Signal bleed with volume at zero

[garytoosweet]

Amp is Mod101. Kit amp that I purchased from someone who built it non-functioning. I have re-soldered every connection, verified all wiring/connections. Amp is now working, all DC voltages match schematic. However 1 problem remains, there is a fairly loud signal passing through when volume is at 0.

With a 100mv sine wave in input, the signal at speaker out is 4V pk-pk.

The leaking signal is not effected by any of the EQ controls. EQ works fine on regular signal once you turn the volume up.

I've verified the volume pot is functioning ok. Checked all grounds multiple times. Tried swapping V1.

I've tried grounding grid and plate of V1 and found that grounding V1a grid, or V1b plate kills the signal. To me this infers that the issue lies between V1a and V1b in the eq circuit, but I've double checked everything and all seems fine. Not sure where to go from here?

Grounding wiper of volume pot does not kill signal
Grounding v1a grid kills signal
Grounding v1b grid does NOT kill signal
Grounding v1a plate does NOT kills giant
Grounding v1b plate does kill signal

[The Dude]

My first thought is a bad or miswired decoupling/filter cap on the "E" node. 1) Make sure wiring is correct for that node. 2) If you have a scope, see if you can see audio riding on that node. 3) Temporarily clip another filter cap on that node to see if that fixes the problem. It could be lead dress or a number of other things, but that's where I'd start.

[Enzo]

We just had a thread with the same complaint on another amp, you might rad through that for all the techniques and such.

[Jon Snell]

In your case separate pins 3 & 8 on V1. Use independent cathode loads. Could be crosstalk problems due to leakage between the common cathode load.

[Chuck H]

My first thought is a bad or miswired decoupling/filter cap on the "E" node. 1) Make sure wiring is correct for that node.

In your case separate pins 3 & 8 on V1. Use independent cathode loads. Could be crosstalk problems due to leakage between the common cathode load.

Along the same lines as what The Dude is saying, but WRT what Jon mentions next. With a fully bypassed common cathode there wouldn't ordinarily be significant cross talk. A miswire of the common cathode, a grossly bad bypass cap at that node, any elevation of the 0V reference there or a change to a smaller value cap sure could cause cross talk. So along with examination of the decoupling circuit in the power supply rail you could also check the preamp tube cathode circuit for errors. Also mentioned by The Dude, you can use a scope. See if any audio is present on the preamp tube cathode.

[Helmholtz]

Grounding v1a plate does NOT kills giant
Grounding v1b plate does kill signal

Not sure how you did it, but never directly ground circuit points that carry DCV (e.g. plates). Use a 0.1µ cap to ground only signal.

Search this tread:

"Signal leaking through with volumes off"

[garytoosweet]

Thanks for all suggestions. I did previously go through the recent thread dealing with a similar issue which was a conductive board.

Plates were grounded with a .1mf cap as Helmholtz suggested.

I verified wiring for E node and all filter caps. I checked it with a scope, again with a .1mf cap in series with the probe, and could see no AC signal. I paralleled a new 22uf cap just for good measure and it had no effect.

Then I separated the cathode loads, which solved the issue completely. MOD has this 'modification' as a pdf on their website. https://www.modelectronics.com/sites...hode_split.pdf

It also solved an issue where the amp would break into oscillation when the volume was up past 8 or 9 and the treble was up in the same area. Seems quieter overall as well.

Thank you Jon Snell, right on the money.

Would there be any reason to share a cathode load other than saving a few cents?

[Helmholtz]

Would there be any reason to share a cathode load other than saving a few cents?

Actually no, except maybe space.
Shared cathodes between different channels are widely used and typically cause no problems. At least as long as the bypass cap is good.

---

[Dave H]

It's a really bad idea to use a shared cathode resistor like that with cascaded gain stages as without the cathode cap the circuit is an oscillator (because there's positive feedback via the cathodes)