A most unusual triode circuit

[Mark]

Hi

I have traced out the circuit of an Australian Sonola made by Ron Pearce.

There are a few photos of it at the garage area.

https://ampgarage.com/forum/viewtopic.php?f=7&t=37968

I have checked the wiring several times and I think I have drawn it correctly.

IMG_2108.jpeg

The obvious question is how does this work?
A 100 ohm cathode resistor in series with a 22uF cap, and the negative side as well?
The 8.2K resistor attaches to the secondary of the output transformer (8 ohm) I suspected that the 8.2K might have played a part in the biasing of triode, but it would have to go through the secondary of the output transformer, which seems bizarre.

I thought the feedback network in a blackface Champ would be a better circuit to implement.

IMG_2100.jpeg

Any thoughts on this one?

[pdf64]

Unless you're desperate to experiment, I suggest to leave the design alone.
Yes the OT secondary does seem to be getting used for the cathode's DC return to 0V common. But that's fine.
The Champ NFB network would increase the open loop gain a lot, which could cause problems, eg instability.
That cap does seem to be reverse polarity though, have you checked whether it is?

[Mark]

Unless you're desperate to experiment, I suggest to leave the design alone.
Yes the OT secondary does seem to be getting used for the cathode's DC return to 0V common. But that's fine.
The Champ NFB network would increase the open loop gain a lot, which could cause problems, eg instability.
That cap does seem to be reverse polarity though, have you checked whether it is?

Thanks for your reply, the amp doesn’t much in the way of gain, it’s a very clean amp. The cap definitely has the positive side to earth. I have never seen a circuit like this one.
I will remove the negative feedback from the amp for my own curiosity and then the cap to observe the change in cathode voltage. I don’t understand why the cap hasn’t blown itself up by now?

[Mark]

I decided to include a picture of the circuit that might show how I came to the conclusions that I did. I think we would agree that the circuit doesn’t really seem plausible or even a smart design.

IMG_2111.jpeg

IMG_2109.jpeg

IMG_2114.jpeg

Hopefully, the pictures show the earth wire grounding the positive side of the electrolytic 22uF cap, the black wire going from the 100 ohm resistor to pin 2 of the 6GW8, and the yellow wire going to the output transformer secondary to provide negative feedback.

[pdf64]

...
I will remove the negative feedback from the amp for my own curiosity

Curiosity (sometimes) doesn't work out too well for the cat!
If the 8k2 is left open circuit, that stage will lose its DC return path to 0V common, go into cut off, and put more voltage across that 10V ecap, which is already reverse polarity.
You could do it safely by moving the leg of the 8k2 that currently connects to the OT secondary to 0V common.
That will maintain the DC conditions.

[/quote] I don’t understand why the cap hasn’t blown itself up by now?
[/quote]
For something to blow up, there needs to be a lot of energy available / stored up somehow.
How much energy can pass through the cathode of a 12AX7, especially one that's tied to within a few volts of 0V common?

... I think we would agree that the circuit doesn’t really seem plausible or even a smart design.
...

I suggest not to rush to judgment
The reverse biased cap may have been a build error, and presumably the amp used to work?
The concept (of using the OT secondary like that) seems valid to me.

[nuke]

I think it was an assembly error. It probably worked as built, just not as designed.

It doesn't really make sense as built, but it isn't a fatal error.

[TUBEDUDE]

No, but it should be rectified.

[Mark]

Thanks to everyone who has taken time to assist me. I really appreciate it. I owe you a Darwin Stubby if we cross paths.

I collected my notes and drew up the circuit diagram.

IMG_2139.jpeg

I’m still puzzled as to what would be achieved with the 100 ohm and 22uF cap to earth regardless of its orientation.

Thanks again for your help.

[Lauri]

I’m still puzzled as to what would be achieved with the 100 ohm and 22uF cap to earth regardless of its orientation.

Voltage divider for the amount of negative feedback.

[martin manning]

I’m still puzzled as to what would be achieved with the 100 ohm and 22uF cap to earth regardless of its orientation.

In addition to the voltage divider effect for the NFB, the 100R and 22u nearly bypass the ~8k2 DC resistance to ground, making it ~100R for AC. That increases the stage gain to almost the level you would get with a fully-bypassed Rk.

[Lauri]

Have you changed the 22uF cap and put it the right way around? There's only about 3dB of negative feedback in that circuit but the cap probably doesn't work correctly the way it's connected and everything is out of whack.

[martin manning]

Seconded. A quick sim shows the cap sees around 3.1V DC, and with an assumed 18V pk on the speaker (20W into 8 ohms) it will not reverse polarity.

[Mark]

IMG_2142.jpeg

Firstly, thank you both for assisting me, I appreciate it.

The drawing illustrates my understanding of your explanation. Is this what you mean?

I haven’t changed the cap yet as I have another friend’s ENGL amp on the bench at the moment. I should complete the ENGL today and get back into the Sonola amp.
The cap looks like it is the original cap in the amp and will be replaced. I’m wondering how it left the assembly line with the caps polarity wrong?

So it’s unanimous that the cap’s polarity is incorrect?
That was also my initial thought though I had never come across a circuit like this one and thought there might be more to it than I’m aware of.

[martin manning]

The drawing illustrates my understanding of your explanation. Is this what you mean?

Yes, that's it. The 8k2 is functioning as both bias and feedback resistor, so to do it like the Champ you'd need to add another one as shown below. The bass response is very slightly increased, but otherwise it's equivalent to your circuit.

[Mark]

The drawing illustrates my understanding of your explanation. Is this what you mean?

Yes, that's it. The 8k2 is functioning as both bias and feedback resistor, so to do it like the Champ you'd need to add another one as shown below. The bass response is very slightly increased, but otherwise it's equivalent to your circuit.

Thanks Martin for your reply and suggestion. Much appreciated, I had thought using the Champ style circuit depending on how the amp sounds. So far it’s very bright as it has a very bright speaker and a 270pF bright cap.
I had thought of using a 1.5K instead of the 8.2K as rightly or wrongly to my ear the 1.5K value sounds fatter.

So I figure the cap has to be wired in with the wrong polarity. I have this strange feeling that I won’t hear any difference but the cap will fail in the fullness of time.

So the 22uF cap in series with the 100 ohm resistor has the wrong polarity?