Interpret this 12ax7 feedback loop

[NoviceOU812]

Take a look at this subcircuit--

It's part of a preamp circuit. The input follows an opamp, which handles the guitar input impedence issues. The output feeds another opamp, which handles any output impedence issues.

So, when the Pot1 is turned full off, tube1 works as a paraphase inverter. It doesn't amplify the signal, but only inverts it.
When the Pot1 is turned full on, tube1 works as gain stage. Here it inverts and amplifies the signal.

Does this subcircuit have a name? Have you ever used it?

[Miles Prower]

You could consider that to be a basic, hollow state op-amp that uses variable negative feedback to set the gain. It avoids putting a potentiometer in the signal path which will also change the high frequency roll off as the impedance it presents varies with the slider position.

[Brian Beck]

But why is there a DC supply (either +7VDC or 47 VDC - it's hard to read) connected to the right end of the pot? No DC current could flow into this circuit due to the two blocking caps, so what's the point of that connection? I think we're missing some pieces to the jig saw puzzle here.

[NoviceOU812]

Brian, The +7v dc comes out of an opamp, which essentially removes any ac component fed into the +7v dc line.

Miles: I think you've got it. So, this is the gain subcircuit. But, what would this do... if I swapped the 330K resistor for a 500K variable resistor (i.e., a pot with the third peg unconnected), would it work (inadvertently) as a lowpass filter? It would surely work to change the amount of negative feedback in the loop, but you're saying it would also change the high-frequency roll-off?

[NoviceOU812]

Like this::

[Miles Prower]

Originally posted by NoviceOU812
Brian, The +7v dc comes out of an opamp, which essentially removes any ac component fed into the +7v dc line.

Miles: I think you've got it. So, this is the gain subcircuit. But, what would this do... if I swapped the 330K resistor for a 500K variable resistor (i.e., a pot with the third peg unconnected), would it work (inadvertently) as a lowpass filter? It would surely work to change the amount of negative feedback in the loop, but you're saying it would also change the high-frequency roll-off?

That won't accomplish anything. Take a look at #1 on attached schematic. This RC network functions as a high pass filter. The critical frequency of this is: f(c)= 3.2Hz. As to what it does, this looks like low frequency compensation to stabilize the circuit to prevent low frequency oscillation. Replacing the 330K resistor with a variable won't do much. Set that control too low, and you're likely to see oscillations and/or ringing. Not a good idea.

As for #2, that has a very low frequency f(c), since all it does is block DC, and it's desireable for it to be as close to an AC short circuit as possible. That's why that capacitor is so huge (and a likely culprit for low frequency instability).

[Brian Beck]

OK, but I still don't get the +7 volt part. How is an opamp connected? Do you have the rest of the schematic?

[NoviceOU812]

Here's the whole circuit, to show where the 7v comes from. I'm regarding it as an ac ground, (dc +7) -- but if it strikes you as something else, let me know.

Great info about the 3.2 Hz oscillation. LTSpice was doing that -- I thought it was a good thing (i.e., feedback that would sound good on a guitar signal). But I see how it wouldn't be all that desirable now.


Next question: I understand wirewound and film resistors to be quieter than carbon comps. My understanding is also that polyproplyene/polyester/teflon/etc caps are more transparent than electrolytic. So, if I were to improve the quality of the preamp, i'd upgrade these caps and resistors:

R2, R6, R5, R11, R10, R22, P2, R23, R39, R24 and R25, and
C5, C1-4, C17, and C10.

(C12, C14, and C16 are already polyproplyene, and C11 & C10 are little tantalum guys.)

Are these the resistors/caps that introduce noise into the circuit?