Help me understand

[ricach]

I've got no output. Tracing signal through preamp and all is well. Hit input to PI and I loose the top half of my signal - what should I see? I would expect something like that after I enter the PI, not before. There is only a 270K resistor between the good signal, and the input to the PI.

Am I correct in assuming I should expect to see a good clean signal entering the grid of the PI tube? And if that is the case, any clue what's causing the clipping?

[Bruce / Mission Amps]

Give us a little more info... like, what kind of amp is this, is the PI a cathodyne driver, long tail pair, paraphase (etc.) and is there a schematic to look at...
it could be something as simple as a bad triode in the PI, etc.

[ricach]

Its a 1960 Ampeg Mercury 12.
The Free Information Society - Ampeg M12 Electronic Circuit Schematic

Maybe you could tell me what type PI it is, I've never seen one like it. I've replaced the tube with no change. I am seeing over 300V on the plates, the schematic calls for something in the 150V range. Likewise on V1.

[Enzo]

WHERE are you missing the top half of the waveform? Right at the PI grid? At the speaker?

There is no signal on the speaker?

No signal on the plates of the PI?

I don;t see voltages on that schematic, but when you see 300v on a trioe plate where it ought to be half that, it is usually a safe bet the tube is not conducting. That can happen when the tube itself quits, or if the tube heaters are not working. Or if there is an open circuit through the tube. For example, the cathodes of the PI tube find ground through that 1k resistor R18 (looks like 18 anyway) plus the extra 240 ohm resistor to the upper triode cathode. If that 1k R18 is open or has lost its connection to ground, then the tube has no path for current and no voltage will drop across its plate resistors - high plate voltage results.

Turn the amp off and pull the PI tube. Now measure resistance to ground from pins 3 and 6 of the socket - at least it looks like 3 and 6. You should see about 1240 ohms and about 1000 ohms respectively. Not worried so much if it is off somewhat, I am worried one or both of them have an open path.

[ricach]

Enzo,

Nicer schematic here: http://www.unofficialampeg.com/schematics/M-12.gif

Yes, seeing proper resistance at pin 3 and 6. With o-scope on grid of PI, no top half of signal. Same signal present at the plate. But before R8 (270K) there is a fine looking signal. And at an amplitude I'd expect to see so the preamp tube is working fine even with the 350V on the preamp plate. Bypassing R8 has no effect on the clipped signal.

FWIW, the signal is pretty ugly at the grid, its not just the top half being chopped, but the bottom half has some square waves present.

I will pull the power tubes to see if that has any effect, but the tremolo 'sound' is present at the speaker so I suspect the back end of the circuitry is operational all the way to the speaker.

[MWJB]

What's the dc voltage at cathodes of the PI? In fact, loaded dc voltages throughout the amp would be useful.

150v on preamp & PI plates? I can't see where it says that, sounds low?

Same problem on both channels?

AC voltage at each power tube grid?

PI is variant of the paraphase PI.

[ricach]

Well things have really changed since last night. While rechecking plate voltages at the Pre and PI I realized something happened. I lost R6 (270K) at the upper plate of Pre - I pulled it and it shows completely open so no voltage getting to the Pre-plate. And both plates of the PI are only getting 30V now. Voltage getting to R6(270k), R7(270k), R20(120k) and R21(120k) are still a healthy 330+ from the supply rail. This is wierd because last night that same voltage was appearing at the plates - as if the resistors were shorted.

I guess I'll need to pull and check/replace the resistors before I can go any further. I feel this is just a symptom of the problem, is there anything thing else I should pull and check as well?

BTW: the schematic is pasted inside the amp and this is where I'm getting the values for expected plate and cathode voltages.

[MWJB]

It's not the same schematic that we're looking at? To save time/confusion can you please refer to the job that each resistor does rather than R No. - e.g. "PI input plate", or "PI out of phase (OOP)/triode B plate"

300v eaten up by a 120K resistor (PI plate?) equates to 2.5mA per triode...possible but sems high. Check/replace all plate resistors at preamp & PI. Shorted PI cathode bypass cap?

Check dc voltages at preamp plates & cathodes, power tubes pins 3, 4 & 8. Check resistance from grounded ends of components on the board, read to chassis ground (eliminate tenuous ground connections - if a cathode was lifted the tube would turn off & there would be no current draw, hence no voltage drop accross the plate resistor).

Tried known good 6SL7s?

[MWJB]

Check tension of the tube sockets too, intermittent connections may give differing symptoms depending on which pin is open. Unplug from the wall, drain the caps & close up the contacts with a stiff safety pin/watchmakers screwdriver, clean contacts by coating tube pins & socket with isopropyl alcohol & repeatedly re-inserting the tube. When happy, fire up the amp, all controls fully up (except trem) & wiggle tubes in the sockets, if any make any untoward noises try again/replace the socket.

[ricach]

The tremolo circuit works, and i've swapped out that 6SL7 and all tubes appear to work in the trem circuit. BTW: it's showing 330v at the plates as well.

I only have 100K and 220K 1W (flameproof) resistors on hand to replace the 120K and 270K's - should that be ok for now? One other note - there is a black oily substance throughout the chassis, on board, tube sockets and pots.

The schematic appears to be the same, and I do reference it to the schematics I posted to make sure there is no difference. Once I get the resistors replaced I'll check DC voltages again.

[MWJB]

Your amp has 120K PI plate resistors? Both schems referenced show 47K. The tidier, easier to read Piazza schem has nor R no's.

100K & 220K will 'work', plate voltages will rise a little...but you'll know if the amp is fixed or not.

"330v on the plates" only means something if we know what the plate resistors are being fed from the B+ rail...that's why I keep asking for dc voltages, if your amp is broken the dc voltages will point to why...AC voltages & scope traces will just keep confirming that it still doesn't work.

[ricach]

My sincere apologies - I should know better than to keep referencing 2 different schematics while following a third (one pasted in the amp). Obviously I'm not as thorough as I think.

Good news, replacing the resistors did the trick. I'm reading about 210V at the PreAmp plates, and 240V at the PI input plate, and 215V at the PI 'B' plate. I'm guessing the Tremolo is next to go so I'll be replacing those plate resistors as well.

[MWJB]

Glad you got a result.

No need to apologise, just been bitten by that particular dog before, along with, "my amp doesn't work, here's the schematic"...well if everything is wired up like the schem your amp works fine :-)

That's why pics and/or dc voltages from the actual amp in question are the first things I want to see.

[ricach]

Well, I initially did look at the DC voltages, and although excessively high, I figured I should at least hear something so my next thought was to trace where I'm losing the signal - thus I came upon the unusual sight at the PI input grid. Fortunately (for me) the resistors failed and made it easier to trace out the problem components.

For any vintage Ampeg experts out there, this Mercury M-12 is an unusal one. Although dated as manufactured in 1960, I cannot find any information on this exact one. It has the navy random flair vinyl covering, but the grill cloth is not what I've seen on other Mercury's. The Ampeg logo is stamped on a piece of tin and mounted directly to the grill. It has 2 instrument inputs and 2 accordion inputs, with one of those being stereo (with a tip/ring jack). Runs 6V6's. Utah speaker. The schematic inside the amp does not perfectly match any of the M-12 schematics on the internet, nor the one that Ampeg sent to me.

[MWJB]

Ha! I think I'd actually be more surprised if I saw a early 60's Ampeg that DID match a schem! :-)

But seriously, many of these amps (inc pre BF Fenders) were subject to constant development and schems were updated much less frequently...if at all. Subsequently, there are many amps that don't match their schems and some schems & layouts with simply NO matching amps (like 5F1). Sometimes, inspecting type/brand of component, looking for disturbed soldering is the only way to establish whether an amp is "as originally built" or not.