Reverb Squeal when turned up (SOLVED- for real)

[waldner]

My Deluxe Reverb clone (Vintage Sound 22) works great and sounds great except, when the reverb gets above 5 it starts to feedback and squeal.  I did some testing and it's not the reverb tank as the problem still happens with no tank connected and both channel volume controls turned all the way down.


I tried adding a .002 cap in parallel with the V4 grid leak resistor to ground as shown in the attached Rob Robinette pic.  Didn't help. 


The only other piece of info on this circuit that may or may not be relevant is that it's wired with both channels having reverb.  I even tried pulling V1 to eliminate the Normal channel as much as possible, but the squeal still happens.


And, one more thing... I have a dwell control but it's set pretty low to tame the otherwise more splashy reverb.


Any ideas?

[tubeswell]

microphonic tube?
NFB loop connected backwards?

[tdvt]

The only other piece of info on this circuit that may or may not be relevant is that it's wired with both channels having reverb.  I even tried pulling V1 to eliminate the Normal channel as much as possible, but the squeal still happens.

I have just been reviewing methods/schematics for two-channel reverb on a Fender circuit, thinking about trying it out on a current project.

How is the normal/non-reverb channel connected to the reverb circuit?

From what I have gathered, depending on the reverb-mod circuit design, there can be some channel interaction. Maybe this figures into your problem.

Did it ever work correctly or is this an issue with a new build? 

[waldner]

microphonic tube?
NFB loop connected backwards?

Thanks Tubwell.  Turned out it was the V4 tube.  It's a newer JJ that didn't show signs of microphonics when tapped, but must have been the cause.  I changed to a different 12AT7 and can now turn the reverb up to 10. 


I wish I would have tried changing that tube first.  I was fooled by it passing the tap test.

[waldner]

TVDT,

Here's a good way to add the normal channel to the reverb from Rob Robinette: https://robrobinette.com/AB763_Modifications.htm#Normal_Channel_Reverb_Mod

[sluckey]

Turned out it was the V4 tube.  It's a newer JJ that didn't show signs of microphonics when tapped, but must have been the cause.  I changed to a different 12AT7 and can now turn the reverb up to 10. 

FYI, V4 should be a 7025 or 12AX7, not a 12AT7.

[waldner]

Turned out it was the V4 tube.  It's a newer JJ that didn't show signs of microphonics when tapped, but must have been the cause.  I changed to a different 12AT7 and can now turn the reverb up to 10. 

FYI, V4 should be a 7025 or 12AX7, not a 12AT7.

Good lookin' out Steve.  I didn't even realize it should be a 12AX7.  For some reason I thought it took a 12AT7 in that spot.  Will change tomorrow when back at my shop and test.


Thanks.

[tdvt]

TVDT,

Here's a good way to add the normal channel to the reverb from Rob Robinette: https://robrobinette.com/AB763_Modifications.htm#Normal_Channel_Reverb_Mod

Thanks for the heads-up. That was the one I had landed on a few nights ago & saved.

I will be interested to know how you like it.

[waldner]

Turned out it was the V4 tube.  It's a newer JJ that didn't show signs of microphonics when tapped, but must have been the cause.  I changed to a different 12AT7 and can now turn the reverb up to 10. 

FYI, V4 should be a 7025 or 12AX7, not a 12AT7.

Hmmm.  Ok, I guess my problem is not solved now that every 12AX7 I try results in squeal / feedback with the reverb turned up above about 6-7.  The only way I can get it to not do that is with a 12AT7.  I even tried a 5751 and though better, it still squeals at about 9-10.


What could cause good tubes that work in any other spot or amp to feedback or act microphonic in the 4V location on this amp?


Thanks

[sluckey]

It seems that a lot of people are having troubles with the reverb recovery tube (V4A) these days. This tube must deal with the lowest signal amplitude in the entire amp. Layout and wiring are critical.

Do you have enough reverb with the pot set just below the squeal threshold? If so, maybe removing the cathode bypass cap from V4 would help.

[tubeswell]

Can you post hi-res photos of the guts of your build?

[waldner]

Can you post hi-res photos of the guts of your build?

Yes, here are a few pics.  Thanks

[waldner]

It seems that a lot of people are having troubles with the reverb recovery tube (V4A) these days. This tube must deal with the lowest signal amplitude in the entire amp. Layout and wiring are critical.

Do you have enough reverb with the pot set just below the squeal threshold? If so, maybe removing the cathode bypass cap from V4 would help.

Yes, I think there is a enough verb just below that squeal threshold.  I believe the deluxe reverb stock build calls for a 25uF/25v cap bypassing the 820ohm V4 cathode resistor.  I have a .1uF in that spot now.  Could that be part of my problem?  What happens when you remove the bypass cap?  I'm guessing high frequencies diminish?


Thanks

[tubeswell]

There's usually a 220k grid leak resistor (for the reverb recovery grid) in this location. Why the cap?

[waldner]

There's usually a 220k grid leak resistor (for the reverb recovery grid) in this location. Why the cap?

That is a .002 uF cap paralleled with the grid leak resistor that I added because it’s said to help prevent feedback / oscillation.  It didn’t solve the problem though. 


Fender apparently used this cap on silver face reverb amps.

[pdf64]

… the deluxe reverb stock build calls for a 25uF/25v cap bypassing the 820ohm V4 cathode resistor.  I have a .1uF in that spot now.  …

That’s the problem, the shared cathodes need to be fully bypassed.
Because the common cathode V4 sections are arranged in cascade.
Any signal at the cathodes will create a positive feedback loop, as the reverb control is advanced, at some point the degree of positive feedback will be sufficient for free running oscillation to initiate.

An occasional cause of reverb oscillation cropping up on forums is the V4 shared cathodes not being properly bypassed. Usually it’s caused by the bypass cap being bad.
Using a dedicated cathode resistor for each V4 cathode will allow partial or unbypassed operation for either or both sections.

Your board photo seems to show the same cap type used for V4 cathode bypass and the adjacent V5 pin3 cathode bypass?
ie more like a 22uF electrolytic than 0.1uF cap?

[waldner]

… the deluxe reverb stock build calls for a 25uF/25v cap bypassing the 820ohm V4 cathode resistor.  I have a .1uF in that spot now.  …

That’s the problem, the shared cathodes need to be fully bypassed.
Because the common cathode V4 sections are arranged in cascade.
Any signal at the cathodes will create a positive feedback loop, as the reverb control is advanced, at some point the degree of positive feedback will be sufficient for free running oscillation to initiate.

An occasional cause of reverb oscillation cropping up on forums is the V4 shared cathodes not being properly bypassed. Usually it’s caused by the bypass cap being bad.
Using a dedicated cathode resistor for each V4 cathode will allow partial or unbypassed operation for either or both sections.

Your board photo seems to show the same cap type used for V4 cathode bypass and the adjacent V5 pin3 cathode bypass?
ie more like a 22uF electrolytic than 0.1uF cap?

Ok, very interesting.  What you are seeing there at the V4 and V5 cathode bypass locations are actually .1uF film caps.  Let me try to change some things there and see what happens.  Thanks PDF.

[waldner]

That solved the problem.  I replaced the .01uF bypass cap on the V4 cathode resistor with a 25uF/25V cap and the squeal is gone.  I can now turn the reverb and dwell up to 10 with no problem while running a 12AX7 in V4. 
Thanks PDF64!!! 


Now I've got one more amp that has a feedback/oscillation problem that I will post a separate thread about.  But this case is closed.

[pdf64]

… What you are seeing there at the V4 and V5 cathode bypass locations are actually .1uF film caps…

It’s also a bad idea to use partial cathode bypass on V5 pin3 the trem oscillator. It needs to be fully bypassed, so as to facilitate full gain down to the subsonic frequencies of the trem oscillator.

[waldner]

… What you are seeing there at the V4 and V5 cathode bypass locations are actually .1uF film caps…

It’s also a bad idea to use partial cathode bypass on V5 pin3 the trem oscillator. It needs to be fully bypassed, so as to facilitate full gain down to the subsonic frequencies of the trem oscillator.

Thanks for the tip.  I will change that cap as well.