Univox U-50 Power Transformer issues

[plabass]

Hello!

First time posting here. I have an amp on the bench right now that's got me a little stuck. It's a Lafayette-branded Univox U-50 Lead guitar head. It came to me with just a ton of problems. I've gone through and re-capped, fixed a couple wiring errors, tidied up the grounding scheme, replaced a couple obviously bad resistors and generally got the thing looking pretty good. I believe there was a tech in there before me that didn't quite understand it. There were a lot of pretty obvious errors.

My issue is this -- the plate voltage on the power tubes is exceedingly high at about 630vdc measured at pin 3 on the socket or directly off of the PT secondaries. It's a 6l6GC amp (I believe. Maybe the issue is that it's not?) so this is way out of spec and obviously an issue. The PT LOOKS original in that it matches the OT in appearance but I don't know for sure.

Does anyone have an idea of what I could look for that would cause the plates to be this far out of whack? My guesses are an incorrect replacement PT (though, again, it looks original), that the amp is actually supposed to run 6550's or similar, or ?.

I don't have the best documentation on this piece. The closest I've got is a schematic for the PA version of this head (mine is the Lead version). The powersupply, I'd guess, would be pretty similar though:

[J M Fahey]

Well, your schematic shows 450V AC on the transformer secondaries, which translate into a very high B+ (nominal 636V) as you are measuring, so .....
The series capacitors hint to that too.
I would use it with 6550's.
JM2CW.

[52 Bill]

I remember working on one years ago, when the only new tubes available were the Chinese 6L6's that wouldn't last more than a few hours in that amp. Had to reduce the B+ to get it to work. The Sovtek tubes will probably work in there.

6550's may have too much heater current draw.

[plabass]

Interesting. I'll have to check out the datasheet for the Sovteks. Do you think they can really take 600vdc+ ? I wasn't aware any 6L6 could handle that. I've got EH's in there now and they definitely don't like it.

[52 Bill]

How you bias them will make a big difference in how happy they will be. How are the EH's biased now?

[olddawg]

You could heat sink and series some 10v zener diodes in the B+ supply. I remember having fits with some Japanese equipment years ago. Line voltage in Japan is 100v at 50hz. A lot of stuff brought back by the military would almost work here. Strange problems and failures because of power supply issues. If you didn't read the spec on the back panel it could drive you nuts.

[Prairie Dawg]

I had a Univox years ago like that that was cookin' at over 600v on the plates. I'd probably drop that voltage down 100v with a zener. Think survival.

[plabass]

Thanks, everyone, for the input!

I'm glad to hear that this is not an unusual problem for this particular amp.

On the bias -- I don't know exactly what the plate dissipation is. I got as far as measuring the plate voltage before I powered the thing down and started scratching my head. I'm reluctant to run it for very long with this kind of plate voltage. I'll check again but I think everything looks good as far as the bias supply is concerned. It's fixed bias and the bias resistors measure close to 220k (as you'd expect). There are two 1.5k grid stoppers in series with the 220k's that aren't in the schematic but I don't think that's going to affect my situation much.

So far I think the best suggestion is addition of rectifier diodes. I guess that will get me where I need to be but I just don't understand the "why". As in "why was this amp designed this way?". I've been trying to find a 6L6 that can operate at anything close to 600v and every one has a max plate voltage of about 500vdc. What's up, Univox? What was the intention here? And more relevant; am I just missing something really basic?

[Jazz P Bass]

Did you consider EL34 tubes?

[plabass]

I have not. I guess that would be another option. It looks like I could:

convert to El34's (assuming the PT can handle the extra current draw. Don't know)
convert to 6550's (same)
Rectify my B+ down to something manageable

[Prairie Dawg]

Mine had the original GE tubes in it. Sadly, I watched one arc internally. Nice. I don't know what they were thinking either. EL34s will handle a lotta voltage, likewise 6550s but you might have to upgrade the filament supply-that or dump some of the voltage with a zener.

[J M Fahey]

Maybe a relatively small and inexpensive add-on filament transformer lets you upgrade tubes, and lighten the main power transformer load to boot.
As of the designer's intentions, maybe he wanted to make a killer amp ... and he succeeded .... a tube killer amp, he he.

[jrfrond]

I have the original schematics here for all three versions (guitar, bass, PA) and all sport the same PT and secondary voltage. That is WAY too high for even NOS 6L6's operating in fixed-bias pentode mode. You're going to need 6550's, KT88's or NOS 7027A's to handle that horsepower.

Or, you can reduce the B+ voltage, change the PT.....

I wonder what those engineers were thinking?

[Gregg Orange]

Just got an old Lafayette Univox u50 head from a friend to work on. I am no amp tech or electronics expert but, I can usually figure out what's going on and have fixed many amps thru troubleshooting over the years. This is a weird amp with a solid state pre amp section, a 12 AX7 tube as a phase inverter and two 6L6 tubes in the power section. The amp fires up and doesn't blow fuses but, sounds as if it were operating at about 1/4th power and sounds very sick. Bad sounding distortion, farts out and obviously has issues. Replaced all tubes to no avail, same problem. I see one paper covered 1000 mfd. cap that looks like it's leaking so I am going to start there. I read previous posts about extremely high plate voltage coming off of the secondaries of the stock power transformer. 630 volts is WAY too high for 6L6 tubes as far as I know. Most of the amps that I have worked on that come with 6L6 tubes have around 500 volts going to the plates off of the PT I have a hard time believing that this is a stock transformer but I could be wrong. I will be delving into this amp and checking some voltages on it this week. Any advice on what I should test with my multimeter to check for bad components? There are no obvious burned components but I am pretty sure it needs to be recapped as this is a more than 30 year old amplifier. Thanks....Gregg-Ft. Worth

[J M Fahey]

You´ll need an oscilloscope.
Feed the amp a clean 100mV 1KHz signal at the input and follow it along its path , until it reaches the speaker.
Somewhere along the way it will distort.
Only then, you start checking around that stage.
Measuring part by part with a multimeter to check whether they have drifted their value , have become open or shorted is an incredibly tedious and never ending way to troubleshoot and even so, leaves out, say, 50% of failure modes so you´ll never be sure.
Google its schematic and post it here, so we all talk about the same.
Good luck.

PS: in that particular case, since you should ditch that Power Transformer anyway, (I consider dropping 200V with Zeners way over what´s practical), you might just strip it, add a couple Noval socket holes, and build a nice classic amp in there, your choice.
A Bassman, for example, is a very versatile amp which will give you from spanky clean to mean crunch in the same package.
Or any other amp you prefer.
But working on an obscure undocumented amp, which we *know* was poorly designed to begin with, may prove frustrating.
jm2c.