Marshall JMP making squealing noise

[Chuck H]

So this problem with the oscillation must have started when I reconnected the 16 ohm tap and left the 8 ohm and 4 ohm unattached. Could this be the source of the problem?

It's a stretch. But since 'according to the customer' the amp never did it before, I suppose we have to treat it as if this problem was caused by you when you altered the impedance arrangement. So...

Those unconnected leads are output secondaries and could be causing instability if they are routed in the wrong place. I would cut the ends flush and shrink tube them. Then twist them with the lead you connected and zip tie or shrink tube them all together. Arranging them that way seems to help with stability issues sometimes. It's pretty important with those amps that the purple presence/NFB circuit lead is routed as far from the preamp as possible. If you removed much length from that lead you may want to add it back and be sure that lead pushed well out and away.

[nosaj]

How about undoing what you did and see if the noise is gone? I'd rule out your work first.

nosaj

[goldtop5]

Yes it's still squealing

[g1]

And the purple is re-connected to the yellow?
Then it depends on how well you trust the customers memory. If it did just start when you did the switch rewire, it probably has to do with lead-dress as some have mentioned.
If maybe the problem was already there before you worked on the switch, it could be a microphonic tube or lack of decoupling due to a faulty filter cap.

[goldtop5]

Update: This amp had multiple issues. There was an opened wire coming off the choke transformer going back into the circuit that was making a crackling noise every time you move it. We opened up the transformer and installed new hookup wire to fix the crackling noise. Also, there were other wires that look like they got in contact with a soldering iron exposing the strands. Obviously, this amp has been worked on multiple times before. Some of those wires had to be replaced. There was a cap, in parallel with a resistor, coming off the low input jack, and going to the preamp gain control pot. We snipped the cap out of circuit and the oscillating noise went away. The schematic called for a .0001uf cap. We swapped it out and the amp is working normal again.

[Chuck H]

It's hard to say what that cap may have been doing. If the amp was wired as a 1959 stock then it doesn't make much sense unless there was some sort of jack controlled channel stacking mod. You wrote JMP originally. Are you sure this isn't a cascade channel Marshall? Is there a master volume? If so you may have altered the amp from stock and what the owner/user expects from it. And, forgive me for saying so but, what 1959 schematic are you using where there is a 100pf cap anywhere in the preamp?

[dai h.]

sounds more like a 2203/4 (which has 470pF over 470k at the LOW SENS input and 1n (or 0.001uF/1000pF--might be misreading the schematic with the ".0001uF" value--which would be 100pF) on the preamp pot).

There are some measures seen in stock amps such as a "hot shield"(basically P to G capacitance) for the grid wire, gluing of the input grid wire to output (with 68k mounted on grid pin if memory serves), gluing P and G wires to chassis, and cap over plate R(IIRC).

[Chuck H]

I was thinking along the same lines. It doesn't take much tinkering on a stock 2203/4 board to make things unstable. And it would explain the switch jack channel stack.

[goldtop5]

My mistake, I was thinking of another amp. It was a 47pf cap. Before, there was a 55pf. I wasn't sure what model the amp was. The customer told me jmp. I called U.S. music corp and according to the serial #, they told me it was a 1959...but which one? By the looks of it, it may have been made somewhere in the 70's. So I went to schematic heaven and downloaded the 2203 100w master vol schematic. It closely resembled what we had here at the shop.

[drewl]

You can also use a shielded input cable with the shield grounded at the input jack and not attached to the V1 plate.

Safer and also cuts down on noise and oscillation.

[Chuck H]

My mistake, I was thinking of another amp. It was a 47pf cap. Before, there was a 55pf. I wasn't sure what model the amp was. The customer told me jmp. I called U.S. music corp and according to the serial #, they told me it was a 1959...but which one? By the looks of it, it may have been made somewhere in the 70's. So I went to schematic heaven and downloaded the 2203 100w master vol schematic. It closely resembled what we had here at the shop.

.00047uf is 470pf not 47pf. Likewise .00055uf is 550pf. Not enough difference there to think it would be the cause of oscillation. Especially since such oscillations typically occur above 5kHz and the 550pf cap would pass lower frequencies in that circuit. Meaning that the frequencies typically noted in oscillations would be about equal for either cap value. That said... If you did change out a 550pf cap for a 47pf cap on such an amp that might stop oscillations because all but the tippy top frequencies would be eliminated. Taking the tone of the amp a very long way from the stock sound. And, FWIW, I've seen 550pf caps in Marshalls from the factory where 470pf is the spec on the schematic, and vice versa. I don't think they troubled much over the value difference either then.

I just did some re read and I see where I got off track when replying to you putting a 1000pf cap across a plate load. In fact, if all you want to change is the oscillation I'll back Bill and say the first things to deal with are the grounds and the filter caps. The amp may be over due for such service anyway.

I would put the amp back to stock as per the schematic. Complete with 470pf caps instead of 47pf caps. Then I would troubleshoot any potential problems with the grounds or power supply filters. Once I was certain the amp was at 100%, if it still oscillated (and they sometimes will) I would add some shielded leads to the input and jack switching circuits. If it still oscillates (it won't) then you can employ some of the ideas I posted in #10 (except for wrapping the plate and grid leads together. Too hard with shielded leads.)

JM2C