[sdiy] Serge and multiples

[Martin Czech]

:::Most folk worry about inputs. Bipolar processes are pretty good
:::for static handling. If you were REALLY leery, you could split that
:::series output resistor, and use MOV, zener, or fast diodes to the rails to 
:::either eat the static discharge, or direct it to a place where it can 
:::dissipate harmlessly.

That would be a good idea.

:::
:::But I've never seen a bipolar opamp die of static discharge at the output, 
:::or even degraded operation. This is more likely if you are using a voltage 
:::follower (output tied to inverting input) than a
:::unity gain inverting stage (with resistors in the feedback).

This is no miracle. Even if the output stage is severely damaged,
you wouldn't notice, because the servo loop in closed loop
applications will hide it. An ota will be diffferent.
Depends on the level of leakage. 1uA may wreck something, other
applications will still work with 10uA....


:::Still I would not worry... I'm not landing my synth on MARS if you know what 
:::I mean  ;^)

If you build it for your own fun, certainly. If you want to sell a good
product, you might consider esd protection. It is not SO expensive,
after all. (most vendors will not consider this, because you can save a
few $0.01 the customer will not know about. what a bargain!).  I aggree
that this topic is somewhat academic. OTOH a good clamping will protect
any i/o against any fool, not only nature. If you have children, apes
or other curious folks in your studio, or if you work very late until
you tend to goof something up, crowbars, flase polartity, overvoltage
protection and i/o protection might be not so stupid. 
How many synth chips die because of blown regulators?
How many circuits have I fried with flase polartiy (a couple!).
How many i/os have I killed during "experiments" with 
sloppy cable routing (also a couple).


The problem is always sitting in front of the hardware...


m.c.