Capacitive coupling?

[GainFreak]

Hi,

I have an amp that has a clean signal leak with all Gain and Volume controls down (2 channels with separate Gain and Volume controls and common TS, PCB less the tubes). It's well audible when it leaks into the drive channel at low gain settings and is very annoying. It looks like the last triode grid (just before a non CF tone stack) is picking up the signal. I know it would be specific for every amp, layout etc but if this is capacitive coupling what general approach and rules would you recommend for handling it ?

[xtian]

Is this a hand built amp with wires you can push around with a chop stick? You should be able to notice a difference in volume as you move wires around, if the issue is capacitive coupling.

[Helmholtz]

General countermeasures are:
1) Increasing the distance between the parts and wires that couple,
2) Shielding.

Of course, this answer is just as vague and general as your question.

[Chuck H]

It's not likely to be capacitive coupling UNLESS you actually have signal leads for multiple channels sharing a multi conductor shielded lead AND the unused channels are not grounded when not in use (rare but not unheard of). This IMHE. More likely something about how the circuit is wired, poorly performing decoupling caps on nodes shared by multiple channels, a less than ideal ground or even a funky tube. I've built quite a few custom design amps and had to trouble shoot problems like this many times. It's never been capacitive coupling.

We need a schematic to do better than this.

[GainFreak]

Moving/touching wires doesn't change anything. Tried to decouple that stage separately without result. I'm waiting for some documentation on this amp and will let you know with more detail. It's a hand built PCB amp, the tubes' sockets are mounted separately and wires run to the PCB.

[GainFreak]

Attached is the schematic and layout of the amp. Only when V3B pin 7 is grounded signal leak is gone (all Gain And Volume controls down).

MC II.pdf

[Chuck H]

Check relay K4 (and K5 which is switched via K4). Check C30. Possibly parallel a known good capacitor at least 10uf and of the same voltage rating and see if that helps. And check for good continuity and contact at the preamp ground via the board standoff.

[Jazz P Bass]

If everything on K4 is working properly, my only thought is the signal is coming in on C3 (V1A)

[trobbins]

Perhaps to further assist faultfinding, short the wiper at VR2 and at VR6 to its local gnd, depending on channel/relay operation, to see if you get the same leakage level for either channel setting.

[Pedro Vecino]

I see that the clean channel muting is done on its volume control. You can try muting the signal right at the relay input and see what happens.

[GainFreak]

Thanks for all your suggestions. Below is a list of things I did and that didn't work:

1/ replaced C3 capacitor (with 2 different types
2/ replaced R50 anode resistor
3/ paralleled a good cap with all the anode decoupling caps one after another
4/ created a separate anode node from R60/C30
5/ all relays worked properly
6/ grounding Gain and Volume pots

After spending several hours was about to give up on this one but from previous experience with the SLO/Mesa Gain 470k grid resistor decided to short R69 (K4 relay). Bingo! No more signal leakage (with all Gain and Volume pots at zero). There's some signal leaking when on channel 2 (with Gain to zero and volume up) but it's nothing like what it was before. When Gain is engaged that minor leak is lost into the main signal and doesn't dominate the sound anymore.

[trobbins]

Just to confirm - with Lead channel selected, there is some minor signal leakage with Lead Gain at zero, and Lead Vol above zero, and with R69 shorted? But with no short on R69 there is quite noticeable signal leakage when Lead Vol is above zero ?

Is it possible to temporarily add a 100k between the wiper of VR6 Lead Vol and R28 as a fairly short linked path? Then to check for leakage when the link from VR6 wiper to K4 is disconnected, and similarly when the link from R28 to K4 is disconnected ? Does the wiring to K4 from VR6 and R28, or the wiring to R69, go near to the Vol 1 and Vol 2 wiring connecting to the other contact set of K4 ?

[GainFreak]

Just to confirm - with Lead channel selected, there is some minor signal leakage with Lead Gain at zero, and Lead Vol above zero, and with R69 shorted? But with no short on R69 there is quite noticeable signal leakage when Lead Vol is above zero ?

Yes.

Is it possible to temporarily add a 100k between the wiper of VR6 Lead Vol and R28 as a fairly short linked path?

The leak is as bad as with the R69 not shorted.

Then to check for leakage when the link from VR6 wiper to K4 is disconnected, and similarly when the link from R28 to K4 is disconnected ?

Still bad.

Does the wiring to K4 from VR6 and R28, or the wiring to R69, go near to the Vol 1 and Vol 2 wiring connecting to the other contact set of K4 ?

Not really. They are both at zero anyway.

So far only shorting R69 produced acceptable results where the leak doesn't dominate.

[Chuck H]

...decided to short R69 (K4 relay). Bingo! No more signal leakage (with all Gain and Volume pots at zero).

So far only shorting R69 produced acceptable results where the leak doesn't dominate.

I think you're reporting is too vague. When you say "short" or "shorting" in context above I assume you mean grounding? Because shorting R69 would simply be a jumper across the resistor You don't mention ground in either report of shorting R69.

Further, if you ARE actually "grounding" R69 then does it make a difference which side of R69 you are grounding? There are two ends on the resistor and on which one signal is eliminated could be important. For example, turning VR6 all the way down effectively grounds one end of R69. If turning VR6 all the way down is not giving the same result as "shorting" R69 (to ground?) we really need more clarification on what you're doing. And...

I would expect grounding (shorting?) the output end of R69 to stop the signal bleed as well as any other signal from the preamp. So there is no solution to be found examining how grounding all signal late in the preamp chain eliminates the problem. Because it effectively eliminates the whole preamp.

Above you say "6/ grounding gain and volume pots" and I don't know what this means. Does this mean just turning them down? Did you check the ground contacts for these circuits? Did you ground signal at the inputs to these circuits? Your description is entirely unclear.

Did you examine the indicated preamp ground point at the board standoff?

I'm not trying to be snarky here. It's just that we can't help if we don't know what's going on.

This shouldn't be that hard. It's a clean signal bleed so it's either coming from the "rhy" channel or the actual amplifier input. If the ground for the rhy gain control is functional you can eliminate channel bleed by simply turning down the gain control. If that ground is compromised it could be channel bleed caused by a bad ground. You also cannot eliminate decoupling if a ground is compromised. You say you have confirmed relay function. So what is left as far as I can tell would be grounding and decoupling.

EDIT: Decoupling may also be an issue depending on the boards ground layout. Which we cannot see in the provided documents. But if grounding is essentially "daisy chained" in the board design this might cause problems. I would still expect no bleed with the rhy channel gain turned to zero though if the ground is secure.

[GainFreak]

When you say "short" or "shorting" in context above I assume you mean grounding? Because shorting R69 would simply be a jumper across the resistor You don't mention ground in either report of shorting R69.

By "shorting" I meant a jumper across the resistor not grounding. Note that even when it's shorted and the nasty leak with ALL Gain and Volume pots at zero is gone there's still some small leak when gain channel Volume is up (Gain down) but this time when the Gain is up it's lost into the distorted sound.

By "6/ grounding gain and volume pots" I meant I ran a separate wire from ground to all pots' terminals just to make sure the relays are working as per schematic (grounding the respective terminals).

Did you examine the indicated preamp ground point at the board standoff?

Yes, it's OK.

Decoupling may also be an issue depending on the boards ground layout. Which we cannot see in the provided documents.

I don't have the complete PCB layout, only what was in the file. The PCB was sold separately for this DIY project. The amp was built by someone and brought to me with this issue.