Am_peg Gemini II... Super loud hum when powered on.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

T-Dogg

Well-known member
Joined
Jun 11, 2004
Messages
81
Location
NJ, US of A
So I'm trying to diagnose and fix this amp, which was given to me in non-working condition a while back. I'm also half trying to understand the basics of tube amp circuitry, and have a couple o' general questions on the schematic, available here:
http://users.aol.com/portaflex/schems/g-15.gif

First off, when the amp is powered up and switched out of standby, there is a very loud, constant hum. Pretty low frequency. Adjusting the volume or tone controls does nothing to change the output or sound of the hum. The input jack does not appear to be shorted. Based on what I've read, this points to a couple of possibilities --

Bad smoothing caps
Bad power tubes
Bad tranny... power, I guess? possibly output?

Does this seem a reasonable assessment? Because the power tubes are tough to find nowadays (7591's I think) I figured I'd test the caps first... (BTW, one of the few things I do know is to discharge them first:) )Then I guess the tranny's, although I'd have to read up on how to do this.

I'm gonna start on it tonight, if anyone has any input I'd love to hear it. Unfortunately, I don't own a scope (nor would I know how to use one at this point!), so my testing is limited to a digital multimeter that can also check capacitance. BTW, should I take the caps out of circuit to test them?

Also, if anyone got to check out that schematic, I have a random question about it. I recognized the two smoothing caps as those following the rectifier stage, and think I understand what they do at a really high level(take much of the ripple outta the DC right?)... But what is the purpose of the cap coming off of the primary going to ground? Is this some sort of smoothing cap too?
 
> 7591's I think

Schematic you linked shows 6L6GC. I think you can still get them.

7591 (or a near-enuff clone) is once again in production, if in fact that is what the amp is made for. 6L6 and 7591 are similar but NOT the same.

The cap on the power transformer connects chassis to either side of the wall outlet. One way will give less hum (once you get rid of the BIG hum!!!). The other way will electrocute you.

Your symptoms really suggest that the main 30uFd 600V cap has failed. You can't get a 30uFd 600V cap. Use two 100uFd 400V in series. Go ahead and get a 10-pack of them: I'd suspect most of the 500V caps have also seen better days and will beg for replacement. Replace the 100uFd 100V in the output stage grid-bias supply too.
 
Thanks PRR. I tested all the caps last night, and it seemed quite a few were bad, or drifting from the expected value -- at least based on what the meter said, though it might be innaccurate cause I left everything in circuit. But as you predicted, those big paper 600v and 500v ones seemed to be among the worst offenders. I'll definately replace them, and see how far that gets me.

Also, thanks for the substitution cap values. That would've probably been my next question in a few days when I realized I couldn't get one! I searched a bit online and found some "cap series/parallel" calculators... But what if I wanted to go backwards, without knowing the cap values... ie -- find out how to create the equivalent of a 30uFd 600v cap by using others in series or parallel like you did, what is the equation to do so? Couldn't seem to find info on this. I'm hoping it's semi-easy to calculate ... but I'm probably being overly optimistic.
 
How long did you leave the amp on with the hum? Sometimes an older amps caps will take about 1/2 hour to reform and the hum will decrease. I would do this outside and watch the amp the whole time don't leave it unattended.
If the tubes are glowing red at the plates shut it down

IIRC that amp is liked by jazz players.

IMO the only way to realy test caps is with a meter like a sencore LC102, with the cap disconnected from the circuit.
 
Hey gus. To be honest, I only left it on for a couple seconds tops... The hum was so bad (and kinda loud!), I was afraid something would blow up at any minute! I'll give it a shot though. When ya think about it, it's pretty damn funny that my amp has a better chance of fixing itself than me getting it working!

I read its supposed to be really clean... In the back of my head, I'm thinkin I'd love to be able to replace the 15" jensen with something a bit beefier, seal the cabinet, and use it as a mid size bass combo. Not to go to waste, I tried that Jensen in a Peavey delta blues amp I had layin around, and it sounded pretty good, despite its wear. I think better than the speaker it came with...

Just looked at that Sencore... $$$
 
I bought an old silvertone Se amp kind of like a champ. It hummed for about a 1/2 hour then it stopped I let it run for a few hours. let it sit a day and then tested the caps they were OK. I also did this with a older fender princeton and it seemed to work.

With other amps the amp hummed still after 1/2 hour and they measured bad as well.


Funny thing I have measured older caps in amps and pulled from amps with the 102 and some might have been low in value but the ESR and leakage(at working voltage.
The 102 secore can go to 999V in 1 volt steps) were good.

Some of the older caps seem to last. I think "blind" recapping of amps like what you read on the web needs to be examined. reminds me of the 3,000 mile oil change myth.
 
Pretty cool, for as many times I've read about recapping, I never once heard that they can sometimes be revitalized.

Would it be really, really bad for the amp if I leave the speaker disconnected while doing this? It's just so loud -- I have a little deck off my apartment, but I could see someone complaining! Maybe a Scholtz power soak before the speaker?
 
It might be better to just change the caps if the hum is that loud, maybe a power soak would be OK. Was the amp sitting a long time years? before you powered it up. I bought a b25 ampeg that hummed IIRC the triode pentode output driver tube ampeg used 7199? was bad I changed that and the hum decreased a lot.
 
> those big paper 600v

Those aren't "paper"; though they may be cardboard-wrap.

Paper capacitor is two layers of foil and a layer of paper insulator. Just one uFd of paper-cap is big; 100+uFd of paper-caps would be bigger than that amp.

These are electrolytic. Aluminum foil in a metal can full of liquid. The insulator is a super thin layer of aluminum oxide, "grown" on the foil by running a current through it. With age, the insulating layer un-grows, dissolves, the cap is sick. While you might be able to re-grow the insulation by leaving it powered-up, it does not work well for me and I would not turn my back on a re-formed electrolytic.

Cut open the old caps. I bet they are two capacitors inside that carboard sleeve. There aren't any 600V electrolytics: max is 450, maybe 500V. Paper and film caps are much more expensive per uFd. So the usual "600V" electrolytic cap was two caps in series, packed together, with overall insulation (the second cap's case is at +300V).

> what if I wanted to go backwards, without knowing the cap values...

Two 100uFd, in series, is 50uFd, etc.

Yes, that's bigger than the original 30 or 40 uFd. But bigger is better, and the $/uFd of modern caps is much better than the old days. Ampeg probably would have used 50 or more uFd, but the cost killed their profit. Now that it has matured into a Classic, and you have invested hours of work, you may as well use better (bigger) caps than their accountant let them use.

You would think that two 400V caps in series could be called 800V. That would only be true if they were perfectly matched in value and in leakage. In real life, you have to de-rate. They are probably running 500V-550V on that "600V" cap (which is probably two selected 350V caps); I suspect that two 400V caps from the same lot, in series, will stand 500V-550V for a decade.

It might be good to put 220K 1 Watt 5% resistors across all the 100uFd 400V caps, to equalize the voltage and swamp the leakage.

> replace the 15" jensen

JBL E-130 for loud, E-140 for deep. Can't kill them with 2x6L6 (I didn't kill mine with 4x8417), and you can't get more air per watt than these guys give. Sadly they are out of production. And the E-series is HEAVY.... D-series would be lighter but pre-1980, maybe hard to find a healthy one. (I knew a guitarist who bought a C-130 new and used it all his life.)
 
Thanks alot for all the help guys... My shipment of Orange Drops and Sprague Atoms just came in, courtesy of the kind folks at Angela Electronics. This will be my Saturday project, hopefully I'll have a functioning amp by the end of the day!

BTW, these "orange drops" have such a vibey look to them! I'm thinkin, if I was to be severely shocked by a cap, I'd like it to be an orange drop. I may even fashion a pair into some earrings for the ol' lady. I'm sure she'll just love 'em...

One more question -- these caps are alot bigger than what was in there. Especially the two 100uF 450v caps that I'll need to put in series to get the 30uF 600v... Not to mention originally (3) 40uF caps were in a can hanging off the bottom of the chassis, and looks like I'll have to stuff the new ones inside. It's gonna get tight in there, alot of the bigger one's will likely be right up against each other. Is it likely heat dissipation be a problem?
 
Orange drops are cool in guitar amps, not so cool in studio equipment, so you made a good choice. You can still get a metal multi-section cap, available at www.tubesandmore.com. There is a picture of it around here in one of the threads.

Don't disconnect the speaker! !! You can pop that output transformer. It needs a load. If anything, short it out. This will not hurt it during testing. Thats why they have shorting speaker jacks.

I am worried that your loud hum may be more than just the caps, like some circuit problem somwhere, but either way, the new filter caps are a good idea.

Ampex used both the 6L6 and the 7591 in a lot of their amps. And even some 7027A's. Westinghouse came out with the 7591. If you really wind that tube up to it's limit, you can get almost 45 watts out of it in push-pull AB1. Close to a pair of 6L6's, and half the space. Get rid of the 100 ohm pot in the bias circuit if it has one, and use a 140 ohm 10 watt in there. (Cathodes to ground) Westinghouse did not really publish a tube manual, but RCA started making them so there is some data there. Not a lot of companiies made the 7591, as the profit margin was slim. For instance, back in 1969, a 6L6 went for $8.45, and the 7591A for $2.30 !

cj
 
Super-loud hum, unaffected by volume and tone controls, usually means that your output tubes are about to fry. Are they glowing red or orange? Power up the amp and watch the output tubes carefully; do you see a glowing spot appear on the plates of one or both tubes as the hum comes up?

Check your bias supply. Also check the coupling caps to the output tube grids. When these become leaky, the leakage can overcome the grid bias and cause the output tubes to burn up. Large unbalanced currents flowing through the output transformer cause a loud hum.

You might need a new set of output tubes by the time you've fixed the original problem.
 
In the back of my mind I've been wondering if it could be more than the caps. But all the large value caps, including the one in the grid bias supply, and a few others in the circuit were showing 0 capacitance on the meter (well, maybe a couple of nF, but nowhere near their respective values.) I'm pretty sure this amp sat around for 30 years or so with no juice, which I guess explains their lack of... ummmm... capacitancificationability :razz:

Then after my original post, I came to realize I had at some point blown a fuse (I think it may have happened when I switched out of standby, but it is possible it occured on power up after a few seconds...) Haven't picked up any additional fuses, so I never really got to try "reforming" the caps as Gus suggested or checking the plates... Hopefully, when the new caps are in, I'll at least be beyond the fuse issue (if it isn't working like a charm that is!) Is there an easy way to test a tube? I'd assume not without a scope and lots of *voltage running through it, right?

EDIT: *Just remembered, I should say current, right?!?!
 
Welp... I replaced the caps, had to get a little creative stuffing them in there... but it didn't get me very far! The amp powered up in standby, no problem... So I let it sit for about 20 minutes or so in standby, watching the tubes. Didn't see anything that looked suspicious to me, nothing appeared overly bright. But when I attempted to switch out of standby, I got about 1 second of loud hum and then a blown fuse. Repeated this twice with the same results.

I guess I'm pretty much stuck at this point, not sure what else to try. I know it was mentioned it could be an issue with the tubes... I'm kinda reluctant to go out and buy some 7591's, seems like they'll be pretty pricey for a pair, unless thats all that's left to try... Gotta admit I'm also a little nervous about testing voltages with the thing powered up.

I also noticed there was a schematic on the inside of the amp chassis -- and surprise, it's pretty different than the one I found online and posted earlier. Maybe just the interpretation of the circuit, but the rectifier section appeared different as did some of the cap values. To my eyes, the one in the amp was easier to follow... Looks pretty much like this one, sorry the image is a bit blurry: http://www.kbapps.com/audio/schematics/tubeamps/ampeg/g15.html

Even if I don't get this thing working, following the circuit from the schematic was a big step for me -- starting to get it now!

And finally, gotta ask my random question: there's a third tap on the output tranny with a 430v potential across it -- why is it necessary/what does it do? I think I understand the two main taps tied to the output tubes (pushing and pulling the speaker in and out, hence push pull?)
 
They WOULDN'T glow red or otherwise do anything out of the ordinary until you switched off standby. I guess I should have been more specific.

Remove the output tubes, then power up the amp. Does it still pop the fuse?

If not: Set your multimeter to DC volts. Connect the negative lead to the chassis and insert the positive lead into pin 6 on either output tube socket. Do you read a negative voltage there? How many volts? Is it present on both sockets? You can take this measurement without even going inside the chassis.

Your ability to troubleshoot this thing is going to be severely hampered if you don't wanna take voltage readings. Yes, probing inside a powered-up chassis can be dangerous; but if you're not careless and you practice proper safety precautions it's no more risky than driving your car.

You do NOT want to stick expensive new 7591s in there until you've rooted out and fixed the problem.
 
Thanks NYD. I just removed powered tubes, but same results... A quick burst of hum, then the fuse blew. Can I also remove the 7199 and 6CG7 tubes from the circuit and attempt powering up? If these were bad, could it cause blown fuses, etc?

If not, then there's probably some random issue with the circuit? What sucks is, someone was in there before I got my hands on it, so I can't be sure other things in the circuit weren't rewired for better or worse. I don't see any obvious wiring issues/breakages -- guess I can try to follow the whole circuit against the schematic to be sure everything is wired right. My head hurts just thinkin' about it!
 
the unit hummed with the power tubes removed? I don't understand

Do you have a hum balence control? A bad humcontrol can cause hum, check the wiped to outside ends maybe the wiper lifted from the element.
 
Back
Top