1940's Amp High Volume Oscillation

[bobloblaws]

Which tube is microphonic, is it the 6N7? Is it metal? I sometimes have a hard time with microphonics from metal tubes.

No, it's the 6SJ7, but it is metal as well. But I'm convinced that's not my issue. Maybe this will shed some light. As I said, with nothing plugged in there is no squeal (I'm not sure what happens if I only have a cable plugged in and no guitar connected, the hum is so loud that I didn't want to turn it up to the point where it begins to whistle. I should try scoping it). Now, with the guitar plugged in, as I turn down the volume knob on the guitar the frequency of the oscillation increases. Does that tell us anything?

[Enzo]

Yes, that the guitar is in the feedback loop.

[bobloblaws]

Yes, that the guitar is in the feedback loop.

I meant does it give us a clue as to what the nature of the problem is? (or more accurately, what might be done about it?)

[Enzo]

Yes, that we have a feedback loop somewhere, and that it includes the guitar and input stage. I wasn't being flippant.

My tech opinion is that increasing the caps hid the problem rather than curing it. After all the original circuit worked with those 4uf and 8uf caps.

[bobloblaws]

Yes, that we have a feedback loop somewhere, and that it includes the guitar and input stage. I wasn't being flippant.

My tech opinion is that increasing the caps hid the problem rather than curing it. After all the original circuit worked with those 4uf and 8uf caps.

I'm not following. The circuit worked with the 4uF and 8uF caps, but it exhibited both the motorboat problem and the whistling problem before I ever touched any of the filter caps. When was either problem ever hidden?

The only thing that is different is the motorboating problem is cleared up.

Edit: FWIW, I scoped the output and the sound I'm describing looks to be in the neighborhood of 10kHz (surprising that I can even hear it :-))

[Enzo]

The circuit worked with the 4uF and 8uF caps, but it exhibited both the motorboat problem and the whistling problem before I ever touched any of the filter caps. When was either problem ever hidden?

I am assuming this amp has not acted this way since the day it was made. It may have had those problem ever since you first saw it, but we assume the circuit used to work in 1947. When it was made it didn't have that problem. That is what I mean that the smaller caps should be sufficient. Sticking larger caps in it now basically makes the symptom stop by brute force, not because the amp always needed larger caps.

Having said that, I have no problem with larger caps, just don't look at them as the solution to your issue.

I could be way off, but I see something like a faulty ground connection in one stage might be covered up with the larger caps.

And I think the biggest clue is that a guitar plugged in and turned to zero makes it stop.

Got a 40 foot cord? Plug the guitar in with that and place it across the room as far as possible. Or for that matter a wireless unit.

[bobloblaws]

I am assuming this amp has not acted this way since the day it was made. It may have had those problem ever since you first saw it, but we assume the circuit used to work in 1947. When it was made it didn't have that problem. That is what I mean that the smaller caps should be sufficient. Sticking larger caps in it now basically makes the symptom stop by brute force, not because the amp always needed larger caps.
Having said that, I have no problem with larger caps, just don't look at them as the solution to your issue.

Hmmm, but in fact I do see them as a solution to one of the two problems in play, that of the low frequency oscillation. You yourself said earlier:

"Caps were expensive back then, so they used 4uf instead of 10uf or 20uf. SO I have to agree with helm, change your filter caps, you are probably suffering from lack of decoupling." The message from helm was:

"As you describe the oszillation as medium speed tremolo effect, it would be a kind of motorboating. This is often caused by a positive feedback between stages via unsufficiently decoupled/filtered supply voltages, in other words too small filter cap C-values. 4µ and 8µ are already extremely low and electrolytics lose capacitance over time. You should try new filter caps and if necessary increase the values moderately."

I could be way off, but I see something like a faulty ground connection in one stage might be covered up with the larger caps.

Other than the motorboating problem being solved as was anticipated, what is different about the scenario now as opposed to before the 20uF caps installed? If the larger caps are covering up some symptom, it follows that that symptom would be apparent prior to installation of the larger caps. What was the symptom that you think might now be covered up?

Got a 40 foot cord? Plug the guitar in with that and place it across the room as far as possible. Or for that matter a wireless unit.

I have a 30 foot cord and I can extend it with a coupler as well. I also have a wireless if needed.

[bobloblaws]

Got a 40 foot cord? Plug the guitar in with that and place it across the room as far as possible. Or for that matter a wireless unit.

So I assumed you wanted to know if moving the guitar far away from the amp reduced or eliminated the squealing. Going wireless and with the same 10 ft cable I've been using from the receiver to the amp input eliminated the squealing, even standing right in front of the amp, so distance played no part there. Then I tried different length cables (straight in, no wireless) and found that the length of the cable determined whether there was any audible squeal at all or whether it was earlier onset etc. So with the original 10 footer, moving away from the amp or side to side made no difference but a 20 foot cable eliminated the squealing even right in front. And with a 5 foot cable the squealing came on earlier, at a lower amp volume level. These experiments were done with the tone pot on 10, reducing the tone pot to 5 for example would eliminate the squeal as well.

[Leo_Gnardo]

So I assumed you wanted to know if moving the guitar far away from the amp reduced or eliminated the squealing. Going wireless and with the same 10 ft cable I've been using from the receiver to the amp input eliminated the squealing, even standing right in front of the amp, so distance played no part there. Then I tried different length cables (straight in, no wireless) and found that the length of the cable determined whether there was any audible squeal at all or whether it was earlier onset etc. So with the original 10 footer, moving away from the amp or side to side made no difference but a 20 foot cable eliminated the squealing even right in front. And with a 5 foot cable the squealing came on earlier, at a lower amp volume level. These experiments were done with the tone pot on 10, reducing the tone pot to 5 for example would eliminate the squeal as well.

It's looking like the squeal noise is a feedback between guitar pickups and speaker. Longer cables have higher capacitance, reducing high frequency signal coming from the guitar similar to the way a tone control does. If there's a tone control on the guitar, I would expect the squeal goes away as you dial the tone control down.

Metal covers over the pickups? If yes, try this: get the guitar/amp/speaker rig to start gently squealing, then gently press a finger onto the pickup cover, see if the squeal stops. If that is the case, nothing unusual here, you can probably fix the problem at the pickup with a wax dip, with the cover on the pickup. You want to get wax to fill the space between the pickup and cover, as well as have it soak into the pickup coils, to eliminate their ability to move. Then the pickup will stop acting as a microphone at high frequencies.

[bobloblaws]

It's looking like the squeal noise is a feedback between guitar pickups and speaker. Longer cables have higher capacitance, reducing high frequency signal coming from the guitar similar to the way a tone control does. If there's a tone control on the guitar, I would expect the squeal goes away as you dial the tone control down.

Metal covers over the pickups? If yes, try this: get the guitar/amp/speaker rig to start gently squealing, then gently press a finger onto the pickup cover, see if the squeal stops. If that is the case, nothing unusual here, you can probably fix the problem at the pickup with a wax dip, with the cover on the pickup. You want to get wax to fill the space between the pickup and cover, as well as have it soak into the pickup coils, to eliminate their ability to move. Then the pickup will stop acting as a microphone at high frequencies.

I may be wrong but I really don't think it is feedback between pickup and speaker, as I stated earlier. If that was the case I would expect that moving away from the speaker, changing the angle of the pickup in relation to the speaker, etc. would have an effect, and that is not the case. You are probably correct re: the tone control, but as I said I observed that when turning down the tone control on the amp, I didn't happen to try it on the guitar. I also mentioned that as the guitar volume knob approaches zero the frequency of the oscillation/feedback increases. I was able to observe that phenomenon on my oscilloscope as well with the output connected to a dummy load, so I am further convinced the speaker is not part of the equation.

[Enzo]

Yet the pickup does seem to be involved.

DO you have other instruments you could plug into the amp that do not have wound pickups? A synth or even a stringed acoustic with a piezo pickup?

[bobloblaws]

Yet the pickup does seem to be involved.

DO you have other instruments you could plug into the amp that do not have wound pickups? A synth or even a stringed acoustic with a piezo pickup?

So I tried an acoustic/piezo with dummy load and there was no observable oscillation. Plugging the Les Paul back in I observed that the (roughly) 10KhZ sine wave's frequency changes in response to adjusting volume control, adjusting tone control (turning down tone control on the amp does not affect frequency, only eliminates the oscillation at some point), moving the guitar/changing the angle closer to and further away from the amp and/or the dummy load. Also, when any of these variables line up to produce a certain frequency I'm actually hearing it coming from the amp, even though there is no speaker connected, albeit at a very slight volume. When I unplugged the guitar but left the cable plugged in I also heard an oscillation, this one shows up as (roughly) 1.5KhZ square wave.

[Enzo]

OK, so the problem requires that a wound pickup be plugged in, correct?

We all know that pickups can acoustically fed back with a scream at high levels. But in super gain circuits, they can feed back at low levels too. I mean levels that do not interrupt conversations.

If you run the amp into a dummy, and crank a signal through, it is normal to hear the OT "sing" along with the music. We just discussed this a day or so ago.

[bobloblaws]

OK, so the problem requires that a wound pickup be plugged in, correct?

It would seem so.

We all know that pickups can acoustically fed back with a scream at high levels.

True, but I always thought this was a feedback loop between the pickup and speaker. I would not expect the same thing with a dummy load.

If you run the amp into a dummy, and crank a signal through, it is normal to hear the OT "sing" along with the music.

Except that in my case I am not running a signal into the amp, just connecting the instrument.


In any case, I'm prepared to leave well enough alone at this point. Chances are this amp will not be typically run near max volume anyway. And other conditions need to be present as well such as shorter cables, tone control up etc.

[Enzo]

The instrument is a signal too.