Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!uakari.primate.wisc.edu!dali.cs.montana.edu!milton!whit
From: whit@milton.u.washington.edu (John Whitmore)
Newsgroups: sci.electronics
Subject: Re: Current shunting
Summary: Perhaps a SCR-output optoisolator?
Message-ID: <11202@milton.u.washington.edu>
Date: 15 Nov 90 23:46:26 GMT
References: <968@eplunix.UUCP>
Organization: University of Washington, Seattle
Lines: 28

In article <968@eplunix.UUCP> raoul@eplunix.UUCP (Nico Garcia) writes:

> We have a set of current stimulators... electrically isolated (running
>off separate batteries) and drive current signals from 1uA to 1 mA
>through nerve electrodes. We also have a 1K resistor and a 1uF AC
>coupling cap between the output and the electrode, so it goes like this:
>
>Iout---->Rtest---->Cap---->Electrode---->Ground
>     |         |       |             |
>    Lead1     Lead2   Lead3         Lead4
>
>We would like to put a current crowbar in: a circuit that would
>detect current above a preset level (positive or negative) and
>shunt Iout directly to ground, (to tie Lead1 to Lead2) and only 
>release that shunt when a panel button is pushed. Those current
>levels would be between 10 and 1000 uA.

	Would an SCR crobar suffice?  You can get SCR-output optoisolators
that would allow you to put a completely isolated crowbar on each
circuit.  If the holding current is as low as your current source,
this would take a single pulse to turn "on", and would remain shunted
until Iout was removed.  If that won't work, you could use a darlington
optoisolator and just keep the LED input turned on (circa 1 mA).
	Either way, you'd have to have an op amp to detect the fault
and a latch or a one-shot to trigger the shunt.

	John Whitmore
	whit@milton.u.washington.edu