When selecting a catch diode for a non-synchronous buck converter, is there way (or even a general rule-of-thumb) to determine a safe current rating? Most diode spec sheets will show the max average current (based mainly on thermal capabilities) and usually a non-repetitive peak current, often 10-50 times higher than the average. But in a switching converter it will be repetitive, and I have very rarely seen this specified.

If for example I expect 800mA output from a regulator, and I expect up to 1.6A peak current through the diode, can I select a diode with a 1A (max. average current) rating? What is safe to assume about the max peak capability? Of course a 2A diode would be safest, but is it necessary?

  • \$\begingroup\$ Often a beefier diode will have a reduced forward voltage at the currents you are seeing helping with efficiency (the reason why I've chosen a 850 mA diode for a 65 mA design). Can you calculate (or simulate) the worst case average current through the diode? I'd take that as a minimum. \$\endgroup\$ – Arsenal Mar 21 '17 at 14:15
  • \$\begingroup\$ I would call it necessary. At a 20% derating the diode will be hotter than it should be. I like 50% derating minimum. \$\endgroup\$ – Robert Endl Mar 21 '17 at 14:53

You have to look at the duty cycle of the converter, which in continuous conduction mode is ~ D = Vout/Vin. Then the diode is carrying the output current for (1-D) of the time. So take the max output load current times (1-D) (at worst case Vin) and add appropriate derating.

Then check your thermals.

Diodes can typically handle peak currents well in excess of the average, so the peak current in a buck converter isn't usually a concern.

| improve this answer | |

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.