Pots in opamp feedback networks, why not?

[airraid]

I've googled the web for sample microphone preamp designs using opamps, and I've found that most have either discrete gain settings via rotary switches, or a high fixed gain and a potentiometer attenuator at the output. What I'm wondering is: is there a good reason why potentiometers are almost never found in the opamp's feedback voltage divider which sets the amplifier gain?

In my breadboard experiments, I've noticed problems with radio stations and hum getting into the amp when I've used metal-shaft pots, or long antenna-like wires to the pots, but this is with unshielded parts (and crappy parts), so I'm not sure if these problems carry over to final designs. When I stick a panasonic plastic-shaft pot right down on the breadboard, I can turn it without these problems, and when using conductive plastic pots, without any scratchiness either. It leaves me wondering if it's OK to do it, if you've got the right type of pots.

-Adam

[sss]

volume controll in the feedback loop

[cunningham]

After a while, a potentiometer wiper will become corroded, rub off the resistive material, which is usually just carbon like as in a pencil, and create noise. Imagine how much noise this would make when in control of the gain of a divice that has such a high open loop gain as a OP-AMP.

[peranders]

A potentiometer in the feedback is OK as long as you use quality parts, has a quality design and of course has a good enclosure, otherwise you will get problems like those are mentioned.

[Centauri]

Most importantly, DON'T have a DC current path through the pot - decouple the pot with a high value capacitor, and use a high value resistor to provide DC coupling from output to inverting input. DC current through the wiper of a pot WILL cause noise.

Cheers

[thylantyr]

You will find designs where a pot is used in the feedback network
to adjust gain, ie, the front end of a preamp to properly match
levels with a source. I wouldn't use this idea for a volume knob

[airraid]

Thank you all, that is what I needed to know. I am not proposing it for a volume knob, I'm building a mic preamp and what I need to do is have some way of setting levels *once* before recording begins. I found several designs on the web that have a fixed high-gain stage followed by an output attenuator, but this seems to me like it would trash the SNR unnecessarily if used for getting only a few dB gain (which is often all I need to bring electret mic levels up to LINE IN).

[HFGuy]

Having a pot inside an op-amp's feedback loop isnt a great idea. Because the pot will be mounted mostly likely on the faceplate your going to have a longer feedback loop. The feedback loop will be long and have a high resistance, it will act as an antenna. The shortest feedback path is ideal, this is why many ppl use SMD and solder them directly to the op-amp's pins.

[mikeks]

Originally posted by airraid
....is there a good reason why potentiometers are almost never found in the opamp's feedback voltage divider which sets the amplifier gain? -Adam

Actually, a pot. in a double inverting feedback network is an exceedingly good way of generating a quasi-log. law volume control from an ordinary linear pot. (due to P. Baxandal)...

Two amps. are needed to ensure gain (and therefore THD+N) at the extremeties of the pot's excursion is well controlled.....

Performance is in fact often superior to that available from the best so-called log. pots. in pratice....