Squealing/positive feedback - explanation of what is happening

[sleipnir]

Sorry for the delay in responding, but I took some time to basically gut the amp and wire it back up as close as possible to the 5F6-A design. Here is a photo of what it looks like now - I have more detailed photos if interested:



The voltages are at the extreme end of the ones found on the 5F6-A schematic (they are at the upper end of the +20% variance) - this is to be expected given that my PT is 380-0-380 and since I don't have the zener low voltage option at this point.

I decided to use the other OT I have from an old Super Reverb Reissue amp and chose to not mount it so I could play with its distance relative to the guitar and chassis. If you look at this video, you'll see that it seems to be the culprit. Am I correct in assuming that this is purely a magnetic field interaction between the OT and guitar pickups, or is there potentially something in the layout that is enabling this and that could be improved to reduce the squeal?

http://vid270.photobucket.com/albums...Squeal%203.mp4

Thanks!

[pdf64]

Maybe connecting the OT frame etc to chassis 0V would reduce / stop the squeal?

[sleipnir]

I experienced the squeal with the OT mounted to the chassis in the past (before I rewired the amp). I am assuming by chassis 0V you are referring to the chassis common (or ground), right? I could run an alligator clip from the OT frame to the chassis, but I don't think it would change. It's worth a try, nonetheless!

[Steve A.]

#1. I would not worry about the interaction between the OT and your guitar pickups, just that between the OT and the circuitry in the chassis which can't step away from the amp as a guitarist can.

#2. To lower the B+ on your amp are you adding the zeners between the CT and ground? I would wire them in now and then experiment with adding a switch to bypass them later if desired.

Good luck!

Steve Ahola

[sleipnir]

That argues for it being either a magnetic leakage path from the OT, or perhaps a shielding issue with the input. Did you try a different guitar cord on these iterations? Probably, but some guitarists use their favorite cord for everything. I had to check.

If this truly is a magnetic issue only and maintaining distance to the amp (OT) is the only solution, it certainly is not a huge deal. However, before I add all of the mods back in and essentially revert back to the way the amp was when I first built it, I would greatly appreciate it if everyone could take a look at one more video clip. I am quoting an earlier post of R.G.'s as I am curious if there is an electrical shielding issue or just magnetic interaction or both. Can anyone explain what is happening in this video? I don't know enough to conclude anything with certainty.

http://vid270.photobucket.com/albums...squeal%204.mp4

[Mike Sulzer]

If this truly is a magnetic issue only and maintaining distance to the amp (OT) is the only solution, it certainly is not a huge deal. However, before I add all of the mods back in and essentially revert back to the way the amp was when I first built it, I would greatly appreciate it if everyone could take a look at one more video clip. I am quoting an earlier post of R.G.'s as I am curious if there is an electrical shielding issue or just magnetic interaction or both. Can anyone explain what is happening in this video? I don't know enough to conclude anything with certainty.

http://vid270.photobucket.com/albums...squeal%204.mp4

I think electric fields are involved. Is it possible that it is purely electrostatic? Remember your test with the humbuckers? They should show a big reduction in the problem since they tend to cancel external magnetic fields, but did they show any reduction at all? Possibly the pickup resonance is involved, but not in a magnetic way. (The circuit impedance is highest at this resonance and provides a high loop gain to all sorts of things.)

Perhaps you need to devise an additional test before reassembly. (And in any case you always want to readdress layout and lead dressing issues because you always want to take advantage of an opportunity to do that.

[sleipnir]

Mike - I just checked again with my PRS humbucker equipped guitar and got the same results as in the most recent video I posted. I was not able to get any closer before the squeal started, and the frequency/pitch of the squeal seemed similar.

Any thoughts on what further testing I can do?

[R.G.]

I can also confirm that the problem has nothing to do with proximity to the speakers, but does indeed depend on how close I am to the chassis of the amp.

The more I think about this, the more it seems like acoustic feedback. I know you think you've confirmed that speaker proximity isn't it, but unless you run a specific test for it, I don't think acoustic is eliminated.

A suitable test might be with the speakers in another room. Or with speakers farther from the chassis than the "squeal threshold" and demonstrate that moving toward speakers stops the squeal, while moving toward the chassis starts it. Or putting resistor dummy loads on the outputs and watching the output with an oscilloscope, showing that it squeals at the same distance from the chassis with resistor loading that it did with speakers.

[Mike Sulzer]

Mike - I just checked again with my PRS humbucker equipped guitar and got the same results as in the most recent video I posted. I was not able to get any closer before the squeal started, and the frequency/pitch of the squeal seemed similar.

Any thoughts on what further testing I can do?

Not sure yet, but one important test you did two videos ago was to put an electrostatic shield over there transformer, stopping the squeal. That was an important clue!

One possibility would be to try this with a guitar that is very well electrically shielded. Maybe one with a preamp. I have found that it is easier to reduce hum from electric fields if the guitar has a preamp and thus a somewhat low output impedance. The same might work at squeal frequency.

[Mike Sulzer]

The more I think about this, the more it seems like acoustic feedback. I know you think you've confirmed that speaker proximity isn't it, but unless you run a specific test for it, I don't think acoustic is eliminated.

A suitable test might be with the speakers in another room. Or with speakers farther from the chassis than the "squeal threshold" and demonstrate that moving toward speakers stops the squeal, while moving toward the chassis starts it. Or putting resistor dummy loads on the outputs and watching the output with an oscilloscope, showing that it squeals at the same distance from the chassis with resistor loading that it did with speakers.

Try a restive load on the amp; that might settle that question. That squeal sounds too high in frequency to be acoustic feedback from a guitar speaker, but strange things do happen! (You can, of course get magnetic feedback involving the speaker under some conditions.)

[Steve A.]

I'm not hearing ANY squeal in your latest video, just hum. Perhaps that is because I don't hear much over 6khz.

I do think that you are extremely close to the amp and the OT, which is usually mounted on the same plane as the PT with the laminations at right angles.

On some of my amps I need to be 3 feet back to avoid the gremlins. Can you move your rig to say a kitchen or dining room table where you could step back a few feet? I would also orient chassis so that knobs were up...

Have you tried rotating the guitar in all dimensions which will often reduce hum and/or noise?

How are your controls set? We could get an idea of that if you were to pluck a string.

Have you added a pF capacitor to the grid of V1? Are you using a 12AX7 or 12AY7 for V1? What is the B+ voltage?

Do you have presence control grounded to tone stack? Not a good idea...

Can you adjust treble and other controls during video so that we can see if they make a difference? Maybe call out the settings as you go...

Do you have a guitar with humbuckers? Unshielded strats can squeal like a stuck pig.

Just some thoughts...

Steve Ahola

[sleipnir]

The speakers I have been using are behind me so I am actually moving farther away from them when getting closer to the OT and getting the squeal, so I don't think this is a speaker issue. You can see the the edge of speaker cabinet (my Super Reverb) in the doorway - this is as far away as I can get it with my current speaker wire.



I took R.G. and Mike's suggestion and hooked up a dummy load and a scope - you can see in this video that I get a squeal in the 10kHz range - and I can hear it coming from the amp somewhere, but I cannot tell if it's the OT, tubes, capacitors or resistors that are squealing.
http://vid270.photobucket.com/albums...squeal%205.mp4

I then hooked the speakers back up and measured a similar, but slightly higher frequency:
http://vid270.photobucket.com/albums...squeal%206.mp4

Finally, although I don't have an active guitar, I have an active bass and gave it a shot - you'll see that with the active electronics the squeal is 14-15kHz but then when I disengage the active electronics, it drops to 10kHz or so.
http://vid270.photobucket.com/albums...squeal%207.mp4

Does this offer any further clues to you guys?

[sleipnir]

Steve,

The squeal is above 6kHz so you might not hear it in the video - I definitely do, though. I tried the amp today with a fuzz pedal and had to be at least three feet away in order to prevent the squeal. 3 feet away is not a huge deal in practical terms, but given that I took the amp apart and still need to reassemble all the mods, I do want to see if there is anything not quite right with the way my layout and wiring is at this time so I can fix it before reassembling. I cannot remember if I recorded this in a video or not, but it does matter how the guitar is oriented with respect to the OT - at some angles I get a very nasty and much louder squeal. I did rest the chassis and OT in one of the previous videos in such a manner as to approximate how they would be situated relative to me holding the guitar, and that same 10kHz squeal is present when close enough.

Volume, Treble, Middle and Presence are all maxed out (I mentioned this early on, but it is probably good to repeat that as it is important) - if I lower any of these, I have to move closer to get the squeal. With presence down very low, I have to get extremely close if my memory serves me right. So I realize this might be an extreme case, but I kind of like the sound with everything except the bass maxed out, so they are settings I would use.

I did try caps across the plates, but I didn't have anything smaller than 47pF. I will order some lower value 1kV ones soon as I need to get a few replacement parts anyway.

The presence control is indeed grounded at the preamp. At one point, I moved it to the power ground but it did not help. I will perhaps try it again now with the simpler layout as I believe tried it prior to removing a lot of the extra wiring and switches.

Humbuckers do not seem to improve the situation.

By the way, in response to something you asked previously, I did have zener diodes between the PT CT and ground as a switchable mod, and I had added some additional bias pots to allow me to have proper bias when switching between low and high voltage mod. Something appears to have happened where those zeners are now shorted. Before I disassembled the amp, I had tried putting the metal box over the OT while it was mounted to the chassis. At that time, I saw some flashing inside the rectifier tube and the fuse eventually blew. I have no idea whether something related to this metal box could have caused this or if it was just coincidence. I will be adding the zeners back as a switchable mod, but now I am wondering if that approach is not reliable. I guess it is possible that the added current flowing as I am inducing the squeal could have damaged the zeners, but I honestly don't know.

Finally, I am using a 12AY7 in V1 and my B+ is around 480V or so, I believe. When I was using the zeners and still getting the squeal, I was at 430V or so B+, so it seems to be happening either way.

[Steve A.]

Volume, Treble, Middle and Presence are all maxed out (I mentioned this early on, but it is probably good to repeat that as it is important) - if I lower any of these, I have to move closer to get the squeal. With presence down very low, I have to get extremely close if my memory serves me right. So I realize this might be an extreme case, but I kind of like the sound with everything except the bass maxed out, so they are settings I would use.

When I first got into modding my guitar amps I thought of them being like test equipment which must be set carefully... turn everything up to 10 and it would scream like a mofo. I figured that production line amps had been neutered so that there were no settings that would go apeshit crazy... but they didn't have that raw sound I was looking for.

In any case for initial testing I will set the tone stack and presence controls to 5 and then adjust them from there looking for anomalies. And set the volume control at a good playing level. Can you do that and film it for us?

I did try caps across the plates, but I didn't have anything smaller than 47pF. I will order some lower value 1kV ones soon as I need to get a few replacement parts anyway.

I didn't mean caps across the plates (yuck!) but from the grid to ground on the tube socket. You can solder two 47pF caps in series to get 23.5pF. I would start with 150pF (three 47pF caps in parallel) to see if that helps and then work my way down towards 22pF. The difference in tone is very subtle but can make an unstable amp stable.

By the way, in response to something you asked previously, I did have zener diodes between the PT CT and ground as a switchable mod, and I had added some additional bias pots to allow me to have proper bias when switching between low and high voltage mod.

I would hardwire the zeners in to bring the B+ to 430vdc and forget about the higher voltage and bias switching. But that is me... FWIW I do like to screw around with switches to change the plate voltage on the preamp and PI tubes but not on the power tubes.

So where are you going to play your amp when it is completed? Can you move the chassis and speakers there to give it a real world test? Your work cubby looks very crowded...

Steve Ahola

[sleipnir]

I didn't mean caps across the plates (yuck!) but from the grid to ground on the tube socket. You can solder two 47pF caps in series to get 23.5pF. I would start with 150pF (three 47pF caps in parallel) to see if that helps and then work my way down towards 22pF. The difference in tone is very subtle but can make an unstable amp stable.

Steve Ahola

Yikes, I guess I was not thinking about what I was writing....I meant that I had tried a 47pF cap across the plate to grid, and I had also tried caps across the plate resistors but needed such large values there to get rid of the squeal that the tone was really muffled. I guess I feel like I shouldn't HAVE to put a cap anywhere since the original design didn't require it, and it seems like more of a band aid than a fix, but then again, I don't know if the designers back in the 50's were testing it with all controls maxed and standing right next to the amp :-)

That being said, I will see if I have a couple of 47pF caps to try in parallel and see if that sounds ok.

I was playing this amp in a bigger room since I finished it back in November. I just got around to looking into the squeal in February, though, which is why the chassis is now sitting on my messy work desk. The room is big enough that I can stand more than 3 feet from the amp if I am using fuzz pedals and such, but I would of course still prefer it not to squeal if I do get closer.