Parasitic oscillation problem untraceable

[TriodeLuvr]

All the higher amplitude of one side does is make the height of the tops larger than the depths of the troughs (or vice versa). It merely attenuates one side of the fundamental first order harmonic.

I haven't read Merlin (Valve Wizard?) on this topic. I'm too newbie for that.

Anyway, when the "top half" of the sine is larger or smaller than the "bottom half," the composite waveform is no longer a pure sine wave. This can be verified empirically with a spectrum analyzer and an unbalanced phase inverter or (more accurately) an arbitrary waveform generator. It can also be confirmed with SPICE simulators and mathematically with Mathcad.

I no longer have access to Mathcad, but I can perform SPICE simulations of balanced VS unbalanced phase inverters and display the resulting FFT plots. I'm pretty sure this will confirm my statement that asymmetry creates even-order products.

Jack

[TriodeLuvr]

I figured the 24V zeners I had should do the job. But to no avail...

24V is probably too much bias (too cold).

Sorry if I've posted this link before. It describes another way to fix-bias the voltage at the cathodes, and it's adjustable. No relation to me or any of my projects.

http://tronola.com/A_New_Look_At_An_Old_Friend.pdf

[crgfrench]

...when the "top half" of the sine is larger or smaller than the "bottom half," the composite waveform is no longer a pure sine wave.

If it's coming from a stratocaster it's not a pure sine wave anyway. But let's work with a signal generator. You seem to believe that an "impure" sine wave implies there are higher order harmonics at play. That is not necessarily true.

[TriodeLuvr]

...when the "top half" of the sine is larger or smaller than the "bottom half," the composite waveform is no longer a pure sine wave.

If it's coming from a stratocaster it's not a pure sine wave anyway. But let's work with a signal generator. You seem to believe that an "impure" sine wave implies there are higher order harmonics at play. That is not necessarily true.

It is true in the context of our discussion involving the LTP.

Jack

[Bieworm]

I figured the 24V zeners I had should do the job. But to no avail...

24V is probably too much bias (too cold).

Sorry if I've posted this link before. It describes another way to fix-bias the voltage at the cathodes, and it's adjustable. No relation to me or any of my projects.

http://tronola.com/A_New_Look_At_An_Old_Friend.pdf

I thought the PR mod zener value was determined by taking the cathode voltage and add a few volts. So the 21V cathode voltage + a few volts would mean that my 24V zeners should be ideal? Or am I missing something in this situation?
It would be a lot easier to determine the cause of the fizz when I have my siglent 1102 to rule out the possibility that the fizz winds up to be a parasitic oscillation earlier in the circuit and gets amplified anyway. gosh!!! Or is the voltage swing on the cathode of them 6V6's a dead giveaway it IS crossover distortion?

[TriodeLuvr]

I thought the PR mod zener value was determined by taking the cathode voltage and add a few volts. So the 21V cathode voltage + a few volts would mean that my 24V zeners should be ideal? Or am I missing something in this situation?
It would be a lot easier to determine the cause of the fizz when I have my siglent 1102 to rule out the possibility that the fizz winds up to be a parasitic oscillation earlier in the circuit and gets amplified anyway. gosh!!! Or is the voltage swing on the cathode of them 6V6's a dead giveaway it IS crossover distortion?

Sorry, I misunderstood what you were doing. I assumed this mod consisted of a Zener in the cathode circuit. Now I see that the Paul Ruby mod is attached to the grids. It's entirely different than what I was thinking about.



This circuit is designed to attack the problem of blocking distortion. Blocking distortion occurs when the grids are pulled negative due to grid current. This problem doesn't cause cathode current to increase. That's what you reported here, a large increase in cathode current, evidenced by the voltage across the cathode bias network swinging to 30V. Based on that, you're experiencing the effects of crossover distortion, but it can't be cured by limiting the negative-going grid voltage.

What you could try is to stabilize cathode voltage directly with a Zener. They can be noisy, so it might not be a permanent solution, but it will serve as a test. 24V is probably too high; you need a Zener value of 18V or 20V, with the Zener just tacked in across the existing cathode resistor and capacitor.

Jack

[JMPGuitars]

The PR mod is a good option to try,

That's exactly what I said - it's necessary to stabilize the bias either at the cathode or grid. Installing Zeners would stabilize the bias at the cathode.

The PR mod is by the grid reference resistors, not the cathode.

The zener across the cathode resistor method is certainly another option. I'm not disagreeing with you. As I said, Goo goo g'joob.

[Bieworm]

Is the zener across the cathode resistor on the 6V6's compatible with the PR mod? Then I could try that method with the PR mod still installed.
Can I put some zeners in series to find the magic value? I have 3 pcs 1N5341 zeners ( 6.2V 5W) in my stash. That's 18.6V... Orientation of zeners cathode to ground? I found a schematic online where the zener cathode is tied to the cathode side of the output tubes..

http://ppamps.blogspot.com/2013/

here's a very comparable situation from somebody who's had the same problem:

Sunday, October 13, 2013
Bodie with zener-assisted cathode bias
I finally got around to trying the zener trick I mentioned a few months ago. Bodie's idle bias voltage was around 21V, and that surged up to 31V under heavy overdrive. A pair of 12V 5W zener diodes in series now clamp the cathode voltage at 24V. Maximum clean output power has risen from 19W to 23W. On the 'scope, crossover distortion is now only barely present during overdrive and the overdriven tone has increased rather dramatically.

maybe I should try the 24V zener first? he did clamp the voltage at 24V in a similar situation...
what's the noisiest?a series of lower voltage zeners or 1 larger voltage zener?

[JMPGuitars]

Is the zener across the cathode resistor on the 6V6's compatible with the PR mod? Then I could try that method with the PR mod still installed.
Can I put some zeners in series to find the magic value? I have 3 pcs 1N5341 zeners ( 6.2V 5W) in my stash. That's 18.6V... Orientation of zeners cathode to ground? I found a schematic online where the zener cathode is tied to the cathode side of the output tubes..

maybe I should try the 24V zener first? he did clamp the voltage at 24V in a similar situation...
what's the noisiest?a series of lower voltage zeners or 1 larger voltage zener?

Series Zeners are fine. Experiment with the value, and see if it works, and how it affects the sound.

A conjunctive filter (like the snubber circuit) might also work, but experiment and see what happens.

[Bieworm]

What's the wattage requirement? I also have 1.3W in lower voltages

[Bieworm]

Ok that's not the problem apparently. Nor 24, 18 or 15V changes anything. I can see my DVM read the values and they stay proper un the zener range.. no change in fizz.
Should I put the 24V 5W zener in anyway for the max bias?

[crgfrench]

It is true in the context of our discussion involving the LTP.

I would love to learn why and how.

[TriodeLuvr]

Ok that's not the problem apparently. Nor 24, 18 or 15V changes anything. I can see my DVM read the values and they stay proper un the zener range.. no change in fizz.
Should I put the 24V 5W zener in anyway for the max bias?

Well, that's disappointing. There must have been some change in the sound, yes? Anyway, I would leave the 18V cathode Zener in place until this is sorted. Stabilizing the cathode voltage can only be a good thing. About the wattage, the Zener is sharing current with the cathode resistor, so 1W is probably plenty. The exact current through the Zener can be calculated by subtracting resistor current (E/R) from the total tube current, or just insert an ammeter in series with the Zener and make a measurement at full output. Be sure to leave the bypass cap in place from the cathode to ground if you measure it like this.

At this point, I think maybe you'll need a scope to identify the problem. It's like I said before, this issue of bias instability might not be the cause of what you're hearing. I'm really sorry this work wasn't more productive for you. Maybe someone else has an idea of something to try.

Jack

[TriodeLuvr]

It is true in the context of our discussion involving the LTP.

I would love to learn why and how.

We were discussing a single frequency sinewave (per your illustration). Any distortion through a circuit that does not introduce an additional signal source at a different frequency can only consist of harmonic energy.

This does change when we move to an actual guitar waveform, because it consists of multiple frequencies. That produces non-harmonically-related intermodulation distortion when the signal encounters non-linearities in the circuit.

Jack

[TriodeLuvr]

Didn't see this earlier...

Can I put some zeners in series to find the magic value? I have 3 pcs 1N5341 zeners ( 6.2V 5W) in my stash. That's 18.6V...

Yes, and that's a good value.

Orientation of zeners cathode to ground? I found a schematic online where the zener cathode is tied to the cathode side of the output tubes..

Yes, the Zener cathode should be connected to the cathodes of the tubes.

[Bieworm]

At grid pin of third preamp stage OD channel
Vol 6
Tone 5
Midbite full on
Signal 1kHz

20210105_204247.jpg

At grid pin of second preamp stage clean channel
Vol full on
Treble 7
Bass 3.5
Signal 1kHz

20210105_204756.jpg

At grid pin second preamp stage OD channel
Vol 6
Treble 5
Midbite full on
Signal 1kHz

20210105_205350.jpg

[JMPGuitars]

Ok that's not the problem apparently. Nor 24, 18 or 15V changes anything. I can see my DVM read the values and they stay proper un the zener range.. no change in fizz.
Should I put the 24V 5W zener in anyway for the max bias?

It was worth a shot, but I'm not surprised. You can leave the 24V 5W Zener in if you want, but I don't know that it will do anything for you.

I think the PR mod or the snubber would more likely be helpful.

What are all your voltages now? Can you find documentation with target voltages?

Another possible cause of this is if your signal is too hot going from one stage to the next. Or if you have too much low end getting amplified in the signal path.

I just looked a little more in depth at your schematic. It's not 2 separate channels so much as it looks like switching on the relay cascades V1A into V2B.

If you already have this issue before switching on the higher gain, then you might be able to start trying a couple things earlier in the circuit.

Did you say the circuit has a lot of bottom end? Maybe C9 should be experimented with. Try dropping that to 1µF (or .68, or 2, whatever, something lower).

If that doesn't help, it would be useful to figure out roughly where the issue begins when the high gain switch is off. It's possible you may need to add some voltage dividers on the signal path to reduce the levels.

Target voltages would certainly help a lot. If your preamp is too hot, that could easily cause this issue.

Thanks,
Josh

[TriodeLuvr]

I thought you didn't have a scope? What does it look like at the output (speaker)?

[Bieworm]

I thought you didn't have a scope? What does it look like at the output (speaker)?

Not really like crossover distortion

20210105_221143.jpg

As for the scope... I just didn't know how to use it I guess. I bought it a couple of weeks ago... but I intended to bring it back. I kinda got the hang of it today... should I change it for the Siglent 1102 ?

[Bieworm]

Ok that's not the problem apparently. Nor 24, 18 or 15V changes anything. I can see my DVM read the values and they stay proper un the zener range.. no change in fizz.
Should I put the 24V 5W zener in anyway for the max bias?

It was worth a shot, but I'm not surprised. You can leave the 24V 5W Zener in if you want, but I don't know that it will do anything for you.

I think the PR mod or the snubber would more likely be helpful.

What are all your voltages now? Can you find documentation with target voltages?

Another possible cause of this is if your signal is too hot going from one stage to the next. Or if you have too much low end getting amplified in the signal path.

I just looked a little more in depth at your schematic. It's not 2 separate channels so much as it looks like switching on the relay cascades V1A into V2B.

If you already have this issue before switching on the higher gain, then you might be able to start trying a couple things earlier in the circuit.

Did you say the circuit has a lot of bottom end? Maybe C9 should be experimented with. Try dropping that to 1µF (or .68, or 2, whatever, something lower).

If that doesn't help, it would be useful to figure out roughly where the issue begins when the high gain switch is off. It's possible you may need to add some voltage dividers on the signal path to reduce the levels.

Target voltages would certainly help a lot. If your preamp is too hot, that could easily cause this issue.

Thanks,
Josh

Don't have target voltages. I wish I had em... top secret tone king gear...

Why do you think It's the clean channel? They only share the 1st triode and PI etc..and that's fine. It situates between the 1st triode and the 3rd preamp triode of the OD channel according to the scope? Or am I reading the wavefirms wrong?