66 Super Reverb no so super sounding

[thumbs]

Did the amp sound that bad before the cap job? If not, I'd go back and check my work. The symtoms sound similar to an amp I worked on a few months ago. Problem turned out to be a broken lead inside the cap can. Because the amp had recently been recapped, and I assumed it was OK in that department, it took me a long time to get there. You might also measure the resistors inside the cap can while you're in there.

[booj]

I set up the amp to drive a 2 ohm resistive load, and listened to dual 4 ohm headphones through two 82 ohm resistors, one resistor going to each side of the headphones. That put out about the right voltage to listen to the sound painlessly.

Super Reverb Amp AB763 Schematic.pdf

Here's the DC voltage measurements. The zero reading at pins 3 and 8 of V5 seems normal because of the tube being turned off by the - bias voltage on the grids.
The power supply is in good shape. the readings may be a little high compared to the scat due to my 123 volt line.
Just for kicks I tried putting a .0022uf/630v cap between grid and ground of the power tubes. The noise was a little less high endy, but was there nonetheless.



Once I try the amp again with the Dual Showman output tranny under proper load, I'll check in again.

[Pedro Vecino]

It seems some kind of oscillation. You can try:
Although not directly related to this, another rectifier. If the rectifier is bad can cause very strange effects with volume. (I use always 5U4 in Superreverb. Not 5AR4/GZ34. With 440V approx. sounds more sweet and relaxed)
Check good earth connection in speaker jack. Tighten nut
Disconnect feedback loop in the board. You will find more overall distortion but probably the transition to overdrive will be "musical".
If is OK. Change input transformer leads and apply the feedback to the board. Put your hand on standby because it can start to scream.
Previously, iIf you distrusted of the insulation on the board, raise the bias feed resistors, ends of 0.1uF caps and the outputs to power tube grids. Connected together but not touching the board.
Regards

[booj]

The latest. I hooked up a listening set-up ala Doug Hoffman. Should have done that sooner! I can hear the signal through my shop amp using a 1 meg pot commonly grounded with the hot side going to the source and the wiper going to the shop amp. Since i've tried a Dual Showman and a Bandmaster known good trannys, it looks like it's not the tranny.
the distortion seems to originate from the feedback circuit. The signal for the bottom driver half is derived from the combination of signals from the 100 ohm/820 ohm junction, and the signal impressed on the 470 ohm/22k junction. If I short the 100 ohm resistor in the feedback loop, the signal loses its nastiness. There is still distortion present, because with the volume turned up to 10 and a strong stum on the guitar (into the normal channel for this test), the driver stage goes into distortion even though the input is totally clean.
so the input to the top half of the driver is clean, but its output is distorted without the nasty sound. The input to the bottom half of the driver is nasty unless I short the 100 ohm resistor.
It's kind of looking like a problem then with the output section. When I remove the output tubes, I still get that driver distortion, but no nasty sounds.
I've tried swapping output tubes to no avail. Perhaps the output tube sockets are leaky? I don't know where t find tube sockets that are an exact replacement, otherwise I'd jump on that possibility. I know Angela has ceramic drop in types, but I'm thinking esoterically here. The output jacks are tight, the tubes are drawing about 33 ma. ea. Voltages normal.
I guess I'll do a cleaning job on the sockets as well as making sure the sockets are tensioned properly. 'Bout time I guess, huh?
Be back with more results. What's the vote for new output tube sockets?

[Chuck H]

Didn't think it was the OT.

I don't think it's the tube sockets either.

You have done some good trouble shooting research. Take a tip from what you've found. The feedback loop is the cause. Something may be causing phase errors in the loop. Are there any tweener kind of silver face circuits in there? And/or check the route for the NFB lead. Make it as short as possible while still routing it away from the preamp. And shorten any preamp grid leads that are near the PI board circuit. Maybe check the .1 bypass cap on the PI for leakage. Have you checked the NFB loop resistors for drift. If the ratio is too high that could certainly cause your problem. Too much NFB is worse than none for ugly distortion.

[Chevy]

I think you mentioned low end distortion... you have a signal generator, right ? Have you isolated the general frequency range that is distorting? If it's just below 100hz or so (just for example), perhaps it's strictly a power supply problem? In other words, check out a 1khz input sine wave, see what it does (and how much output power you can generate), and then do the same with an 80hz input. Any notable oddities?

[booj]

the latest. Replaced the tube sockets with porcelain types from Angela. Actually found a bad solder connection to one of the filament leads. The lead was oxidized. Thought fixing that would help, but no. Replaced the screen resistors (again) and the grid resistors. Also changed the 820 ohm and 100 ohm feedback resistors. I've scoped, measured and listened to this thing to death. The space between the fiber boards is clean. All of the power supply caps and bypass caps jare new, as are the coupling caps in the driver stage. (I know that doesn't necessarily mean anything) The problem persists. The layout looks good.
The problem sounds the worst at the junction of the 100 ohm and 820 ohm neg. feedback resistors.
Apologies if I haven't acknowledged everyone's suggestions individually. I really do appreciate your input, and have indeed taken them into account. The amp sounds good with the 100 ohm resistor shorted. That may be where it stays unless someone comes along with exactly the same problem and has a fix that works.

[MWJB]

Your grid wire to V8 pin 1 is very, very long indeed, try shortening it. Perhaps route it over the V7 socket, rather than looping under past the standby etc.

If the ground is good at the end of the 100ohm NFB load & connection from 820 NFB dropper to speaker jack is good, values check out, then what can be wrong with the loop? I think that you are just seeing the problem manifesting itself here...a symptom rather than the root cause. There is a slight imbalance at dc grid voltages at V6 pins 2 & 7...can't really say whether it is pertinent, but you may as well change the 1 megs.

[MWJB]

Maybe also run the PI grid wires suspended over the plate wires (which I'd push down to the chassis floor), you might be able to shorten them a little too.

[thumbs]

On the layout drawing, I see 4 resistors and a cap flowing into the junction of the 820 and 100 ohm resistors. Could one of those be throwing things off?

[booj]

I wonder how the length and layout could make it sound bad considering its the same as on the Silver Face amp? All the resistors measure well within tolerance. Believe me I'm not jumping to conclusions here. Just wondering.
Tonight I tried swapping output transformers, putting the SF (known good) tranny into the BF chassis. It made no discernable difference. I also tried lifting the whole junction of the 820 ohm, 100 ohm, .1uf cap and 22k resistor off the fiberboard to no effect.
All while the customer was watching. No effect. In the end, I put the two amps back together as they were, tried them both out again and sent him on his way. We were both satisfied we had tried everything possible and both frustrated because the problem persisted.

[MWJB]

I looked at other Super Reverbs and you V8 grid wire seemed much longer than them...you didn't post pics of the SF amp? Things are only "the same" when in fact they are "the same", anything else is "different". SF amps have grid to ground caps to prevent oscillation...sometimes they are needed other times not. In some amps just an extra 1" of grid wire can screw them up. As you have never presumably heard the SR in question sound great, we may as well assume that the reason it is lacklustre could go back to the day it was made.

Let's recap - OT good (hadn't you already eliminated the OT as the culprit before swapping with the SF SR OT? The problem isn't going to be the component you want it to be, just because it's what you want it to be), voltages good, resistors good, tubes good, speakers good....then it's something else. Troubleshooting is about eliminating the things that it isn't, then whatever is left should be the issue.

[Chuck H]

+1 on stringently checking the layout / lead dress. AND...

An important observation is the that the amp sounds much better with the NFB loop disabled. This leads ME to think that the NFB lead may be carrying peripheral emf back to the PI tail and /or there are phase errors causing borderline instability. This may be lead dress and / or ground point selection.

Just for kicks... When you disconnect the NFB loop, does the amp get louder?