Best path to eliminating or reducing RFI / oscillations?

[LessPaul]

Adjusting the volume of my Carvin X-amp makes oscillating sounds like tuning a radio, but without radio content. I can "tune" to a volume with no interference, but where it is present, the amp's audio is distorted.

The input is a typical chassis-grounded jack with a 1M resistor to ground, and connected directly to the grid of the first 12AX7 preamp stage. With the input jack grounded, there is no interference, and I am using a quality cable from a shielded guitar.

I understand that there are lots of RFI sources other than radio, such as light dimmers, switching power supplies, etc.

What would be the best approach(es) toward taming this interference?

[mozz]

A grid stopper resistor directly mounted on the tube socket. Anyway, that is standard/accepted practice, if there are better ideas nobody seems to update that.

[Helmholtz]

A 33k grid stopper certainly is a good measure. Any input triode should have a grid stopper of at least 10k.

Do you have a scope? It helps to know what frequencies are involved,

Please post a schematic.

Was the amp always acting like that?
Have you tried a new input tube?

Can you produce the effect with a signal generator connected to the amp's input?

What effect has rolling back the guitar volume?

[Chuck H]

I think it's implied by the opening post that the perception is RFI getting into the amp. I don't think this is RF getting into the amp though. The description sounds (to me) like HF instability. That is, "the phone call is coming from inside the house". As it were. It's probably cross talk between the amps own circuits causing a positive feedback loop. This is usually managed by installing grid stops and/or low value caps that trim a little HF to ground at some strategic place in the signal path.

But an important question that Helmholtz asked needs to be answered. Has the amp always done this? And I would ask another question. Has the amp been modified?

[Tom Phillips]

I second Chuck's thoughts.
I'll add that the problems described can also be caused by faulty filter/decoupling capacitors.
The last time I fixed a similar problem in an amp it was a broken lead on a preamp decoupling cap. With the cap open circuit it allowed positive feedback in the preamp. When this happens it often causes erratic oscillations that are affected by various settings of the volume and tone controls.

[Chuck H]

I second Chuck's thoughts.
I'll add that the problems described can also be caused by faulty filter/decoupling capacitors.
The last time I fixed a similar problem in an amp it was a broken lead on a preamp decoupling cap. With the cap open circuit it allowed positive feedback in the preamp. When this happens it often causes erratic oscillations that are affected by various settings of the volume and tone controls.

I just looked up a bunch of X amp schematics (there are many) and it seems the earlier 100's did take care to keep like phase stages on separate PS nodes. Then later (mid 80's?) the lower wattage X amps did have three cascade gain stages on the same PS node. They might have had some trouble with that because around '90 they changed things. They still had three cascade stages on the same PS node but they added a series resistor for the first and third stages. The third stage does have a .47u cap to ground. I wouldn't consider this value a DC filter so it's probable it was put there just to decouple HF from that PS node.

[LessPaul]

I really hate making a stupid mistake, but I did, and I have returned to confess my sins:

"the phone call is coming from inside the house". As it were. It's probably cross talk between the amps own circuits causing a positive feedback loop

This is indeed what was occurring.

.Has the amp been modified?

That's the $10,000 question. When I got this amp, it had been heavily modified by a previous owner, and had all kinds of stability issues. This particular amp is a rev-b, 1982 X-amp. Using the schematic from the original 1981 X-amp, I slowly reverted the amp back to how it likely was. It has been a bit of a guessing game when I encounter a changed value component or a piggybacked part -- owner mod or Carvin change?

In any case, the output driver of the reverb is a typical non-inverting op amp. (schematic attached) When I was de-modding the amp I noticed a 220pF cap, not on the schematic, in parallel with the feedback resistor of the op amp. A cursory check showed no real changes, so I took it out.

At this point, you are probably way ahead of me.That cap prevented the reverb output driver from amplifying unwanted high frequencies. Its removal caused the problems I was encountering. I reinstalled the cap and all is well.

Thank you to all who contributed and helped me identify my cockpit error.

One positive: While chasing this, I finally got fed up with the crumbling "cable winding feet" on my Tek 435B and 3d printed a new set.

Carvin_x100a_1981_Preamp.pdf

[Helmholtz]

In any case, the output driver of the reverb is a typical non-inverting op amp. (schematic attached) When I was de-modding the amp I noticed a 220pF cap, not on the schematic, in parallel with the feedback resistor of the op amp. A cursory check showed no real changes, so I took it out.

At this point, you are probably way ahead of me.That cap prevented the reverb output driver from amplifying unwanted high frequencies. Its removal caused the problems I was encountering. I reinstalled the cap and all is well.

What do you mean with reverb output driver? Do you rather mean the recovery amplifier as the reverb driver drives the input of the tank.

Caps in the NFB path are typically used to improve stability against self-oscillation.

[LessPaul]

What do you mean with reverb output driver? Do you rather mean the recovery amplifier?

Caps in the NFB path are typically used to improve stability against self-oscillation.

I apologize for my incorrect terminology: I am referring to the op amp on the output side of the reverb tank as shown in the schematic.

The reason I removed the cap (clearly a mistake) was because it was a modification, piggybacked on the 100k feedback resistor. There were no traces for it, and it was not on the schematic.

[Helmholtz]

Ok, that cap also limits the HF bandwidth of the recovery amp.

[Helmholtz]

The reason I removed the cap (clearly a mistake) was because it was a modification, piggybacked on the 100k feedback resistor. There were no traces for it, and it was not on the schematic.

Don't worry. That "mistake" might have happened to anybody trying to restore the amp to original condition.

Glad you fixed it.