Marshall JCM 800 presence issue

[Randall]

Early 80's JCM 800 50 watt combo, ST 1 iss. 202 board. Model 4104. Presence control is extremely noisy, When it is full up there is 48v on the lugs that are tied together. Although the schematic states Phase Splitter all models, this one is not exactly the same. From right to left, I have the 100K resistor coming from neg feedback F, to the .1uF cap and 10K resistor, but not the second .1 cap nor the 4.7K resistor to ground as far as I can tell. I pulled the .1 cap feeding the second grid, but still have the voltage on the presence pot. The other resistors measure as good in circuit.

There shouldn't be 48 volts on that pot, should there? And if not, where in the heck is it coming from?

[Enzo]

Where is it coming from? The cathode current. As you describe it, they left off the tail resistor (4.7k) and just used the presence control as the tail. SO whatever appears on the PI cathodes, is voltage divided with the 10k tail resistor and the 22k pot. SO yes indeed I expect a bunch of volts there. Exact voltage depends on the setting of the control.

Control will always be noisy then. I would try this. Put the other 0.1 cap in and the missing 4.7k resistor and wire it up like the schematic.

[g1]

Does the pot have the wiper connected to a .1uf that goes to ground, like the 1959T ?

[Enzo]

He mentioned in post #1 that two legs of the pot are soldered together.

[Chuck H]

Early 80's JCM 800 50 watt combo, ST 1 iss. 202 board. Model 4104. Presence control is extremely noisy, When it is full up there is 48v on the lugs that are tied together. Although the schematic states Phase Splitter all models, this one is not exactly the same. From right to left, I have the 100K resistor coming from neg feedback F, to the .1uF cap and 10K resistor, but not the second .1 cap nor the 4.7K resistor to ground as far as I can tell. I pulled the .1 cap feeding the second grid, but still have the voltage on the presence pot. The other resistors measure as good in circuit.

There shouldn't be 48 volts on that pot, should there? And if not, where in the heck is it coming from?

The other .1uf cap should be AC bypass for the non inverting side of the LTP. It's not feeding it. The 100k should be the NFB series resistor.

The schematic does show the 25k pot wired as a variable resistor. Which should explain the joined lugs.

It sounds like all the parts are there. I think you must be misinterpreting their position or function in circuit or (more likely because you have voltage on the pot) the circuit has been wired incorrectly. Like Enzo said, there is a standing DC voltage on the tail resistor because it elevates the PI function from 0V. Though 48V does seem like too much for where you're measuring it and in fact there should be none on the 25k presence pot. So either someone has kludged the wiring, the .1uf presence cap is leaky or both. Also like Enzo said, wire it up as per the schematic.

[Helmholtz]

The other .1uf cap should be AC bypass for the non inverting side of the LTP. It's not feeding it.

Not quite sure which cap you mean, but C15 in the schematic is a coupling cap feeding the NFB signal to the non-inverting PI grid, which is actually the inverting input of the power amp. Without C15 no there is essentially no NFB.

If the presence circuit is wired as in the schematic, but R23 (4.7k) is missing, there would be a missing DC path to ground and R22 (100k) will be used as additional cathode/tail resistor. This high tail resistance could explain the high cathode voltage. Consequently PI plate voltages would be high because of low current.

But that would not explain DCV on the presence pot. Actually the presence wiring is not clear to from the OP's description. A little drawing could help clarify things.

[g1]

Photo's would also help.

[Enzo]

Helms, the way he described it, it sounded to me like he had the result of shorting C17 and removing R23. But they did it like putting the control in place of R23 and removing C17.

Wiring it like the schematic should make it work.

[Helmholtz]

Helms, the way he described it, it sounded to me like he had the result of shorting C17 and removing R23. But they did it like putting the control in place of R23 and removing C17.
.

Thanks Enzo. Somehow I refused to read it this way, as this means that the amp had no presence control but variable gain instead.
But 48V across 22k means a little over 1mA per tube and that makes a lot of sense.

Wiring it like the schematic should make it work.

Completely agree.

What a lemon!

[nevetslab]

That ^^^^^^^^^ If C17 isn't ahead of the pot, then there WOULD be the DCV on the Presence Control. My 2205 schematic (50W Reverb Combo) has the pot ahead of the cap to ground, so there would be DCV on the pot, and no doubt the source of the noise. Put the cap ahead of the pot, as Enzo suggested.

[Helmholtz]

That ^^^^^^^^^ If C17 isn't ahead of the pot, then there WOULD be the DCV on the Presence Control. My 2205 schematic (50W Reverb Combo) has the pot ahead of the cap to ground, so there would be DCV on the pot, and no doubt the source of the noise. Put the cap ahead of the pot, as Enzo suggested.

It's not DCV on a pot but DC current through the pot and/or wiper that makes it noisy. In other words there needs to be a DC voltage drop across the pot.
As a series cap blocks DC current, the order of the series arrangement should not make a difference.

But if I understand Enzo correctly, he thinks that the presence cap as well as the 4.7k NFB shunt resistor are missing and the PI cathode current flows through the pot.

[Randall]

Sorry, I thought I linked the schematic, but it was a dud.

In the above schematic, I do not have C17 or R23. Please forgive the way I have tacked on the presence pot wires, it is for easier connection while I am flipping the board. I have not wired it like the schematic as Enzo suggested yet, but I intend to. In the meantime, here are photos of what I now have.

Purple wire coming off presence pot, going to junction of 100K NFB resistor, 10K R20, and .1uF C15. Purple wire below R22 goes to impedance switch, Green wire below C15 goes to the second grid of V3. Yellow wire below R16 goes to shared cathodes.

[Helmholtz]

So Enzo was right. Doesn't look factory original to me. Probably someone sloppily modded the amp to variable NFB instead of presence.

[Randall]

I tacked in the 4,7K and .1uF, and now it works, except that there is a glitch in the pot, and it is wired backwards so that it gets darker as the pot is turned CW. So, how to explain this having missing parts? There is no apparent place designated on the board for them. I will have to fly those parts off pot lugs. Did this come from the factory like this? I mean, it has obviously been hacked previously, wires replaced, components snipped topside and resoldered. But, where were those two missing parts?

[nickb]

I wonder if someone had tried to make it like the model 1959 which did have DC through the presence control and does sound very noisy.....