[sdiy] SSM2164 protection

[David G. Dixon]

> Hi David,
> 
> > So, I decided it would be a hassle to try and retrofit 
> negative-rail 
> > protection to all the 2164s in my system.  Also, the idea 
> of putting a 
> > resistor on the +V pin gives me willies, and I don't relish 
> the idea 
> > of having to lay out special power connections for that 
> chip on all my 
> > designs.
> 
> Ok, so you protect at the rack level, and assume that the 
> individual power connections to your boards never, ever fail, 
> and that when you fit them you will never, ever make a 
> mistake.  If these modules are for your own enjoyment then do 
> whatever you feel happiest with.

Well, obviously, I get it wrong sometimes with the cable, but that's a
"one-shot" sort of deal.  What I'm worried about is something not under my
control happening and wiping out all the 2164s in one fell swoop during
otherwise "normal" operation, or perhaps during power-up.

> If, however, you're designing modules for production that 
> will travel the world, to be fitted by customers who are not 
> you, then I would suggest taking a more pro-active stance.

Yes, excellent advice.  And I'm actually in this situation now with a
company, so it's certainly something we'll be looking at.  However, I can
take some comfort in the fact that neither the Mankato nor all those
2164-based VCA units sold by Doepfer and others have no special protection
for the 2164 chip, so I suppose this problem doesn't arise very often.
Still...

> In light of these recent discussions I will be reviewing all of my
> SSM2164 designs and adding some sort of protection scheme, 
> either a plain resistor or something more involved (PNP + 
> zener), for the next board revisions.

Yes, again, you're probably right.  Oh, and by the way, I simulated the
zener and it only pulls about 1mA.