Hybrid Circlotron Amplifier with only 3 components on the signal path

[adamus]

looking forward to more details. I feel a hybrid will be my next project and this looks good (although expensive!).

 

[audiodesign]

In the best configuration for 50W with 6 chokes the cost should be about 1200 euro including HiFi2000 chassie h=5 units

but now I prefer it to my actual GM70/813 with 2400 of components.

 

[P.Lacombe]

Sure. This is the basics of what anyone interested in a Circlotron has to know. In the rather simple schematic, we see a CF-coupled triode amplifier. The load is the the cathode resistor so I depicted it with a loudspeaker symbol.

The power supply is conventional and so is depicted by a box. In this case I have drawn it so that the cathode load goes to ground as does the power supply. The input is the grid. As you can see, no capacitor in the power supply of this example can also be construed to be a coupling capacitor.

(...)

The current supply which flows across the load is the same that flows across the power supply, ie DC and AC. If you don't agree, try to connect an inductor in the plate circuit... and listen

 

[atmasphere]

The current supply which flows across the load is the same that flows across the power supply, ie DC and AC. If you don't agree, try to connect an inductor in the plate circuit... and listen

I'm not used to seeing AC currents (unless the supply is not bypassed very well) flowing in a DC supply, save for those that are inputs to the rectifiers.

So, if I get you right then, in your book, all filter caps in all power supplies are coupling caps, in all audio circuits?? Or do you know of an example where this is not the case?

 

[banat]

Actually only real AC current is present on the Main transformer pimary/ secondaryes and Nowhere else.

After Deflectors(Bridge Rectifier) everything else inside the Circlotron circuit is flow of pure DC current,and this DC curent flow through the PP Circlotron Balanced Bridge Circuit.This DC current flow is regulated with Output Tubes( Basicly Output Power Tubes are just one Big- One way- ultrafast regulation Reostat ).Only exception here is resulting AC current passing trough the Load(Speaker) which is generated from PP Circlotron circuit oposite DC current arangement.

Trough smoothing kapacitors don`t pass any AC or DC curent.If any of AC or DC current start to passing trough this kapacitors results well be instant Amplifier failure caused with Kapacitors failure.

Smoothing Kapacitors here is just only one big DC current reservoirs which is constatly charching and discharching,Speed of this charching/discharging rate mostly depends from Electrolityc Kapacitor quality,and that is the reason why we sometime put some Blok Kapacitor in parallel with this Elko`s , Blok`s are simply faster.

Best regards

 

[Alastair E]

Hmmm.....

Well, not sure of the realities of the PSU caps being coupling-caps,--

BUT, if you've ever tried to build a Stereo Circlotron and run both channels off only 2 circlotron rails, you find that you've made a Mono Two-channel amp.

The Crosstalk is terrible from the channel with signal to the channel thats not driven......
--I did this, I know!

If the PSU's didnt form part of the signal route, WHY did I make a dual-channel Mono Amp??

 

[banat]

Hmmm.....

Well, not sure of the realities of the PSU caps being coupling-caps,--

BUT, if you've ever tried to build a Stereo Circlotron and run both channels off only 2 circlotron rails, you find that you've made a Mono Two-channel amp.

The Crosstalk is terrible from the channel with signal to the channel thats not driven......
--I did this, I know!

If the PSU's didnt form part of the signal route, WHY did I make a dual-channel Mono Amp??

Don`t want to` Hi Jack`this thread started by respected Audiodesign DIY Member

Just quick answer:

Reason for this Stereo Circlotron Amps `Crosstalking` when PSU is common for both Amp is next:
With this two common rail PSU arangement both independent Loadspeaker became DC conected in parallel ,whatever if You drive only one Circlotron Amp both loadspeaker well `sing`. And actually thats `efect` is not real clasik defined Crosstalk efect.

Simply Circlotron Power Amp Stage need two Galvanic independent DC power rail for Each chanell.
Best Regards for All

 

[Alastair E]

Ah.....

Yes, of course.... Doh!

I was just testing out an amp that I had built confirming both channels alive. I had not completed the PSU.

Since writing the above post I reasoned out that I had in fact placed both output stages in parallel.

 

[P.Lacombe]

(...)

So, if I get you right then, in your book, all filter caps in all power supplies are coupling caps, in all audio circuits?? Or do you know of an example where this is not the case?

Filter caps in power supplies are couplig caps if, (and only if) they are in the path of the signal, ie if the AC current flowing is part or totality of the AC current flowing across the load. This can be proved by inserting very small resistor (0.1 ohm) and connect cheap earphones on this resistor (listen !)

If the power supply uses pi cell (one capacitor, one inductor, one capacitor) the first capacitor is not coupling capacitor, but the second is. One can verify with the earphones.

In a pure class A double serial push-pull (bridged) the total current supply is constant, in this case the PS capacitors are not coupling capacitors.

In ordinary class A SE with output transformer (300B...) the final capacitor of the PS is a coupling capacitor. Sad, but unavoidable.

In ordinary pure class A parallel push-pull, with output transformer, the total current is constant, hence the final capacitor of the PS is not a coupling capacitor. If the amplifier is driven in class AB , one can hear very distorted music...and capacitor becomes coupling capacitor... and so on !

Because of my bad knowledge of the english language, I perhaps don't understand all you wrote, for instance I don't understand "in your book" . Apologies

 

[Bigun]

I maybe flogging a dead horse here, but I still don't get it.

I've attached a sketch.

Forget the DC currents, there's no signal in there.

The arrow indicates an AC current, the signal, flowing out of the top active device and then through speaker. Current goes in loops, it has to find it's way back to the other side of the top active device again. It does so by going through the top psu, which is floating.

So the signal current goes through both psu's.

There are only so many ways we like to build an audio power supply. Transformers with rectifier diodes or switch mode power supplies are not conducting constantly. Recitifier diodes only conduct when they are forward biassed and have to be followed by either a capacitor or an LC filter. Switch mode supplies have to have an LC filter. Any AC currents flowing through the Circlotron power supplies flow through the last capacitor in the power supply.

So as far as I can see, the Circlotron always has a capacitor in the signal path.

Sure you can use batteries, but they're just really large capacitors !

I don't believe I've ever seen an audio amplifier design without a capacitor in the signal path somewhere.

 

[Bigun]

I don't believe I've ever seen an audio amplifier design without a capacitor in the signal path somewhere.

Except this one - psu is out of the way. Balanced inputs.

 

[banat]

Here is two simplified schematic with DC flow analysis.
One picture is only for half Circlotron circuit,and second is for whole Circlotron circuit.
Same analysis is valid for SS Circlotron Amps when we use SS aktive elements,similar like in
Audiodesign Diy Member Hybrid Amps.
Conclusion:AC power goes Only trough the Load(Speaker), like produkt of Oposite kathodes
PP DC curent flow, aranged in Circlotron PP output power stage circuit.

Best Regards for All

 

[Bigun]

Look at your picture on the left. The signal (i.e. the AC current, not the power from the wall socket) flows through the two capacitors you've drawn.

I hope we're talking about the same question ? I'm just trying to explain why the Circlotron has two capacitors in the signal path. I'm not talking about DC at all.

 

[Sheldon]

Arguing whether or not capacitors are in the signal path is beside the point. A narrow definition of signal path, would imply only capacitors that are completely in series with the signal (i.e., blocking capacitors) However, as DF96 pointed out, other capacitors can affect the signal.

So the real question should be; How much of the non-linearity of the capacitor can be imposed on the signal? For a blocking capacitor, it is 100%. For a final power supply capacitor, to a first approximation, it will be a function of the PSRR of the stage. It should be a simple matter to calculate the proportion of distortion products of the capacitor that will be imposed upon the output. Discussions about how much a capacitor is "in" or "out" of the signal path are pointless, without reference to proportion (and, one might add some quantitative knowledge about the non-linearity of the capacitor in question).

Sheldon

 

[banat]

Look at your picture on the left. The signal (i.e. the AC current, not the power from the wall socket) flows through the two capacitors you've drawn.

I hope we're talking about the same question ? I'm just trying to explain why the Circlotron has two capacitors in the signal path. I'm not talking about DC at all.

Nope,This four small arrows on the half Circlotron circuit picture sign the charge/discharge current direktions of this kapacitors(C).

They are constantly charged from the bridge rectifier with 100Hz DC pulsating current,and constantly discharged from Output power Tube and Load(Speaker), both conected in series.Mentioned before:This two kapacitors are just two Big energy(DC current) reservoirs.
But trough this Kapacitors dont flow any DC or AC current,except if this kapacitors `leak` or is internaly shorted,(but that well be produce Amp mallfunction kondition).

Any way ,signal path` termin was allways strange to me,think such of thing don`t exist.

Best regards

 

[Bigun]

Sheldon, I disagree that the psrr is relevant here too in the context of the conceptual drawing (no gnf).


I admit defeat, we will agree to disagree. So long as we're all having fun !!

 

[Wavebourn]

PSRR can be measured, estimated, calculated. But PS is always in signal path. However, if like on Bigun's drawing CSS is in series with filter capacitors, influence of PS is less due to much higher dynamic resistance of the CCS than of the filter capacitor, but anyway they are in series, and the corresponding PS capacitor is always in a signal path.

 

[banat]

. So long as we're all having fun !!

Yes Man

 

[Sheldon]

PSRR can be measured, estimated, calculated. But PS is always in signal path. However, if like on Bigun's drawing CSS is in series with filter capacitors, influence of PS is less due to much higher dynamic resistance of the CCS than of the filter capacitor, but anyway they are in series, and the corresponding PS capacitor is always in a signal path.

"is always in a signal path." Agreed, never disagreed. I still say that it's a pointless distinction without quantifying how much it affects the signal. It's said all the time here that a good design means making the right compromises. Can't make the right ones without knowing the magnitude of the
effect.


Sheldon

 

[Bigun]

Can't make the right ones without knowing the magnitude of the
effect.


Sheldon

D. Self has had stuff to say on distortion arising from signal passing trough large electrolytic capacitors - such as those used on the psu or on cap coupled outputs. I don't have a link to hand but I think it's one answer to your question. His findings, if I remember correctly, were that it depends on the ac voltage drop across the capacitor - you want it to be low. When it starts to increase distortion takes off quickly. The solution was to use large valued capacitors, much larger than would be suggested from an analysis of the roll-off frequency you need for the audio signal. In this way the impedance of the capacitor is lower for frequencies circulating through them and therefore the a.c. voltage across the caps is reduced.

If this is true then it matters not the dynamic resistance of the CCS relative to the psu capacitors but only the magnitude and frequency of the current flow through the capacitors (along with the details of the capacitor). The current and frequency are the same as that which flows through the load since they are in series.

I've never worried about 'em myself - never been willing to pay out for boutique (expensive) capacitors for a psu. I did try paralleling some electrolytics with film capacitors once but this doesn't address the issue identified by D. Self - which suggests the need is not for parallel film caps but bigger and better electrolytics.

It seems to me that Nelson took this route with his Zen amplifiers - a very large electrolytic at the output.