We're Jammin,
I have a Bassman clone that is carved from a Dukane chassis. The Dukane Transformers have been used. It sounds great with Sozo Blue caps!
The amp sounds fine below a volume of 6. After 7, an instability begins to show during hard transient slaps, rakes, or chords. When the NFB loop is attached, the oscillation has the ability to perpetuate and self-sustain. When the FB loop is disconnected, the "wobble" will only occur for a second, then dissipate into the unknown abyss. The speed of the oscillation is about the same as a fast heart beat, and has the same kind of beat-pulse energy, too.
Is this an issue of:
1 - Phase Margin?
2 - Resonance in the Power Supply
3 - Too much impedance in the ground scheme or supply lines
4 - Bad cable layout
HERE'S WHAT I'VE TRIED:
I have a Bassman clone that is carved from a Dukane chassis. The Dukane Transformers have been used. It sounds great with Sozo Blue caps!
The amp sounds fine below a volume of 6. After 7, an instability begins to show during hard transient slaps, rakes, or chords. When the NFB loop is attached, the oscillation has the ability to perpetuate and self-sustain. When the FB loop is disconnected, the "wobble" will only occur for a second, then dissipate into the unknown abyss. The speed of the oscillation is about the same as a fast heart beat, and has the same kind of beat-pulse energy, too.
Is this an issue of:
1 - Phase Margin?
2 - Resonance in the Power Supply
3 - Too much impedance in the ground scheme or supply lines
4 - Bad cable layout
HERE'S WHAT I'VE TRIED:
- Short choke
- Double capacitance of each stage with parallel capacitor of same value and excellent ESR and ESL ratings
- Disconnect NFB
- Swap OT (original had a weird conductivity reading on 25V and 70V taps, internal short suspected)
- Swap ALL TUBES with new ones
- Use a SS rectifier tube socket (for testing)
- Shield input jack to V1 cable
- Shield input to V2 cable
- Lift Tone Stack ground
- Bypass Tone Stack
- Move 1st stage preamp reservoir from PS board onto Preamp board
- Apply Valve Wizard's Grounding scheme: Short and contained current paths, meaning C1 is in a closed loop with PT and rectifier
- More Grounding: 2nd node = C2 20uF= GND for P. Tubes and Bias Supply. 3rd node = C3 20uF = GND for PI and Speaker
- Run all signal and ground leads against chassis
Since you made up the wiring - as opposed to wiring it like a Fender - any chance one or more of the preamp tubes share a ground with a filter cap for other stages?
The fact disconnecting teh NFB makes it better, connecting the NFB makes the thing a bog oscillator. Perhaps your output transformer is wired "Backwards. In other words exchange the brown and blow wires to the power tube plates. (Or whatever color if your OT is not standard colors for USA.)
From your description, I get a picture of an amp going unstable and "motorboating". That can come from ground separations, it can come from wrong phase NFB.
The fact disconnecting teh NFB makes it better, connecting the NFB makes the thing a bog oscillator. Perhaps your output transformer is wired "Backwards. In other words exchange the brown and blow wires to the power tube plates. (Or whatever color if your OT is not standard colors for USA.)
From your description, I get a picture of an amp going unstable and "motorboating". That can come from ground separations, it can come from wrong phase NFB.
Schematic, is it the same as the 5F6?
Layout?
If there is a decoupling issue between the output stage and the driver/pre amp, this will cause your issue, as will layout errors and the wrong value choke.
NFB wired out of phase can be an issue but tends to be a higher frequency oscillation.
A capacitor of a certain value, is a capacitor of that value. The only differences are reliability and tolerances. (temperature, voltage etc).
Because they are 'blue' doesn't improve the quality.
Is the output transformer loading correct?
Have you used the correct value choke?
I trust and hope you are not using the 'death cap' with switch and used an earthed mains lead to chassis!
Layout?
If there is a decoupling issue between the output stage and the driver/pre amp, this will cause your issue, as will layout errors and the wrong value choke.
NFB wired out of phase can be an issue but tends to be a higher frequency oscillation.
A capacitor of a certain value, is a capacitor of that value. The only differences are reliability and tolerances. (temperature, voltage etc).
Because they are 'blue' doesn't improve the quality.
Is the output transformer loading correct?
Have you used the correct value choke?
I trust and hope you are not using the 'death cap' with switch and used an earthed mains lead to chassis!
When the choke is shorted, the oscillations are damped completely within 3 seconds. This is with the FB connected. Thanks for the tip about choke value!
The FB leads are connected correctly. I can recognize the squeal from flipped primary leads. When the OT leads are incorrect, this would be an issue of phase margin? Yes, some will say positive FB, I am asking if the term Phase Margin essentially means that a stage is receiving positive feedback at some frequency.
No Deathcap, no poisonous mushrooms, no dying.😉
No change in osc. with different load, I also thought about that last night! Thanks. To be thorough, the correct 4/8 ohm taps have been tested as such by the power they dissipate into a load.
Power cord is 3-prong and wired correctly.
**Fender certainly didn't use a buss wire or star ground, etc. grounding scheme. I usually see them route preamp grounds to chassis as quickly as possible. But, they did attach 3 stages to the same 1st stage 10uF filter. I moved this 1st stage filter to tighten the loop between these 3 stages (1st and 2nd gain, Kfollower). I am double checking the ground routing again to make sure the loops are correct.
I will disconnect all of my grounds and play around with different connection points.
How should I go about testing a decoupling issue between driver/pre and power amp? This sounds interesting. All of the coupling caps are new.
The FB leads are connected correctly. I can recognize the squeal from flipped primary leads. When the OT leads are incorrect, this would be an issue of phase margin? Yes, some will say positive FB, I am asking if the term Phase Margin essentially means that a stage is receiving positive feedback at some frequency.
No Deathcap, no poisonous mushrooms, no dying.😉
No change in osc. with different load, I also thought about that last night! Thanks. To be thorough, the correct 4/8 ohm taps have been tested as such by the power they dissipate into a load.
Power cord is 3-prong and wired correctly.
**Fender certainly didn't use a buss wire or star ground, etc. grounding scheme. I usually see them route preamp grounds to chassis as quickly as possible. But, they did attach 3 stages to the same 1st stage 10uF filter. I moved this 1st stage filter to tighten the loop between these 3 stages (1st and 2nd gain, Kfollower). I am double checking the ground routing again to make sure the loops are correct.
I will disconnect all of my grounds and play around with different connection points.
How should I go about testing a decoupling issue between driver/pre and power amp? This sounds interesting. All of the coupling caps are new.
Attachments
The very low oscillation frequency, combined with your observations about the choke, strongly suggest the oscillations are caused by the way the amp interacts with the power supply rails. It's motorboating, in other words.
As has already been suggested, make sure B+ is properly filtered and decoupled at least every two gain stages. Also look into your grounding scheme - make sure input stage grounds don't share long wire runs with the ground wires from output valves or speaker.
I wonder if wiring a suitable power resistor in parallel with the choke will help. (The choke, along with the B+ filter caps, can cause the power supply rail to "ring" in response to a transient, such as the spike in output valve current each time you pick a guitar string. It seems your amp is currently so sensitive to this that it causes outright oscillation.)
-Gnobuddy
I don't have much in the way of direction, just to say I had instabilities in a Bassman styled amp (the difference was having one channel being a pentode input) and I solved it doing something to the input jack. Nice pic by the way.
Yes, the pic is a schem.
THIS IS WHAT HAS BEEN ISOLATED:
THIS IS WHAT HAS BEEN ISOLATED:
- Issue stops when V1 coupling cap is removed. This amp only uses one input and 1/2 AY7, so please disregard the second V1 coupling cap.
- When V1 tube is removed, the issue can still be triggered using a sig gen on V2a grid resistor.
- V1 Coupling cap and P.S. must be coupled together to cause issue.
- At V1, Cap without P.S., and P.S. without Cap have been tried. No benefit.
- A different Coupling cap has been installed, with better fit against the chassis. No change.
Attachments
An explanation for the "mess of discovery" seen inside the amp! All changes have been made AFTER amp displayed exact same issue.
The yellow jumper, top right connects the new OT secondary lead to output jack. Pri side leads are Ape Hangers right now!
The black F+T caps are soldered in parallel with original IC's, as well as the brown radial who is sleeping.
Green jumper, top right shorts Choke. Yes, choke is in a weird place with lines that run parallel to main ground, which is also disturbingly long, and bias supply (middle right under clothes-pin).
Red jumper, left side connects TS to ground. At one point I was removing all variables.
The yellow jumper, top right connects the new OT secondary lead to output jack. Pri side leads are Ape Hangers right now!
The black F+T caps are soldered in parallel with original IC's, as well as the brown radial who is sleeping.
Green jumper, top right shorts Choke. Yes, choke is in a weird place with lines that run parallel to main ground, which is also disturbingly long, and bias supply (middle right under clothes-pin).
Red jumper, left side connects TS to ground. At one point I was removing all variables.
- Status
- Not open for further replies.