sizing a freewheeling diode for a coil

Jamie wrote in news:Mxk8i.295$ snipped-for-privacy@newsfe06.lga:

You're mother is a retard.

Do tell.. how would that work?

Best regards, Spehro Pefhany

I think you mean this one:

-- Paul Hovnanian mailto: snipped-for-privacy@Hovnanian.com

------------------------------------------------------------------ In theory, practice and theory are the same, but in practice they are different -- Larry McVoy

I've seen 1n4148s fail across small relays, the 'off hook' relay in a modem. The cause? Surges in the switched circuit. Big ones. The loop area of the relays' contacts and connections--as laid out on the PCB--was sufficient to couple said surges into their coils & pop

This modem was in my PC. After losing several diodes in succession, I fixed the problem by 'improving' the layout to reduce the loop area & proximity to the coil.

Designing industrial stuff, we picked relays wired such that the one- turn switched loop wouldn't couple back to the coil, and laid out traces carefully on both sides exactly to avoid this problem. Otherwise, lightning-induced surges can make quite a mess of things as they propagate backward from the relays and onto your board.

Cheers, James Arthur

That's further off track than I'd expect from you Paul. What has this to do with suppression diodes?

Robert

last I checked 1N4148 was rated for 200 mA continuous.

Bye. Jasen

you get contact bounce on the relay.

Was that suppose to be a joke?

Not really, I really have no idea what s/he is talking about in terms of potentially harmful heating. Do you?

Best regards, Spehro Pefhany

And?

Robert

The person you need to blame this on is "Lamey The Cable Guy" or whatever f****ng stupid nym he is going by now.

The posts is a forgery/clone or whatever you want to call it. I did not write it. That goes for the next one you read as well. I was never in this thread.

Backspacing the URL to "appnotes/" presents a selection of pdf's.

Thanks for all the helpful replies.

It seems that the overall reliability of the device is best served not by a single diode, but by a transient voltage suppressor. The diode is a little better at protecting the transistor that switches the coil, but apparently it can cause a not-so-clean unmaking of the relay, leading to welding or erosion of the contacts in the case of a heavily-loaded normally-open relay. This is described in

If the contacts operating the relay become dirty or the control transistor begins to oscillate, then you will get repetitive surges at quite frequent intervals. It is better not to rely upon the single-pulse surge current rating of the diode.

A diode with a continuous rating equal to (or greater than) the maximum coil current will give peace of mind under all possible conditions of operation and mis-operation.

I did not mention the single pulse surge current rating, but the repetitive surge surge current rating. But as with any limiting case, it is a good idea to include some safety factor. By the way, if the coil is driven by rapid pulses, its current will not reach the full, DC, steady state value, but will bounce between some lower limits.

Agreed. Especially if the diode is in a good thermal situation.