Cirklotron Preamp with ECC88.
I had this idea a couple of years ago, but lack of high value bifilare winded chokes has stopped me. Im seriously thinking of building this one as I think its a very simple and elegant design.
Its a Circlotron equivalent to a long tailed pair / diffpair. You can see the cathodes being connected ontop of the big chokes.
Any and all technical comments are most welcome, since I have never seen anything like this anywhere before.
Cheers Johannes.
I had this idea a couple of years ago, but lack of high value bifilare winded chokes has stopped me. Im seriously thinking of building this one as I think its a very simple and elegant design.
Its a Circlotron equivalent to a long tailed pair / diffpair. You can see the cathodes being connected ontop of the big chokes.
Any and all technical comments are most welcome, since I have never seen anything like this anywhere before.
Cheers Johannes.
An externally hosted image should be here but it was not working when we last tested it.
And here is the chematic.
I don't understand the purpose of the chokes or the direct connection of the cathodes to one another. The circlotron that I know is just two tubes and two power supplies in a circle with the load taken between the cathodes. A reference for the input signals is usually derived from the load somehow.
I dont want any capacitors in the circuit at all. I want as few active devices as possible, yet I want balanced operation and the cancelation of second harmonics.
I dont like having the output at a high voltage because of the startup peaks that will destroy the poweramp and my speakers. I like the symmetry.
The chokes are there to isolate the plates from the cathodes to stop the feedback mekanism that makes them into cathodefollowers. The connection between the cathods are there to provide a low impedans connection , kind of like "ground". Its basicly a LTP. Works in the same way. The chokes also serves as plate load, It converts the current variation inte voltage variation.
Johannes.
Normally, the Circlotron is used as a work-around for mediocre transformers in power amps. As a demerit, the signal is now run through two sets of capacitors and the circuit becomes much more complicated.
You can get all the balance and harmonic cancellation with a simple differential amp and an output transformer. Much, much simpler and there's only one set of power supply caps (they ARE in the signal path, after all).
You can get all the balance and harmonic cancellation with a simple differential amp and an output transformer. Much, much simpler and there's only one set of power supply caps (they ARE in the signal path, after all).
Circlotron....
I am familiar with the circlotron, having just built a circlotron power amp....
BUT I cannot for the life of me, see how the posted schematic can possibly work, let alone solve your problems.....
Whats wrong with caps anyway.....Your source CD player and other devices probably have them, as well as your power amp.....
This scheme is slightly similar to my own but without the chokes, or the cathode short and taking the out-put accross the cathodes......Shorting both cathodes together, effectively shorts the output in your scheme as far as I can see.....
I am familiar with the circlotron, having just built a circlotron power amp....
BUT I cannot for the life of me, see how the posted schematic can possibly work, let alone solve your problems.....
Whats wrong with caps anyway.....Your source CD player and other devices probably have them, as well as your power amp.....
This scheme is slightly similar to my own but without the chokes, or the cathode short and taking the out-put accross the cathodes......Shorting both cathodes together, effectively shorts the output in your scheme as far as I can see.....
I think the circuit does work as shown.
The DC biasing works in terms of a conventional circlotron circuit.
The AC gain works similar to Circlomanens quaditron circuits posted in Pass Labs forum.
I guess I would call it some sort of circlotron/floating differential balanced amp thing???
With the feedback network shown, it exhibits NP's SuSy effect.
Also removing the feedback network might make understanding the circuit easier.
Circlomanen posted a similar circuit in Pass Labs forum
using his Zenitron circuit with Power JFets.
It does seem like overkill for a preamp though.
But hey, this is DIY audio. If it makes you happy build it.
I don't belive it, I'm in the tube forum.
Tom
The DC biasing works in terms of a conventional circlotron circuit.
The AC gain works similar to Circlomanens quaditron circuits posted in Pass Labs forum.
I guess I would call it some sort of circlotron/floating differential balanced amp thing???
With the feedback network shown, it exhibits NP's SuSy effect.
Also removing the feedback network might make understanding the circuit easier.
Circlomanen posted a similar circuit in Pass Labs forum
using his Zenitron circuit with Power JFets.
It does seem like overkill for a preamp though.
But hey, this is DIY audio. If it makes you happy build it.
I don't belive it, I'm in the tube forum.
Tom
After reading your responses, and redrawing the circuit a few different ways I might have an idea what you're trying to do. It looks to me like what you have operates something like an ordinary plate loaded PP stage (it has voltage gain) but with a separate floating supply for each tube. This does in fact eliminate coupling caps on the output; the grid referencing resistors put the load at ground potential so it can be direct coupled to a power amp. However, I can't forget that the supplies have capacitors in them, and the signal must go straight through them on its way from the plates to the output.
My first thought was to ground the connection between the cathodes (or simply ground the cathodes) but I realized that that would put the output connection at some DC potential.
And, I finally realized why you need a bifilar dual winding choke; that's what makes it a longtail pair.
Hmmm... It's not the way I would have gone about it, but that's not to say that it's wrong.
I probably would have gone straight for a 1+1:1+1 transformer. That would simplify the power supply A LOT. Besides having only one supply, the PSRR would be tremendously better. And, although the single supply would necessarily have caps in it, the signal would not need to travel through them (being a PP stage.)
But hey, those are all 50 year old text book arguments. Your idea might just sound fantastic. WTFDIK?
-- Dave
My first thought was to ground the connection between the cathodes (or simply ground the cathodes) but I realized that that would put the output connection at some DC potential.
And, I finally realized why you need a bifilar dual winding choke; that's what makes it a longtail pair.
Hmmm... It's not the way I would have gone about it, but that's not to say that it's wrong.
I probably would have gone straight for a 1+1:1+1 transformer. That would simplify the power supply A LOT. Besides having only one supply, the PSRR would be tremendously better. And, although the single supply would necessarily have caps in it, the signal would not need to travel through them (being a PP stage.)But hey, those are all 50 year old text book arguments. Your idea might just sound fantastic. WTFDIK?
-- Dave
Hey Circlomanen, are you sure you want bifilar chokes? It seems to me that separate chokes would work, dual winding chokes might buy you extra inductance and any coupling wouldn't hurt. But I think that the added capacitance of bifilar would actually hurt the hf response in this application.
Maybe I'm missing something?
Maybe I'm missing something?
I just did a little visualizing to see how this circuit works. ( see attached diagram) I tilted the components to indicate the voltage levels when one tube puts AC current through it. Tilts the other way for the other tube.
Appears to work fine, but it does EXACTLY the same thing as a normal circlotron with the feedback resistor (R5, R6 ...) values shown. The feedback resistors reduce the grid-cathode voltage by half from the input. The normal circlotron likewise subtracts (by series subtraction) half the total output from the grid -cathode voltage, ending up with the same gain and same signals on the triodes.
As Dave suggested, the inductors really don't want to be bifilar, as they have half the load voltage across them at each corresponding point.
Assuming that the "supersymmetry" aspect is related to taking balanced feedback from both differential output polarities back to each active device input, the normal circlotron circuit already does this too.
Only advantage I see is elimination of the DC grid bias voltage from either the input or the output so no coupling capacitor is needed. But the signal is still passing thru the power supply capacitors.
Don
Appears to work fine, but it does EXACTLY the same thing as a normal circlotron with the feedback resistor (R5, R6 ...) values shown. The feedback resistors reduce the grid-cathode voltage by half from the input. The normal circlotron likewise subtracts (by series subtraction) half the total output from the grid -cathode voltage, ending up with the same gain and same signals on the triodes.
As Dave suggested, the inductors really don't want to be bifilar, as they have half the load voltage across them at each corresponding point.
Assuming that the "supersymmetry" aspect is related to taking balanced feedback from both differential output polarities back to each active device input, the normal circlotron circuit already does this too.
Only advantage I see is elimination of the DC grid bias voltage from either the input or the output so no coupling capacitor is needed. But the signal is still passing thru the power supply capacitors.
Don
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.