Marshall 2205/4212 problem with master volume

[g1]

Whatever the culprit is, it's getting around the FX loop, which should not be possible.
The strangest part to me is that the problem does not occur with the clean channel, where the master works properly.
If there are no messed up mods, I'd think there may be a power supply filtering problem or grounding issue.
With only V5 removed, does it still bleed around the loop?

[g1]

Attaching complete schematic with power amp page.

[terrykvan]

Attaching complete schematic with power amp page.

thanks for putting that up there but I do have the complete schematic. no bleed with V5 pulled. I replaced the multi caps about 7 or 8 years ago, Rubys, and they've been fine. The amp idles very quietly fwiw.

[Enzo]

yeah well, my first suspect with all these multiple symptoms revolving around the channel switching, was the IC. Look up its data sheet, it is just a collection of transistors all in one container.

[terrykvan]

yeah well, my first suspect with all these multiple symptoms revolving around the channel switching, was the IC. Look up its data sheet, it is just a collection of transistors all in one container.

Actually, I pulled that thing last night just to see what happens. Still bleeding. I'm gonna replace it anyway, with a socket. Now the other thing I've noticed is the more the gain, channel volume, and eq are turned up, the more the mas. volume behaves correctly so it's working like it should with the signal sent to it from the tone stack circuit. And the bleed over signal is post channel vol. (gets louder when ch. vol. is turned up).

[teemuk]

After checking the plate voltages on the preamp tubes, a couple of weird things I found were V1b had only 80vdc on it but after lifting the clipping diode on the cathode it went up to 212vdc. I don't know if that's normal or not.

It is normal, so at least from here you wil not find any issues in operation. The diode in the cathode circuit is not a clipping diode. It's a pseudo constant current source. It is permanently forward-biased so the cathode voltage is equal to voltage drop across the diode, which is about 500 - 600 mV (DC). If you remove the forward-biased diode there is no longer "clamp" in cathode voltage, and with 10K resistance at cathode the cathode voltage will rise to about 3 volts (DC). The cathode bias will naturally affect where plate gets biased in terms of DC: The higher the cathode voltage, the higher the plate voltage.

Also, V2a had 276vdc.Too high right?

How come? With 10K at cathode (and no clamp) it supposed to bias very close to such figure. I suppose they didn't even aim for perfect "center bias" at the plate of V2A in order to 1) attain lower headroom so clean channel can be more easily overdriven to "crunch" -tone territory and 2) attain asymmetric clipping and resulting harmonic distortion (even and odd order) by establishing different headrooms for signal swings at positive and negative half waves. With "center bias" the stage would just have higher headroom, clipping would be fairly symmetric and harmonic distortion would be mainly odd order. Although it's a clean channel it would have probably sounded too "sterile".

[J M Fahey]

Yes, no change. I made another observation. I disconnected the ground wire at the power socket inside the chassis and measured between earth and the chassis- 207vac. !!

Are you crazy?
Why did you do that?
You must NOT repeat NOT mess with **safety** ground.
Reconnect it back and the way it´s expected: *firmly* connected.

I then checked the wires from the OT secondary to see if any were damaged, moved them around a bit while observing the voltage. It did fluctuate down to about 155vac but that's it. So I will check all the wires some more. Not sure what else could cause that.

You caused that, who else?

EDIT: you are always grounded, unless you float on the air.
IF chassis is 200V away from that, whatever the reason, and you touch the chassis (you will), then it´s >200V at your fingertips and 0V (ground) at your feet.
Guess *what* joins feet and fingertips .

[terrykvan]

yes I am crazy. I felt tiniest hint of voltage when touching the chassis so I shut it down and checked my ground connection and I had a faulty ground on my power strip. After fixing that I was curious what that voltage was so I momentarily lifted the ground, connected the meter, stood back and powered up. I had a feeling someone was going to call me out on that. I also measured current. It was only a small fraction of a milliamp.Then the ground wire was replaced.

[terrykvan]

It is normal, so at least from here you wil not find any issues in operation. The diode in the cathode circuit is not a clipping diode. It's a pseudo constant current source. It is permanently forward-biased so the cathode voltage is equal to voltage drop across the diode, which is about 500 - 600 mV (DC). If you remove the forward-biased diode there is no longer "clamp" in cathode voltage, and with 10K resistance at cathode the cathode voltage will rise to about 3 volts (DC). The cathode bias will naturally affect where plate gets biased in terms of DC: The higher the cathode voltage, the higher the plate voltage.



How come? With 10K at cathode (and no clamp) it supposed to bias very close to such figure. I suppose they didn't even aim for perfect "center bias" at the plate of V2A in order to 1) attain lower headroom so clean channel can be more easily overdriven to "crunch" -tone territory and 2) attain asymmetric clipping and resulting harmonic distortion (even and odd order) by establishing different headrooms for signal swings at positive and negative half waves. With "center bias" the stage would just have higher headroom, clipping would be fairly symmetric and harmonic distortion would be mainly odd order. Although it's a clean channel it would have probably sounded too "sterile".

Wow thanks for that tutorial. I could use more of that so feel free to read all my posts on this thread and see what you think. Oh, I put the diode back in.

[terrykvan]

Just made an important observation. My amp is wired with V3b post efx loop and V4a is used in the reverb send. I have seen other posts where people had the same issue. I guess Marshall did this for a reason. That's why the bleed still occurred with V4 pulled. So, I wonder if it is practical to rewire like the schematic. I'm about ready to put this thing back together and call it good. If I turn up the MV and back off on the channel volume the bled signal is insignificant to the main signal so it's really not a problem. BTW I did replace the switching IC but no improvement. The amp otherwise seems to function properly.

[g1]

So the master volume does work on the boost channel? Your original post said it did not and a fair bit of tail chasing ensued.
I'm sure you can understand if this thread kind of went off the rails under that assumption?

If the schematic is all correct except for the labeling of those tubes then it's probably a typo. The transistor switching version (earlier) shows the tubes labeled as you found in your amp. (http://drtube.com/schematics/marshall/4210.gif)
In this case I don't see any need to change over.
As you still had bleed with something breaking the circuit at the FX return jack, the only possible cause of the crosstalk is between the 2 halves of V3. Maybe different tubes would give more or less bleed.

[terrykvan]

I'm really sorry about that. it appeared to me that the mv didn't work because, at lower mv levels, you mostly hear the bleed signal. Turn the mv up and down and there's no change. Turn mv up high and ch. vol.low then not much bleed comes through. But the best way for me to use this amp is to plug my digitech 2112 into the efx return and guitar into efx unit. It sound incredible, very pristine on clean sounds and the 2112 with 2 12ax7s has a nice overdrive, pairs well with this amp. But I want to thank you for all the help that you and others gave me. this amp sounds so beautiful with the 2112 into the efx return, simple and pure.

[g1]

Glad to help. Hopefully I'll remember the typo for next time.
I vaguely recall hearing of the crosstalk issue with these, and if it's getting around the FX loop, then the cause being bleeding between the 2 sides of V3 is about the only thing that makes sense.

[terrykvan]

Yes, definitely crosstalk. After trying several tubes I found a sovtek 12ax7wb that all but eliminated the bleed. Still, I will only go through the Effects return. It sounds so much better than the effect unit through the entire preamp. I have enough (amp type) knobs on the front panel of the 2112 to make this possible. I guess I could pull V1 V2 and V4. Would that pose any problems to V3? Maybe causing the plate volts to rise?

[g1]

Is there any advantage to pulling them? Going into the return disconnects them anyway. If you pull them all your supply voltages will rise, for pre and power tubes, including the heater voltage. I suppose it could sound better or worse, as V3 will probably end up biased a little different.
In any case, check that none of the raised DC supply voltages exceed the ratings of the filter caps C43,44,45.