Marshall JTM30 output and bias question

[Randall]

Amp came in dead, with a smoked AC inlet board. The neutral wire side caught fire, and the fuse did not blow. I replaced the inlet socket and fuse holder. Now it powers up, but has a mild hum. I checked the tubes and they are drawing 96mA and 99mA! I get 406v at the plates, 403 at the screens and only -29v at the grids. I don't see a way to adjust the bias on this one, and don't really understand the zener array. I also must point out that it came in with 6L6s but the drawing shows 5881s.

So, does this amp have a problem that caused the inlet board to fry? I had initially thought it looked like a connection gone bad, raising the resistance and causing it to burn. Shouldn't the bias be adjustable? Is the problem that it has the wrong tubes in it?

https://www.thetubestore.com/lib/the...-Schematic.pdf

[Enzo]

5881 is a 6L6, making that swap won;t hurt anything. True (old) 5881 had slightly lower spec than 6L6, they were an industrial version of 6L6. This is a low power amp with B+ that won't threaten them. Modern 5881, current production, are just 6L6s with a different name.

So you have three zeners adding up to 39v setting the bias, but you have 29v. I suspect the 9v zener is shorted or one of the 15v zeners is leaky. Measure voltage across each to see who is underperforming. Or if that bias 29v is FULL of ripple, then maybe C126 is not working. Lastly, the HV AC is tapped through a resistor and cap. Are they healthy?

Not much to understand here, the AC is rectified by D107 to make a small negative supply, then the zener string sets the voltage there.

[Randall]

Hmm. I can only get probes on ZD 101 and ZD 102, and they are both low. This thing has a sideways mounted board that makes things tricky to get to. I think pulling it and replacing those components as a set while it's out is the way to go in this case. Will changing the 9v zener to 15v give me a little more negative bias voltage?

And there is ~5vac of ripple.

And what is the function of C127 X class cap feeding the zener array?

[Chuck H]

I think I will change the 9v zener to a 15v while I'm at it.

Why? To raise the bias voltage?

[Randall]

"Why? To raise the bias voltage?"

Yes.

[Dave H]

And what is the function of C127 X class cap feeding the zener array?

It's because the bias supply is derived from the B+ supply which uses a bridge rectifier. It can't be done in the usual way because there's no centre tap. It has to be AC coupled.

[Randall]

Must it be an X Class cap? I seem to be having difficulty finding one in the proper rating/value.

[nickb]

Must it be an X Class cap? I seem to be having difficulty finding one in the proper rating/value.

Try Mouser. I came up with 17 hits

https://www.mouser.co.uk/Passive-Components/Capacitors/Safety-Capacitors/_/N-fb8xl?P=1z0wpt1Z1y92jkrZ1yzshdiZ1yp0uc3Z1z0x6d8Z1z0x63t& Keyword=47nf+250v&FS=True


PS: I think the reason for the zener string is so the amp will work on 50 & 60Hz. The problem with the capacitive coupling is that the impedance varies with frequency so rendering a passive divider out of the question.

[Dave H]

Must it be an X Class cap? I seem to be having difficulty finding one in the proper rating/value.

The cap may be OK. You could try disconnecting the zener string to see how much -ve voltage it can supply.

[g1]

Your bias is out by 10V. I would see what you get when it's back in order before trying to make it even more negative.
Like DaveH mentioned, the bias circuit may not be able to go much more negative without the zeners anyway.

[Randall]

It looks to be a bugger to access that cap, so I am opting to just replace all of them while the board is out to save myself time and aggrivation. I asked about subbing a 15v zener for the 9v because I have seen other threads on this subject suggesting that.

[Dave H]

Can you easily lift one end of one of the zeners? It should have a lot more bias voltage than you need then, if not higher voltage zeners won't help.

[Randall]

I think I can get to one side of ZD101. I'll give it a try while I am waiting for my Mouser order, and report back in a couple of days.

[g1]

If you are going to replace them anyway, just clip one from topside. They are a series string so a break anywhere in the series string will disconnect them. Then take your bias voltage reading at the power tube socket pin 5.

[Randall]

Update. I pulled the board and replaced the whole bias circuit while I had it out, this thing is a pain to get into. Now I have -37 volts on pins 5. The only component that shows any failure out of circuit is the safety supply cap C127, it is supposed to be .047mF, but it reads .017mF. Also ripple now gone on bias voltage. I kept everything stock and did not make any adjustments to lower the bias as I was asking.

Now I have 66mA and 64mA at 431v, this still seems high at 28 and 27 watts for 6L6s. Do I still have something wrong?

Swapping out 5881s gets me 53mA at 23 watts, again, why so high?