0
\$\begingroup\$

This seems like it should have been asked before, but I can't find anything. I've got a large 12v DC motor (11A startup surge, 7a running current) that I'm PWM-ing at ~20khz through a mosfet.

I, naturally, need a flyback diode to keep the highly inductive motor from doing naughty things with my mosfet.

Up to now, I've been using the CDBA540-HF (5A) as my kickback snubber, and I've never had any problems with it.

I've heard that you should match your motor's max operating current with the diode current. However, this diode has a peak current of 125A. Does this mean it can handle my motor easily?

\$\endgroup\$
  • 1
    \$\begingroup\$ Will the diode handle a continuous stall scenario? I don't think it will. Can you 100% rule out this scenario? \$\endgroup\$ – Andy aka Jun 14 '16 at 12:35
  • \$\begingroup\$ @Andy Nope, I can't rule out a stall. \$\endgroup\$ – 0xDBFB7 Jun 14 '16 at 12:44
  • \$\begingroup\$ @Andy Do you think I should get a diode that's got an average current equal to the motor's current? Reading online, I get this quote from a paper: "A diode that has a peak intermittent current specification that exceeds a motor’s stall current will usually suffice." I've got a peak current that 12 times the max current. Do you think it'll survive in the long run? \$\endgroup\$ – 0xDBFB7 Jun 14 '16 at 12:48
2
\$\begingroup\$

You have to meet all your diode specifications, average current, peak current both have to be met. Neither is more important than the other, though in a specific application, it may be one or the other that fries the device.

You need to meet average current for average dissipation and so case temperature rise.

You need to meet peak current for max die current, max bond-wire current, and maximum pulse temperature rise of the die above its steady state 'running' temperature rise due to the average current.

Looking quickly at your application figures, 7A at slightly less than 100% duty cycle (delivering very low voltage at stall) looks like it exceeds your 5A average rating. At low case temperatures, and with a generous spec by the manufacturer, you have got away with it. 11A versus 125A will certainly be OK for peak (as long as you respect that 8.3mS time limit, you saw that, didn't you?)

In your shoes, I would be grateful that I'd not blown anything up so far, and choose a beefier average rated diode for production / long term use.

\$\endgroup\$
1
\$\begingroup\$

If you PWM a stalled motor at 50% the diode average current could be over the 5A rating, even assuming the 11A surge is right- sounds a bit low to me if the run current is 7A.

The diode current will not be higher than the motor current, so the peak rating should be okay.

\$\endgroup\$
  • \$\begingroup\$ Is the average current that important for a kickback snubber? It's only getting that current for milliseconds, each PWM pulse, right? Am I right in thinking that the peak current is the important value? \$\endgroup\$ – 0xDBFB7 Jun 14 '16 at 12:58
  • \$\begingroup\$ Both are important. The motor current flows alternately through the diode and the PWM switch. Typically the PWM frequency wrt motor inductance is high enough that the motor current has only a bit of ripple. So the full motor current flows through the diode during the PWM 'off' time. \$\endgroup\$ – Spehro Pefhany Jun 14 '16 at 13:45

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.