(2) dirty/clean ground again [sdiy]

[René Schmitz]

Hi Harry,

>I submit (to the gentlemen and Cynthia of the jury ;^)
>
>That very often the loads on an opamp DO indeed go to ground when using
>a bipolar supply...and they are often greater than picoamps.

Yes, they are, but that doesn't mean one shouldn't try minimizing the amount 
of ground current. 

>Case in point:  A potentiometer used as a "volume control" in a mixer circuit.
>
>We use a reasonable value of summing resistor, maybe 100K so as not to get
>too much noise... so that dictates a pot of maybe 1/10 that value so loading is
>not
>an issue... so 10K to ground.  We could be in the milliamps here...

Some possible cures, if that would be a problem, or if one wants to do some 
over-design: Returning the pot to a "ground follower" instead of GND. 
Or choose a higher impedance pot, and place a unity gain buffer between the wiper
and the summing stage. Advantage could be that you can use a 100k pot and a 10k 
summing resistor, which hopefully compensates for the noise of the extra opamp.
(In extreme cases one could use the dirty ground, but use an instrumentation amp
to get at the voltage from wiper to GND.)

Well, I don't say that its a must to to this, but just want to say that even in
this case there are ways of minimizing (or eliminating) ground currents by design. 

>I say decouple positive and negative busses to ground with caps. Better if you
>can stand series resistance in the supply lines as well...

As has been said before, there is no one-size-fits-all solution. To decide what to
do in which situation one needs some insight what is going on and what options are 
available, that will dictate what the best solution is.

Cheers,
 René




-- 
uzs159 at uni-bonn.de
http://www.uni-bonn.de/~uzs159