens of mA before its hFe falls off. So the output current you'll get from this driver probably won't be more than a hundred mA. This should be enough for low frequency PWM though.

\$\endgroup\$
2
  • \$\begingroup\$ great tips. i completely missed the Q3 base current. what's a maximum safe current for a MOSFET gate that will not destroy it? and yes i'm outside Vgs. it's +/- 24V. i suppose my prototype worked because I was testing it at 12V! \$\endgroup\$
    – hjf
    Commented Dec 3, 2020 at 17:11
  • \$\begingroup\$ For max MOSFET gate current, check datasheet, but it'll probably take several amps so that won't be a problem at all. In fact if you use PWM for a load at rather low frequency like 25kHz it's better to switch a bit slowly to avoid EMI, which means gate current isn't that high. \$\endgroup\$
    – bobflux
    Commented Dec 3, 2020 at 17:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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