Univox U45 – 2 Service Techs, Still Not Fixed

[gtr_joe]

Hi Folks. Long story short, I have a Univox U45 tube amp. (Different circuit than the U45-B.) Developed a minor hum. Took it to a reputable tube amp guy. He replaced components, tubes, then had me buy another matching power transformer. He still couldn't get it to work. Took it to a 2nd guy. Also very reputable. He said he tested or switched out all the components. He got it to work, but here's the weird thing:

There's a hum at the lowest volume. The hum clears up and goes away from 12 o'clock to 2 o'clock on the volume dial. And then returns very loudly when you dime it.

Has anybody ever heard of this? The 2nd guy even got the tremolo circuit to work, but couldn't eliminate the hum as stated above. The schematic looked pretty simple. (There's a copy of it on this site.) And the amp used to run beautifully flat out without significant noise.

Anybody got an idea of what's wrong? Thanks.

[Ed_Chambley]

How are your heaters.  Are they center tapped?  If not check the 100ohm artificial centertap resistors for drift.  There are lots of things that cause noise.  Plate resistors, even if replaced.  Ground loops.  Try some chopsticking to see if you can find a component that makes a change in the noise.  Finding noise can be difficult, but if you start with the heaters, then the plate resistors this is where most times I find the problem you have that shows differently at the volume pot.

I had 2 princetons in for repair which were doing the same thing.  I installed artificial center taps on both and cured the problem.

[HotBluePlates]

There's a hum at the lowest volume. The hum clears up and goes away from 12 o'clock to 2 o'clock on the volume dial. And then returns very loudly when you dime it.

...

Anybody got an idea of what's wrong? Thanks.

You didn't happen to get a new speaker for the amp, did you? And did either of the repairmen replace the filter caps? Did they use values larger than the original 10μF and 3μF?

This amp is much like a Fender VibroChamp, but with a 7-pin version of the 6V6. The Fender amps used a small 8" speaker in an open cabinet in part because that setup has weak low bass response right where the 120Hz power supply hum will occur. So if you swapped out to a better, newer speaker you might have gotten a model that has better bass and lets you hear hum that was always there but you never noticed before.

So let's say that isn't your issue. What else could be the problem?

You have hum at 0 volume and at full volume, not in the middle.

Look at the schematic and see that if the volume control is full up, any hum present at the input 12AX7 plate will be passed on to the output tube. Hum may be there because the filter caps are old and worn out. It may be there because the 1.5MΩ resistor from plate to ground is too close to a source of hum like heater wiring. It may be there because overall the power supply has pretty weak filtering (and a lot of the stages seemed choked down to low current to allow the small filter caps).

Look again at the schematic and see that if the volume control is at zero, the only source of hum is the 6AV6 and the 6AQ5 output tube. How would those have hum? Well, a single-ended output stage doesn't cancel hum at the plate of the output tube and push-pull amps do. So a common problem is hum at all volume settings (but obviously more noticeable at lower volume) because the single filter cap at the power supply point feeding the output transformer and output tube plate just isn't enough ripple reduction.

A common fix for the second possible problem is to add another filter cap and smallish resistor between the rectifier output and what is currently the first filter cap after the rectifier. Try a 16-20μF and a 500-1kΩ resistor. It wouldn't be a bad idea to bump up all the other filter caps to 16-20μF. That should address the hum at no volume.

Another possibility for hum with no volume (for an entirely different reason, and not peculiar to SE amps) is the need for a 3-prong power cord as the existing plug has a cap attached to one of the prongs which is intended to be the neutral lead. If your plug isn't polarized, try flipping the plug over in the socket to see if hum is reduced in one direction over the other. If it is, you should cut out the cap connected to the power cord and have a 3-wire grounded cord installed. The amp will be safer and this source of hum will be corrected.

The first potential cause of hum that I mentioned may be cured by fixing the power supply issue I mentioned with regard to hum at zero volume. If it isn't, the big 1.5MΩ to ground may be a problem. Really, that resistor is not needed, and it also presents a high impedance which tends to pick up hum out of the air, and injects it at the plate of the input stage. Essentially, a high resistance point like this near any significant a.c. (like the heater wiring of the tube or any power supply wiring) will pick up hum from the a.c. source which is then present to be amplified by the amp.

I suspect there is more than one issue in your amp. I also am assuming that basic things are correct (like circuit grounds).

[PRR]

> I suspect there is more than one issue in your amp.

The symptom suggests _2_ hums. One, small, in the output stage. One, larger, in the input stage. Out-of-phase. At some midway point, they cancel.

> I also am assuming that basic things are correct (like circuit grounds).

I would not assume that. After re-capping (which "has to be" part of what _2_ techs did?), my 2 cents is on a loose ground rivet.

[gtr_joe]

Thank you everyone for your thoughts and advice. I greatly appreciate it!

[jjasilli]

Ground rivets on old amps are a source of issues.  At each ground connection, use a meter to measure the continuity (resistance) from the top of the grounding lug to the chassis.  If more than a small fraction of an Ohm, you have a culprit.

If that doesn't work, I would temporarily disconnect the signal feed from the preamp > power amp.  Then:

1.  Feed clean signal from an outside source into the U45 power amp.  If there's hum, then there's a problem in the U45 power amp or the portion of the power supply feeding it.

2.  Feed clean signal from an outside source into the U45 preamp. Use a listening amp to detect any hum coming from the tail end of the preamp signal.  If yes, then work backwards, one signal connection at a time toward the input jack.  If the hum stops at some point then you have focused in on the problem area.

[PRR]

> Ground rivets on old amps are a source of issues.  At each ground connection, use a meter

Bah. Ground-through-chassis is a cost-cutting technique. When it fails, instead of "testing" (with dubious results: pressing the probe on a rivet may momentarily fix the contact), it is MUCH cheaper to do the right thing: run and solder a wire through all ground lugs. Follow the flow from input to output and power-cap.

[adammwhite]

Hey Everyone,

I know this is an old topic, but I wanted to ask some questions along the same lines. I just picked up a Univox U45 on craigslist for a good deal. I did my usual routine and immediately replaced the caps (all of them). The first filter cap had visibly given up the ghost, so it was a no brainer. I fired it up slowly on the variac and it works and actually sounds great. But it has the same problem/s expressed in the original post. At about noon on the volume, the noise is not that bad. On either side of noon, the noise is pretty bad (worse all the way up and all the way down). I have a few suspicions. For one, the resistors have all drifted pretty badly and I'll be replacing those as well. I'll try to do that this week/weekend and maybe that will help. I also feel like the shared ground bus in the amp is a flawed design. There is no separation between the power/preamp from a grounding perspective. I have included two pictures, one that shows the schematic and where I would propose cutting the ground bus. The second is a picture of the back of the board just for reference. Thoughts on the proposed ground bus split location? I would just remove the trace and run the wires to ground these sections separately. Thanks in advance for your help!

Also, the center tap of the 6.3v PT tap is grounded with a wire over to the HT center tap...  I assume that's ok. I think I've just never seen it done that way, because I have mostly worked with transformers that have leads coming out of the bell... not ones that have external connection tabs.

[sluckey]

On the schematic you have placed a red vertical line on the B+ rail, not ground?

[adammwhite]

Yes, that's true... there's no ground rail represented on the schematic, so I thought this would work. Let me be more explicit. What I meant to indicate is a split between the second 10uf filter cap and the 3uf filter cap. Making sure that the 6AV6 cathode bypass is on the preamp side and the 6AQ5 cathode bypass is on the power amp side (ground wise). Think of the line as what's on the left is preamp ground and what's on the right is power amp. Make sense?

[shooter]

try these;
all the grounds I circled should come together at only 1 point, the input jack, or close.

can't tell what the PS filter caps are, I typically use 47uF.
guessing it has been converted to a 3 prone, dead cap removed?

I'd look to a fender 5E1, 5F1 for NFB

[adammwhite]

Yes, the power cord has been changed out to a 3 prong and the death cap has been removed. The NFB question was a leftover from the original post on this topic. I don't think I'll add any NFB, I'm just trying to tame the noise:) I think the REALLY out of spec resistors are part of the issue, but I think the grounding can be improved. Thanks for the help, I really appreciate it!

[adammwhite]

One other question on the diagram where you circled the ground. I was thinking the 6AV6 cathode would be grounded at the input as well, correct?

[sluckey]

That's correct since it is powered by the same filter cap.

[adammwhite]

Awesome, thanks! BTW, I did replace the first two 10uf filter caps with 10ufs, but I bumped the 3uf up to a 10uf. So I could go up more on those caps. I'm sure there's a point of diminishing returns on filter caps, but would you be able to notice a difference in 60hz noise between 3x10uf caps and 3x47uf caps (hypothetically)? The 6x4 datasheet recommends a 10uf cap for a full wave capacitor input filter, but is that really a minimum value? Just curious:)

[kagliostro]

6X4 Rectifier


datasheet say max 40uF as first e-cap, so you can use

First cap 40uF and the oter one as you like more (in increase respect to the original)

BTW, 40uF and 47uF are surely close values and there will be no much problem if you use a 47uF as first e-cap

Franco

[ac427v]

I worked on one for my brother. Cool amp with great overdriven sound! The biggest noise culprit was the ground loop at the trem jack. Here is a summary of what I did:

edit--OOPS Wrong U-45. The one I worked on was a push-pull amp with the same model number. You might want to look at my efforts anyway in case both amps share the same layout/construction issues.
--Craig

[adammwhite]

Thanks for sending your notes on what you did! Please take this the right way... it was most helpful for what you did that didn't work:) I was thinking about doing a lot of those things and that saved me a lot of time! I still need to change a lot of resistors because they are all really out of spec. I replaced the first two 10uf filter caps with 47uf caps and that made a HUGE difference! I can actually live with the noise now. I do want to experiment with the Trem ground loop you mentioned. The U45 doesn't have a footswitch jack, but I might try to isolate the trem pot with a plastic washer, just to check it out. Also, I want to try to split the ground plane between the pre/power side. They didn't really make that easy with the board layout, but I still might try it. It probably would have been faster at this point to have just done a perfboard rebuild (still might do that:). Thanks again for everyone's help. Amp work is what I do for fun when my family goes to bed:) Y'all have a great day!

[adammwhite]

Upping the filter cap size has helped noise immensely, but there's still a baseline that I believe is heater noise. The heater center tap is jumpered right over to the HT center tap. I have been reading more about heater CT elevation and possibly running the 6.3v CT to the 6AQ5 cathode. Is this a viable thing to try?

[Deric]

It's worth a try and I'm curious to see if it helps.  I have one of these that sounds awesome and has just a tiny bit of hum.  If it helps yours at all I may give it a shot.


Mine was very noisy (hum) and not fully functional when I got it.  I ended up replacing every single cap and resistor.  A lot of the resistors were WAY out of spec.  I used 10u, 10u, 4.7u for the filtering.  Cleaned/serviced jacks and sockets per usual.  Fired right up and sounds killer.  Fun little amp to play.

[adammwhite]

Ok, so I couldn't leave well enough alone with this U45. The noise was bothering me and I knew there had to be a solution. All of the steps I took (complete recap, moving wires, etc...) made some difference, but even without any tubes in there was noise coming through the speaker. The OT and PT in this amp are crazy close to one another and the PT Secondary faced the OT Secondary head on (with maybe an inch between them). I considered drilling some new holes and mounting the OT sideways, but it would have been tight. Instead, I had an epiphany, inspired by Leo himself. I just mounted the OT on the backside of the chassis and re-ran the wires. The difference in noise is night and day. The previous noise level turned way down is now the noise level all the way up (and there's just nothing you can do about that, it's called gain:). Anyway, I thought I would share that and I appreciate everyone's input.

[Weslito2]

Has anyone been able to solve this?  I have the exact same issue with this U-45 amp.  Any advice would be greatly appreciated. 

[John]

Ok, so I couldn't leave well enough alone with this U45. The noise was bothering me and I knew there had to be a solution. All of the steps I took (complete recap, moving wires, etc...) made some difference, but even without any tubes in there was noise coming through the speaker. The OT and PT in this amp are crazy close to one another and the PT Secondary faced the OT Secondary head on (with maybe an inch between them). I considered drilling some new holes and mounting the OT sideways, but it would have been tight. Instead, I had an epiphany, inspired by Leo himself. I just mounted the OT on the backside of the chassis and re-ran the wires. The difference in noise is night and day. The previous noise level turned way down is now the noise level all the way up (and there's just nothing you can do about that, it's called gain:). Anyway, I thought I would share that and I appreciate everyone's input.

[rsilverst]

Ok, so I couldn't leave well enough alone with this U45. The noise was bothering me and I knew there had to be a solution. All of the steps I took (complete recap, moving wires, etc...) made some difference, but even without any tubes in there was noise coming through the speaker. The OT and PT in this amp are crazy close to one another and the PT Secondary faced the OT Secondary head on (with maybe an inch between them). I considered drilling some new holes and mounting the OT sideways, but it would have been tight. Instead, I had an epiphany, inspired by Leo himself. I just mounted the OT on the backside of the chassis and re-ran the wires. The difference in noise is night and day. The previous noise level turned way down is now the noise level all the way up (and there's just nothing you can do about that, it's called gain:). Anyway, I thought I would share that and I appreciate everyone's input.

Hey Adam - I'm new to the forum, with someone in another forum having pointed me to your post on the U45B. I am working through the same issue that you described. I'm going to replace PCB components first, since many of them are way out of spec, but I'm ready to do the OT move if need be.

My question:

How, if at all, did you protect the leads coming off the OT - there are high voltage points exposed - normally an OT has these solder points internal when it is meant to be mounted on the outside of the chassis. I worry that I or someone could inadvertently reach under and touch 260 Volts. I could imagine fashioning a little folded aluminum box over it that screws to the chassis. Not sure if just taping the leads would be sufficient? Your photo only showed the absence of the OT, not what it actually looks like underneath :)

Thanks!
Bob

[Latole]

I have already repaired a Wem Dominator Mark III amp.
P.T show external lug where wires are connected. No insulation.
Many European ( old ? ) PT are built like this

IMO no need more insulation, access inside a amp chassis is for who know what he is doing, not neewbee.

[ac427v]

I now believe these amps hummed from day 1 because the OT and PT were next to each other mounted inside the chassis. 
The speaker hums even when there are no tubes installed. 
The question is "how to safely move the OT and its' exposed terminals to the outside of the chassis?"
Can you solder wire extensions to the terminals and cover with heat shrink insulation? I can't remember the details of the one I worked on.

[rsilverst]

I now believe these amps hummed from day 1 because the OT and PT were next to each other mounted inside the chassis. 
The speaker hums even when there are no tubes installed. 
The question is "how to safely move the OT and its' exposed terminals to the outside of the chassis?"
Can you solder wire extensions to the terminals and cover with heat shrink insulation? I can't remember the details of the one I worked on.

That's definitely possible to solder extensions and heat shrink. That was my plan. But... the "dangerous"-ish part is that the terminals where the wires attach to the OT are exposed on what would be the side of the OT that a hand would first run into if fumbling blindly in the back side of the amplifier. Here's a photo of how the leads attach. If you contrast that with the OTs that are standard on Fender amps where the OT is external to the chassis, there would not be any exposed junctions where one could touch the high voltage.

[sluckey]

Silicone RTV should fix that.