Gain pot instead of volume pot

[boyofwax]

Hi all ! Currently I'm experimenting with opamp's to create a preamp to drive my FET monoblocks. I'm planning a gain pot (feedback) instead of a volume pot for several reasons. My PC and CD player have a digital volume control. To keep out the output coupling caps (my monoblocks have 1). I'll use 2 separate pots for left and right in order to have some balance control. Has anyone done this before ? What are the (dis)advantages ?

[GRollins]

Assuming that I'm reading your post correctly, you want to vary the gain by varying the feedback. This idea gets re-invented about once every six months or so. The problem is that by varying the feedback, you're varying the distortion. Will it work? Yes. However, the amp will sound subtly different at different volume levels, and you could conceivably run into stability problems. You'll also find it difficult to get infinite attenuation (zero volume).

Grey

[Steven]

This has been done before. Douglas Self uses it in his Precision Preamp, based on the well-known Baxandall circuit. Advantages are more headroom and the possibility of better channel tracking of a stereo potentiometer because the gain depends on the resistance ratio of track sections instead of absolute values. Disadvantages are already mentioned by Grey.

See:
http://sound.westhost.com/project24.htm or the book "Self on Audio" or Self's Precision Preamp in Wireless World 1996.


Steven

[Centauri]

In addition to the above, do NOT allow the pot to provide a DC path - scratchy operation WILL result.

Cheers

[dreamer]

This kind of volume control, or gain control is used
in MF's A-100/A-200.
They sound different at different volume due to
something like Grey said, change of NFB
at different volume, change the distortion too.

One more problem of that circuit is......if your VR(volume)
fail and become open circuit, your speaker may die with it!
Which may not happen in normal design.

Roger

[Eva]

Changing the amount of NFB on a power amplifier is evil since gain has to be higher than a certain value to ensure stability

This minimum 'stable' gain is usually the one used on the circuit so reducing it is not reliable

In the other hand, you can increase that gain but THD of the amplifier will increase proportionally to the gain increase so it's not the best solution

To circumvent all those problems, you could simply use an op-amp variable gain stage just before the power amplifier. Most Op-amps are unity gain stable and have THD below .005% at unity gain
[If you have some hum, you could also add a differential input stage]

Finally, if you use a potentiometer in a feedback loop, try to place a high value resistor between OUT and IN- in order to provide some DC feedback in case of potentiometer failure. Also don't allow high frequencies in a feedback loop to go through any potentiometer and its leads [ie: place a small capacitor 20-100pF between OUT and IN-]