univox u-1000 with a suspicious mod?

[mykle]

Hello, I wonder if I can ask y'all some advice on an amp I'm trying to get fixed up for resale. It's a Univox U-1000 -- the schematic is here:

http://www.univox.org/pics/schematics/u1000.gif

... only what's inside the amp doesn't quite match. The B+ voltage is significantly higher than listed at all 5 reference points. The large 10u/2W resistors between the 6L6GC power tubes and ground have been replaced with 1000u/5w resistors, which appear to be of much newer vintage than all the other parts. I'm thinking the high voltage and the new, larger resistors might be related. At any rate, all of the tubes are running on higher plate voltage than listed, with one exception (see below).

The amp basically works, but for some reason the tremolo circuit is sending a putt-putt noise through the amp even when the tremolo footswitch is off. Also, the reverb section has tons of buzzzzzzz when turned up. The voltage on the plate of the reverb half of the 12ax7 is too low, only 90v -- my theory is that somehow the other half of that tube is running on such high voltage that it's causing current to cross between the two halves. Replacing that tube with a NOS reduced, but did not entirely eliminate, the problem.

I guess my first question is: is it possible for current to bleed across the two halves of a 12ax7 like that? And my second question is: did someone increase those resistors as a tone-adjustment mod? Is that a legit thing to do? It seems nuts to me, but i'm new at this.

I should add that I tested all the other resistors and caps, and found nothing severely wrong.

Thanks for any advice.
-mykle-

[Steve Conner]

Changing those resistors to 1000 ohms would bias the power tubes extremely cold and cripple the power output of the amp to about 10 watts. The lack of current drawn by the power tubes would cause all the voltages to rise. I'd replace them for the original 10 ohms ASAP and let the amp rip at its full 100 watts again

[mykle]

That makes sense ... but it sounds quite a bit louder than a 10 watt amp. I wonder if whoever changed those resistors also adjusted the bias current to compensate. I guess I'm trying to judge whether the person who did this knew what they were doing or not.

Truth is, the tone of the amp is pretty good. I just want to get rid of these issues in the reverb & tremolo.

I'm going to measure the bias as it stands now & see if it's drastically low ... if it is, I'll try those resistors. Otherwise ... ?

[mykle]

weird wattage ...

(i guess i should mention that i never really studied electronics in school, so i'm still kind of coming to terms with Ohms Law. i'd appreciate if you let me know what i do wrong.)

so i'm seeing voltage drops between 1.65 volts and 1.95 volts across those 1k resistors. so that's current of something like 1.65-1.95mv, right? that's with the plate voltage at 511 volts on all four tubes. (i'm also reading a bias of -53 volts on the grids, if that makes a difference.)

this seems to be giving me a static dissipation wattage of just under 1 watt ... my god, could that be right? how is the thing even making noise? i thought a 6L6GC wanted 12-24 watts ...

i have a hard time believing this is right. i bought this amp from a legitimate local shop with a good reputation for service. hmmm ...

-m-

[Steve Conner]

Yes, your math is right. Either the amp is making a pathetically small amount of power, or you read the values of the resistors wrong and they're not really 1k at all.

[Enzo]

No, one side of the 12AX7 is not going to run really low because the other side has high voltage. If the plate is running low voltage, check the other end of its plate load resistor - well first make sure that resistor is close to its value - and see what voltage is dropped across it, and thus the current through the tube. If current is more than 1ma or so, find out why.

[mykle]

About the general overvoltage: I changed the 1k cathode resistors on the power tubes back to the 10 ohm spec. i was surprised that this didn't change the plate voltages at all, or the current over the resistors. But then i discovered a bias resistor was set to 660 instead of the 6600 that was spec'd ... hmm ... i changed that back to spec, which brought down the bias voltage to -45 from -53, and that seemed to push things in the right direction -- dissipation went up to .2 watts. i'm wondering if the bias ought to be significantly lower. I'm re-reading my sources on tube biasing ...

About the undervoltage on one tube: i found something fishy in that schematic, actually. The two preamp stages are spec'd for plate at 165v, cathode resistor at 1k. But the reverb stage is spec'd for plate at 165, cathode at 220 ohms -- and the number 220 looks like it was hand-written on the schematic! They all get their B+ current from the same source and they're all 12ax7 ... they can't have drastically different cathode resistors and still have the same plate voltage, can they?

However, I do indeed see a 220ohm resistor right there, and it sure looks like it came from the factory. It's just weird. There are some other differences between the actual circuit and its schematic, in the tremolo circuit. So I'm beginning to doubt the schematic ... maybe this design evolved, maybe these voltages came from the factory ... sigh.