At 10:02 PM 2/24/2008, you wrote:
>Spehro Pefhany wrote:
> > Do you *really* need to lock it out for seconds? A shorter time, such
> > as 100msec would be a lot easier-- just a pot + R from OCDa to
> > discharge aceramic cap of say 47nF and (eg. 100R pot + 10R) and charge
> > the cap through a 2M resistor to Vdd. That would give you a lock out
> > time constant of about 100msec, and a trigger time constant of about
> > 5usec +/- 2usec. No active parts at all.
>The problem seems to be that 100msec is too low...  The part will self
>destruct with a 22uf /22K combination after 5-10 seconds when dealing
>with a dead short.   I would expect that the times with a 47nf/2M
>pair+~110on OCDa (dead time of about 1/5 of the 22uf/22K pair) to be
>even more dismal from my testing.   The 22uf seems to be about the
>smallest cap I could deal with unless I added a R to OCDa, in which case
>it basically permitted me to use a smaller cap, but with the same effects.
>
>I think I'm actually at the point of saying "trigger the overcurrent
>detection, game over, require power cycle"... which might be easier to
>deal with, since a short should never happen, and if it does, requiring
>a human to figure out why is probably a good thing.
>
>-forrest

Interesting that the built-in thermal protection is not protecting the
part. You might want to find a way to test your proposed method before
taking it too far.. I smell trouble (or maybe that's burning epoxy).

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist