Squealing/positive feedback - explanation of what is happening

[sleipnir]

So, I was thinking about this some more after having confirmed in the videos above that the oscillation happens with speakers and with a dummy load, regardless of using single coils or humbuckers, etc., again, as long as the volume of the guitar, volume of the amp as well presence and treble are maxed or pretty high. At the end of the day, since I get oscillation without playing the strings, there has to be some signal present at the output which then gets picked up by the guitar, fed back to the amp, and then starts the oscillation. The only signal present would have to be noise, right? In looking at the output of the amp with an oscilloscope, with the guitar volume down all the way and with all amp controls down all the way except the volume, there is not much evidence of noise. If I max the middle pot, I see some noise. If I lower it again and max the presence, I don't see any noise. If I lower the presence and max the treble, there is significantly more noise compared to when the middle pot was maxed (something like 12-20mV at the speaker jack), but if I leave the treble maxed and max the presence, the noise is significantly higher (at 35-50mV or so) than with just the treble. Would it be a logical conclusion that the noise present at the output with the volume, treble and presence controls up high, is sufficiently high so that when I bring the volume up on the guitar and get close to the OT or even the speaker jack where it connects to the dummy load, there is enough strength in this noise signal to get picked up by the guitar pickups? Further, since a pickup generates a current by having its magnetic field "disturbed", does that mean that it is therefore electromagnetic radiation from the OT or speaker jack (caused by the noise) that is causing the pickup to send a signal down the guitar chord back to the amp?

If these are valid conclusions, then is the solution perhaps to try and lower the noise in the amp (if at all possible)? I already have metal film resistors for the input jack 1M and 68k resistors, but everything else is carbon comp. I don't know how much the contribute vs. capacitors and the tubes themselves.

Does any of this make sense?

[Steve A.]

For some reason this thread brings to mind the old Henny Youngman joke...



I really can't imagine a guitarist setting a 5F6A with the volume, treble and presence controls dimed. The 5F6A was not designed with that amount of gain in mind. Just because the knobs can be moved to those settings doesn't mean that they should be.

Steve Ahola

[nickb]

So, I was thinking about this some more after having confirmed in the videos above that the oscillation happens with speakers and with a dummy load, regardless of using single coils or humbuckers, etc., again, as long as the volume of the guitar, volume of the amp as well presence and treble are maxed or pretty high. At the end of the day, since I get oscillation without playing the strings, there has to be some signal present at the output which then gets picked up by the guitar, fed back to the amp, and then starts the oscillation. The only signal present would have to be noise, right? In looking at the output of the amp with an oscilloscope, with the guitar volume down all the way and with all amp controls down all the way except the volume, there is not much evidence of noise. If I max the middle pot, I see some noise. If I lower it again and max the presence, I don't see any noise. If I lower the presence and max the treble, there is significantly more noise compared to when the middle pot was maxed (something like 12-20mV at the speaker jack), but if I leave the treble maxed and max the presence, the noise is significantly higher (at 35-50mV or so) than with just the treble. Would it be a logical conclusion that the noise present at the output with the volume, treble and presence controls up high, is sufficiently high so that when I bring the volume up on the guitar and get close to the OT or even the speaker jack where it connects to the dummy load, there is enough strength in this noise signal to get picked up by the guitar pickups? Further, since a pickup generates a current by having its magnetic field "disturbed", does that mean that it is therefore electromagnetic radiation from the OT or speaker jack (caused by the noise) that is causing the pickup to send a signal down the guitar chord back to the amp?

If these are valid conclusions, then is the solution perhaps to try and lower the noise in the amp (if at all possible)? I already have metal film resistors for the input jack 1M and 68k resistors, but everything else is carbon comp. I don't know how much the contribute vs. capacitors and the tubes themselves.

Does any of this make sense?

To some extent you are right - noise of some kind is needed to start the oscillation. The problem is you have positive feedback meaning that even the tiniest noise is enough to get the ball rolling. So, lowering the noise floor will not help.

There are many production amps (JCM800 4010 springs to mind) that squeal when you crank everything up, so you're not in bad company

Given that it seems you have eliminated magnetic and acoustic paths that we get back to electric. I see the chassis is open. Have you tried covering the opening with something metallic as a shield, making sure the shield is in contact with the chassis? I keep a sheet of cardboard covered with aluminum foil ( the sticky type) handy for such purposes.

[R.G.]

Would it be a logical conclusion that the noise present at the output with the volume, treble and presence controls up high, is sufficiently high so that when I bring the volume up on the guitar and get close to the OT or even the speaker jack where it connects to the dummy load, there is enough strength in this noise signal to get picked up by the guitar pickups?

What they said.

But it would be an interesting test to take a 100K resistor, wrap its leads tightly around the tip and sleeve of a guitar cord, then plug the non-resistor end into the amp and try it. If the amp oscillates with this input and it's not specific to one cord, then the problem is inside the amp.

Since this is a home-built, what does the grounding setup look like? It is possible for ground impedance to cause oscillation if the forward gain of the amp is high enough. Tone and volume controls work by lowering the forward gain of the amp.

[Steve A.]

From the Tube Amp Debug Page at R.G.'s GEOFEX site (as suggested previously)...









Steve Ahola

[R.G.]

Oh, sure, throw that back in my face.

[Justin Thomas]

Hey, that geofex thingy STILL saves my butt on occasion! Thank you, R.G.!

Justin

[Chuck H]

I don't think it's a positive feedback "problem" per se. Otherwise you could flip your guitar over and render it negative feedback. Which would stabilize the amp more. Does that happen?

I know you've mentioned that the problem started with amp assembled and you now have it on the bench in an effort to trouble shoot it. Makes sense. But you may be hyper sensitive looking for trouble at this point. I know it's a pita putting it in and out of the cabinet, etc. But to be honest if any Marshall or 5f6a Bassman were sitting open chassis on my bench with the treble, volume and presence cranked up I would be surprised if I could get open face pickups within three feet of it without SOMETHING happening. Maybe try putting it back into the cabinet, speakers and all and see if it operates normally. I assume your cabinet has a shield for the open end of the chassis?

Also, I know you disabled the switches, but did you make sure any OT secondary leads from the "speaker in" circuit are no longer running under the board, near the preamp or otherwise untended and in a bad place under the hood??? It doesn't even matter if their not connected. No OT leads should be anywhere near the preamp. Including the presence lead from the feedback circuit.

[Mike Sulzer]

I don't think it's a positive feedback "problem" per se. Otherwise you could flip your guitar over and render it negative feedback. Which would stabilize the amp more. Does that happen?

It might work that way for purely magnetic feedback, but I would not be so sure if it is electric. Metal modifies electric fields. You have a guitar cable and various metal parts in the guitar. Turning it could have complicated effects.

But anyway, if you move the guitar closer and the oscillation starts what other realistic possibilities (besides feedback) are there?

[g1]

Perhaps a problem would more likely be an amp that did not feedback under said conditions.

[sleipnir]

But it would be an interesting test to take a 100K resistor, wrap its leads tightly around the tip and sleeve of a guitar cord, then plug the non-resistor end into the amp and try it. If the amp oscillates with this input and it's not specific to one cord, then the problem is inside the amp.

I gave this a shot. I tried wrapping a 100k resistor across the tip and sleeve on one cable and also tried soldering a 100k resistor on another just to ensure a good connection. I also tried a cable with no resistor. I recorded a video:
http://vid270.photobucket.com/albums...ueal%208_1.mp4

I figured I would also give it a shot with a metal cover/shield over the chassis while also moving the OT and its wires to the other side of the chassis away from the power tubes.
http://vid270.photobucket.com/albums...squeal%209.mp4

Since this is a home-built, what does the grounding setup look like? It is possible for ground impedance to cause oscillation if the forward gain of the amp is high enough. Tone and volume controls work by lowering the forward gain of the amp.

In addition to a separate safety ground, I have a power ground, preamp ground and the speaker jack ground. In the photo below, you will see six wires connecting at the power ground. Two are from the output tubes, one from the PT center tap, one from the filament supply artificial CT, one from the bias circuit, and one from the filter caps - specifically the first two sections which connect to either side of the choke:


The pots and input jacks are isolated from the chassis with the exception of one input jack which serves as the single grounding point for the preamp section. There is a wire that runs along the back of the pots which is connected to the one grounded input jack. The various grounds from the preamp circuit are grounded here, as are the remaining two filter cap sections. I currently have the presence pot grounded here as well, although I have tried grounding it to the power ground and speaker jack ground with no improvement. I did try grounding each pot to the chassis but that did not make a difference. I tried grounding ALL the filter caps at the power ground, but that did not change anything. I also tried grounding the speaker jack to both the preamp ground and power ground, but it did not change anything

[nickb]

Have I got this right? You are putting the input right next to the output transformer and are surprised it's oscillating? That's what I would expect it to do under such circumstances - you are doing exactly the thing you should not do i.e. coupling the biggest signal swing output back to the tiny signal level input.

First get the OPT back in the chassis with neat short wiring well away from anything sensitive and try again. What you are aiming for is no tendency to oscillate under reasonable conditions - you'll have to define what you think is reasonable for yourself. But hopefully we can agree that what you are doing isn't reasonable.

[Steve A.]

Oh, sure, throw that back in my face.

I think that half of the boutique amp builders today cut their teeth on your Tube Amp Debug Page...

Steve Ahola

P.S. I really liked your advice on GeoFex to not damage a $100 printed circuit board trying to desolder a 5 cent device. Just cut the damn cap or resistor out and save the board!

[Chuck H]

Have I got this right? You are putting the input right next to the output transformer and are surprised it's oscillating? That's what I would expect it to do under such circumstances - you are doing exactly the thing you should not do i.e. coupling the biggest signal swing output back to the tiny signal level input.

First get the OPT back in the chassis with neat short wiring well away from anything sensitive and try again. What you are aiming for is no tendency to oscillate under reasonable conditions - you'll have to define what you think is reasonable for yourself. But hopefully we can agree that what you are doing isn't reasonable.

If you read the thread he was having some real problems with the amp assembled before. Maybe just spooky now to accept something less than ideal, which most things never are. But I might do something like that too. Looking at it from the outside is a lot easier. That's why I suggested the real world test with a reassembled amp.

[nickb]

If you read the thread he was having some real problems with the amp assembled before. Maybe just spooky now to accept something less than ideal, which most things never are. But I might do something like that too. Looking at it from the outside is a lot easier. That's why I suggested the real world test with a reassembled amp.

Ah! The thread has got longer than my ability to retain the info . Thx.

Still the fact remains, putting the input lead next to the OPT leads is asking for problems.