1980 Fender twin Reverb 135 Hum

[Lou Popadoo]

I recapped the power supply, and all the electrolytics of my 1980 Fender Twin Reverb 135.  The amp worked really well, but it had an annoying 120 cycle hum, which is the reason for the recap.  Once inside the amp it appeared that someone had done a poor attempt at a mod on the Tube Balance pot.  I restored all wiring to the original schematic, and replaced the heater hum pot, which was bad.  My bias voltage is spot on, as well as all the other voltages.  Everything works, except that 120 cycle hum, is now louder than before.  It is there with all pots down and nothing plugged in.  It is there with tubes 1-6 pulled.  It is there until I pull the last power tube.  And it gets lower in level as I turn the master volume up!??  I’ve been through this amp 100 times now, checking every ground and chop sticking every connection.  I have a 1975 Dual Showman Reverb that I also recapped, which sounds amazing.  It has an identical master volume pot, which I swapped out to the twin, no change.  I’m at a loss.  Any insight will be much appreciated.

[shooter]

And it gets lower in level as I turn the master volume up!??

after all the tubes are out?

whenever I get 120hum, I measure the 1st PS tap for VAC (ripple), do the same to your showman, are they close?

[NewYorker]

Lou:

Did you replace the bias cap?  (70mF / 100V on the schematic)  If so, is it connected the right way?  Since the bias voltage is negative, "+" on the bias cap is grounded, not "-".

If you didn't replace the bias cap, and the old one is bad, I think that could cause the hum that you're describing.

However, the thing about the hum going down as you turn up Master is puzzling.

Ed

[Lou Popadoo]

When all the tubes are out there is no hum, but there is hum up until the last power tube is removed.  I replaced the bias cap with an 80mF @ 160v cap, and the positive end is going to ground.  Can you give me a better description of the 1st PS tap? I'm not sure which wire you meaning. 
Thanks to all.

[shooter]

1st PS tap?

the one that feeds the power tubes

measure it with all the power tubes in, compare it to your showman, I would think they should be close.

[Lou Popadoo]

Here's the whole voltage story. The showman on all power tubes;
Bias -51.7
Plate 457v and 4.35VAC
Grid 457V and .075VAC

The Twin;
Bias -58
Plate on all 4 tubes 512v
Plate on Tubes 7&8 4.4VAC
Plate on Tubes 9&10 1.325VAC
Grid on all 4 tubes 512V
Grid on Tubes 7&8 3.12VAC
Grid on Tubes 9&10 2.7VAC

[shooter]

you show plate voltage on both amps grids, I would expect to see bias voltage there, not plate voltage.

.07vac seems like a good # to me
3.0vac seems suspect to me

[Lou Popadoo]

Sorry, I gave you Screen grid voltages.  the Control grid voltages are the negative bias voltages, with OVAC.

[shooter]

so are the tubes from 1980?

got any spare Ecaps?  If so try doubling up - in parallel - with the new ones and see if the hum goes down.
(you can just gator clip)

as I turn the master volume up!?? 

Is that a post PI MV?

[NewYorker]

"Is that a post PI MV?"

According to the schematic, no.  However, it's wired a bit differently.

But getting back to the varying-hum vs MV-setting issue - what tubes are in place when you notice this?

I've heard sort-of similar behavior in a self-built amp that I did years ago - hum was probably coming from multiple stages of the amp, but in different phases because of how each gain stage inverts the signal.  My theory was that these hum components can cancel each other at certain Gain and MV settings.

Just a thought.

Ed

[NewYorker]

To clarify - the Master Volume wiring shown in the schematic is a bit different from what I normally think of.

http://el34world.com/charts/Schematics/files/fender/Fender_twin_reverb_sf_135_schem.pdf

Ed

[shooter]

were just the power tubes in when you noticed the MV effect?

for testing, disconnect the wire I marked and see if the MV still affects humm

[Lou Popadoo]

Not sure about the age of the tubes.  They are;
1 RAM 12ax7
2 EH 12ax7
3 No Name 12AT7
4 RAM ax7
5 Groove Tube 12ax7
6 Fender 12at7
Power tubes are Sovtek 6l6 wxt+

It is not a post PI MV.
The amp hums and has the MV effect with all the tubes or with just one power tube, or any other combination you can think of.  It only goes away with no tubes.
I will try both of shooters ideas

[Lou Popadoo]

I removed the MV shield from the 100ohm, 820ohm, 22K juncture as requested by Shooter.  No change.  I'll try the parallel caps, but have a question.  Am I just going to parallel the 2 220mF caps?

[Lou Popadoo]

I don't know if this is related to my hum issue, and i don't know if this is a mod or not, because it doesn't appear on my schematic, and this component is not on my Dual Showman reverb.  But, there is a 10M resistor going to ground that is tied to the Bright Switch end of the 120pF that goes from the volume pot to the Bright Switch on both channels.

[shooter]

If it's humming with just 1 PA tube and nothing else, there isn't much "in the circuit"  I suspect there might be a problem in the bias supply, or some ground loop positive feedback problem.

with all tubes out, meter on Volts DC, measure your bias at the power tube socket, pin 5, (tube side of socket), to ground.  Note the value.  adjust the bias balance pot both ways and note values. re-set to original value and switch meter to VAC and record value.


And yes, just clip cap in parallel make sure of POLARITY.

[NewYorker]

"there is a 10M resistor going to ground that is tied to the Bright Switch end of the 120pF"

Lou:

I think this is unrelated to the hum issue.  The 10M resistors are likely intended to prevent "pop" or "click" noises when the Bright switches are turned on.  Putting the resistors across the Bright switch terminals would accomplish something similar.  But in any case this should not be your problem.

Shooter:

Any reason to suspect a problem in the OT?  Or maybe the last 6L6 that Lou left in is bad?

Ed

[sluckey]

Something to consider... There is normally a lot of 120HZ ripple (hum) at the power supply node that feeds the output tube plates. No big deal because that ripple is cancelled by the common mode rejection characteristics of the push/pull tubes and OT. But, when you pull the tubes from one side of the OT, you lose that common mode hum cancellation. I would expect the hum to increase if only half of the push pull circuit is working. That's also why grossly unbalanced OT currents (mismatched tubes) will cause hum.

[shooter]

+1

for me, I would get a known good set of PA tubes, and PI tube, bias and get that section working idling no signal, all other tubes out, all pots 0.  Then I can move left, fixing as I go til I get to my guitar

[Lou Popadoo]

First, thanks to all for your input.
To New Yorker; I've tried all 4 6L6's as the last tube in, all with the same results, and yes I do suspect the OT.
To Sluckey; The hum actually goes down in level as I remove the power tubes one by one.
And to Shooter; With all tubes out with tubes balanced;
#9 -58.67V
#10 -58.67V
#11 -58.44V
#12 -58.44V
With Tube balance pot fully CCw;
#9 -60.24V
#10 -60.24V
#11 -52.37V
#12 -52.37V
With Tube balance fully CW;
#9 -52.83V
#10 -52.83V
#11 -60.23V
#12 -60.23V
With Tube balance reset;
0VAC on all
And your last response about new tubes is definitely under consideration

[Fonz]

Hi all,

I've been working on a '78 TR135 with the same hum issues. It had some weird breakup that was fixed with new power tubes (the whole thing was stock when it came to me) so I replaced those with a match quartet and all the filter caps that were leaky. Then measured everywhere and theres no VAC on the plates or bias. The bias capacitors measured fine so I left them. All VDC measurements where spot on according to the schematic. Hum went down, but it's still there. The hum balance does squat and the hum kinda of goes away as you turn the MV up. Granted it's super loud it hard to tell if it's really going away. I noticed the hum disappeared even more when I had an 8Ω load on it and turned the volume up. But ultimately it's still there.

I've poked it with a chopstick everywhere, jiggled wires and tubes, swapped preamp tubes, checked all diodes. Even tried pulling power tubes like was suggested to OP. Kinda stumped on this one so I thought I'd chime in.

One thing I've been thinking about checking is the coupling caps. Maybe some of those are leaky. Just had an '65 Ampeg Gemini II that needed all the coupling caps replaced and it got dead quiet. Was thinking maybe thats the case?

I will say, the specimen before me is quite messy. Tons of extra wire, lots of signal-carrying wires sloppily jammed here and there in between components. Jeff Bober has a couple articles about these amps for Premier Guitar and the pics show much nicer craftsmanship than mine, even tho this is all in the CBS era. Maybe all the haphazard wiring is to blame? Or the layout in general? The balance/bias circuit does goes across the amp to get to its spot on the rear of the chassis, where the wires to the grids go around/across a couple tubes. Blackface bias layout didn't get near as much signal.

I've changed the balance pot to real bias pot, and it has made no difference in terms of hum. The original balance circuit added hum if anything.

Bober also says this about the balance circuit; https://www.premierguitar.com/articles/22353-ask-amp-man-restoring-a-1979-fender-twin-reverb

 "Balanced tubes in the output stage are a great idea, and I believe Fender engineers added this feature because tubes were still being made on old machinery. This machinery was not well maintained because—let’s face it—tubes were on their way out, and precision fell by the wayside. This led to tube sets that were not as closely matched as they’d once been, causing output-stage hum. Hence the output-tube-matching control, which alters the bias voltage on one side of the output stage relative to the other. This matches the total current flow of one half of the output transformer primary relative to the other half, eliminating hum. (This control was referred to as “hum balance” at the time.) We really don’t need this function nowadays because matched output tube sets are readily available. I feel it’s better to install matched output tubes and turn this control into a true bias control..."

Sluckey is kind of backing this up if the ripple ac on the plate nodes is cancelled out.

This is the first UL amp I've ever worked on, and I noticed the power supply has a full wave rectifier but has a center tap on the winding that is tied to the MIDDLE of the 220uf capacitors.  I looked up other UL designs like from the Marshall Major series, and they all do the same thing. But heres the weirdest part... I put my meter on the center tap and find not just 250VDC (to be expected) but 545VAC  Is that normal?

Lou - do you measure any VAC on the center tap (red/yel)?

Thanks

[PRR]

Coupling caps don't make hum. They should just couple hum along. Forget them.

> center tap and find not just 250VDC (to be expected) but 545VAC

Per the plan, that seems to put 500V AC across a large e-cap. Which would have burst by now. Something is wrong with the measurement or the amp.

> hum kinda of goes away as you turn the MV up.

What about all the other knobs, which ones change the hum?

MV "up"- turned up a little? Turned up a lot? At half-volume the pot wipers are most susceptible to crap around the wires from pot to power tubes. How tangled is that wiring with power wiring?

[pdf64]

...I've been working on a '78 TR135 with the same hum issues. It had some weird breakup that was fixed with new power tubes...

What idle plate or cathode current are each of the power tubes drawing?

...tried pulling power tubes like was suggested to OP...

What happened to the hum?
How about with the power tubes in place but the tube for phase splitter V6 removed?

...Maybe all the haphazard wiring is to blame? Or the layout in general? The balance/bias circuit does goes across the amp to get to its spot on the rear of the chassis, where the wires to the grids go around/across a couple tubes. Blackface bias layout didn't get near as much signal.

I've changed the balance pot to real bias pot, and it has made no difference in terms of hum. The original balance circuit added hum if anything...

The power tube matching / bias balance control is pretty much a necessity with this range of amps, due to the combination of the distributed load OT and poorly filtered HT. Normally with UL / DL power amp arrangements, the HT to the OT is well filtered with a CLC supply, eg http://schems.com/schematicheaven.net/manu/sunn/sunn_model_t.pdf
It's all down to the screen grids; the clue is in the name, these are grids and have gain  so any HT ripple fed to them will modulate plate current.
I doubt many tube vendors include g2 gain in their tube matching characteristics, so failing that the bias needs tweaking to get those ripple currents to match on both p-p sides of the OT and so cancel out at idle.

This is much less of an issue with a regular pentode p-p output arrangement, as the HT supply to the screen grids is almost always much better filtered.

The wiring to the bias balance control may have looked messy but it's unlikely to have caused any performance issues.

...Bober also says this about the balance circuit; https://www.premierguitar.com/articles/22353-ask-amp-man-restoring-a-1979-fender-twin-reverb ...

Although he's a great tech, I think Jeff Bober is off target with some aspects of his analysis above.

1/ I think Fender introduced bias balance as early as 1965 with the AB165 Bassman; that's still in the golden age of tube manufacture (indeed, when did vintage tube manufacture ever go bad as he suggests?). Tubes have always had wide spread of acceptable performance characteristics either side of their type bogie, eg see p2 of http://www.mif.pg.gda.pl/homepages/frank/sheets/084/k/KT88.pdf re the recommendation for a bias voltage range of +/-25%.
So to have reasonably similar power tube idle currents for p-p amps, the options have always been to buy selected power tubes, select suitable sets in-house, or to have independent bias or bias balance.

2/ With Fender, the 'hum balance' control was always in the heater circuit.

3/ The 'power tube matching' control function is still beneficial even with apparently 'matched' tube sets, as the tubes' g2 characteristics are not generally included in the matching criteria.

...I've changed the balance pot to real bias pot, and it has made no difference in terms of hum. The original balance circuit added hum if anything...

For the reasons noted above, I suggest to re-instate the 'power tube matching' control, and to fit an additional trimmer inside to set the overall bias voltage level.

... I put my meter on the center tap and find not just 250VDC (to be expected) but 545VAC  Is that normal?...

Yes, I think that's normal, the issue is that your meter can't provide Vac measurements when there's Vdc present.
A better meter, eg Fluke 87, should give reliable measurements in such circumstances.

[2deaf]

Wow. pdf64 is here.  How cool is that?

[sluckey]

I think it's very good. Welcome to the forum.

[bnwitt]

Welcome Pete.  I've seen your posts over at the MEF quite often and enjoy your perspective.
Barry

[pdf64]

Wow, I didn't expect that, thanks for the welcome 
This thread turned up in a search, so having had some experience with these amps, I thought I'd chip in, hopefully it's helpful

[Lou Popadoo]

To Fonz,
I get 1.5VAC on the center tap (red/yel).

[Lou Popadoo]

So, per Shooters advice, "for me, I would get a known good set of PA tubes, and PI tube, bias and get that section working idling no signal, all other tubes out, all pots 0.  Then I can move left, fixing as I go til I get to my guitar". So I bought a new set of tubes with matched power tubes and now the amp is just fine. I didn't have to go any further.  As part of my original trouble shooting I swapped out each of the old tubes in every mathematical combination known to man, and my hum always persisted.  I'm perplexed, but quite happy.  It turns out that my work was just fine, just needed some new tubes.  Thanks to all.

[shooter]

I'm perplexed, but quite happy.

I swapped out each of the old tubes in every mathematical combination known to man

when you put a noisy tube, back in the same path, regardless where, it will probably "modulate" the audio every time

glad you're up