As has already been mentioned, the problem will be what happens as the supply turns on. In theory, as VCC ramps up, the 431 will start to regulate and so limit its own cathode voltage. If VCC exceeds 36V before the 431 has started drawing current then all bets are off. Crude but you could try putting a zener in series with R3/NPN_B and 431 cathode to drop the excess voltage so that VCC - Vzen < 36V.
Any series combination of components from VCC to the cathode will suffer from this problem. The VCC will be present at the cathode if nothing is conducting, right. So what makes them break? If it's just a potential thing, like mosfet gate punch-through then this can never work. If it's the power dissipated in the part then I guess it won't matter until it really starts to conduct and this method should be safe.
And another: they are notoriously unstable (and the spice models don't always show it) with certain cathode loads.
Yep, I may have TI's model and the thing goes unstable with increasing load capacitance. That was a surprise. The feedback capacitor suggested helps, but does not eliminate the problem.
> All in all, may be more trouble than it's worth.I agree and have given up :) Instead, I'm using LM317, and some extra stuff to drop the voltage down to its input. Not so bad on part count.
Cheers,
Andy.
thanks, Andy. gene _______________________________________________ geda-user mailing list geda-user@xxxxxxxxxxxxxx http://www.seul.org/cgi-bin/mailman/listinfo/geda-user