Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

Graham

"OTL" for "output transformer less" amps exist, but they tend t use mny
tubes in parallel to get the ability to drive a speaker. And you have
to deal with the DC in the output, which speakers don't like.

As the B+ on the plate would be essentially constant, you might as well
use pentodes.

"Bob"

As the B+ on the plate would be essentially constant,


** Huh ???

Constant relative to the chassis.

Not constant relative to the cathode.



.......... Phil

Bob wrote:

"OTL" for "output transformer less" amps exist, but they tend t use mny
tubes in parallel to get the ability to drive a speaker. And you have
to deal with the DC in the output, which speakers don't like.

I wasn't meaning an OTL design. Stick the transformer in the cathode
circuit instead.

As the B+ on the plate would be essentially constant, you might as well
use pentodes.

Ah but the 'tubists' want triodes.

Graham

Pooh Bear wrote:


I wasn't meaning an OTL design. Stick the transformer in the cathode
circuit instead.


Its been done. One of the Audio Anthology books has one single ended
6V6. The RDH4 describes one using eight 6V6s in push pull.


Adam

On Sun, 18 Dec 2005 23:33:01 +0000, Pooh Bear
wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

It was done back in the classic era, when many approaches
were being tried. Later, the drive difficulties forced
designs in the direction of combined plate and cathode
loading; for example the Quad and McIntosh designs.

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

You may be interested in the "Circlotron" topology. It
takes the McIntosh to an interesting next step.

Always the best,

Chris Hornbeck

"Angels are, dreaming of you" - Sonic Youth

in article , Chris Hornbeck at
wrote on 12/18/05 11:11 PM:

On Sun, 18 Dec 2005 23:33:01 +0000, Pooh Bear
wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

It was done back in the classic era, when many approaches
were being tried. Later, the drive difficulties forced
designs in the direction of combined plate and cathode
loading; for example the Quad and McIntosh designs.

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

You may be interested in the "Circlotron" topology. It
takes the McIntosh to an interesting next step.

Always the best,

Chris Hornbeck

"Angels are, dreaming of you" - Sonic Youth


A good example of the genre is one patented by Bruce Rozenblitz.

Atmasphere also sells 6AS7 OTL amps.

Jon

Chris Hornbeck wrote:

On Sun, 18 Dec 2005 23:33:01 +0000, Pooh Bear
wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

It was done back in the classic era, when many approaches
were being tried. Later, the drive difficulties forced
designs in the direction of combined plate and cathode
loading; for example the Quad and McIntosh designs.

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

Why any need for split primaries ? I'm talking about a single tube.

I see no trouble achieving the neccessary drive volts with clever design.
Indeed the simplest method uses a transformer.

So, basically, it seems that the idea hasn't been given a decent chance.

Graham

Jon Yaeger wrote:

in article , Chris Hornbeck at
wrote on 12/18/05 11:11 PM:

On Sun, 18 Dec 2005 23:33:01 +0000, Pooh Bear
wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

It was done back in the classic era, when many approaches
were being tried. Later, the drive difficulties forced
designs in the direction of combined plate and cathode
loading; for example the Quad and McIntosh designs.

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

You may be interested in the "Circlotron" topology. It
takes the McIntosh to an interesting next step.

Always the best,

Chris Hornbeck

"Angels are, dreaming of you" - Sonic Youth

A good example of the genre is one patented by Bruce Rozenblitz.

Atmasphere also sells 6AS7 OTL amps.

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain. Inherently
flat frequency response ! Much more linear.

Graham

On Mon, 19 Dec 2005 05:05:36 +0000, Pooh Bear
wrote:

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

Why any need for split primaries ? I'm talking about a single tube.

I see no trouble achieving the neccessary drive volts with clever design.
Indeed the simplest method uses a transformer.

So, basically, it seems that the idea hasn't been given a decent chance.

You're working your way through French and Japanese
thought of the 1990's. Still fashionable in some circles,
but current r.a.t fashion is to minimize the number
of transformers, while still demanding linear loadlines.

BTW, if you accept signal transformers and all that
they imply, you're on a slippery slope to the Dark Side
of audiophoolery (to use your term). And Luke, I'm
not really your father.

BTW, I *knew* you'd be interested in this stuff! Thanks
for all thoughts and arguments,

Chris Hornbeck

"Angels are, dreaming of you" - Sonic Youth

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain. Inherently
flat frequency response ! Much more linear.



** A power triode inherently has a flat response from DC to hundreds of KHz.

Using it as a cathode follower allows a modest amount of local NFB to
improve linearity at the expense of a much more complex drive arrangements
or the HORROR of making a wide band audio transformer for * step up *
voltage drive.

Poopie Bear needs to go easy on the laxatives.




......... Phil

Chris Hornbeck wrote:

On Mon, 19 Dec 2005 05:05:36 +0000, Pooh Bear
wrote:

Heroic efforts at acheiving 1000 volt grid swings seem,
on balance, to be less worthwhile than the compromises
of split primaries.

Why any need for split primaries ? I'm talking about a single tube.

I see no trouble achieving the neccessary drive volts with clever design.
Indeed the simplest method uses a transformer.

So, basically, it seems that the idea hasn't been given a decent chance.

You're working your way through French and Japanese
thought of the 1990's. Still fashionable in some circles,
but current r.a.t fashion is to minimize the number
of transformers, while still demanding linear loadlines.

The transformer is simply one option.


BTW, if you accept signal transformers and all that
they imply, you're on a slippery slope to the Dark Side
of audiophoolery (to use your term). And Luke, I'm
not really your father.

So-called 'small signal' transformers have far fewer problems than their power
cousins. The number of turns may be a little on the high side though.

Actually on that subject - I got here partly on account of a related pro-audio
matter re : mic pres and subjective thoughts about their audio quality ( esp
when transformers are involved ).

Of course the *easy* way to get the grid volts is to use some semis. Makes it
a total breeze. I'm considering a hybrid since the use of semiconductor active
loads doesn't seem to incur so much wrath.


BTW, I *knew* you'd be interested in this stuff! Thanks
for all thoughts and arguments,

You're welcome. It's kinda intruiging.

Graham

Phil Allison wrote:

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain. Inherently
flat frequency response ! Much more linear.


** A power triode inherently has a flat response from DC to hundreds of KHz.

With a *pure resistive* load - yes.... The impedance presented by a transformer
loaded with a loudspeaker is *anything but* pure resistive however !

With a reactive load in the plate circuit the gain will be approx proportional
to the load Z. Hence any variation in Z leads to a variation in gain and hence
changes in the amplitude/frequncy repsonse.

That's *basic* stuff !


Using it as a cathode follower allows a modest amount of local NFB to
improve linearity at the expense of a much more complex drive arrangements
or the HORROR of making a wide band audio transformer for * step up *
voltage drive.

I'm looking at some interesting drive arrangements. If I use a solid-state
current mirror ( not something that tubes can do ) it's done.

Graham

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain.
Inherently
flat frequency response ! Much more linear.


** A power triode inherently has a flat response from DC to hundreds of
KHz.

With a *pure resistive* load - yes....


** That is how amps are tested for frequency response.

Output Z and speaker loading effects are a separate issue.

A triode amp can easily have low output Z across the audio band, by use of
NFB.

Fool.



......... Phil

Phil Allison wrote:

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain.
Inherently
flat frequency response ! Much more linear.


** A power triode inherently has a flat response from DC to hundreds of
KHz.

With a *pure resistive* load - yes....

** That is how amps are tested for frequency response.

And it's not how you listen to them you cretin.

The big difference here is that an SS amp with inherently large indifference
to load impeadance by virtue of its emitter follower output simply won't care
about the load impedance ( or its phase angle either come to that ) and will
deliver the same voltage regardless.

Output Z and speaker loading effects are a separate issue.

They are actually related.

A triode amp can easily have low output Z across the audio band, by use of
NFB.

And we know that Jootikins preference as being discussed is for *zero* NFB.

The hilarity is that the SET ( and indeed P-P ) output can only begin to have
a *flat frequency response* and also lowish output Z by using serious NFB.

Graham

"Poopie ****WIT Bear"


** That is how amps are tested for frequency response.

And it's not how you listen to them you cretin.


** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.



Output Z and speaker loading effects are a separate issue.

They are actually related.


** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.


A triode amp can easily have low output Z across the audio band, by use
of
NFB.

And we know that Jootikins preference as being discussed is for *zero*
NFB.


** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.


The hilarity is that the SET ( and indeed P-P ) output can only begin to
have
a *flat frequency response* and also lowish output Z by using serious NFB.


** The former is easy even with no loop feedback - fool.

If you keep moving the goalposts - you will eventually FORCE the ball to
go between them.

Fool.



........... Phil

Phil Allison wrote:

"Poopie ****WIT Bear"


** That is how amps are tested for frequency response.

And it's not how you listen to them you cretin.

** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.

Output Z and speaker loading effects are a separate issue.

They are actually related.

** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.


A triode amp can easily have low output Z across the audio band, by use
of
NFB.

And we know that Jootikins preference as being discussed is for *zero*
NFB.

** If you keep moving the goalposts about - you will eventually FORCE the
ball to go between them.

Fool.

The hilarity is that the SET ( and indeed P-P ) output can only begin to
have
a *flat frequency response* and also lowish output Z by using serious NFB.

** The former is easy even with no loop feedback - fool.

If you keep moving the goalposts - you will eventually FORCE the ball to
go between them.

Fool.

You're useless.

How do you think an active filter works ? Voodoo ?

I don't suppose a spice model would even convince you ?

Graham

** Poopie Bear is completely insane.

He spends all day chasing his own tail.





......... Phil

Phil Allison wrote:

** Poopie Bear is completely insane.

He spends all day chasing his own tail.

Spoken like a good loony.

Get a life Allison. I don't expect you to understand the finer points
of circuit theory that *toaster_repair_man 101* doesn't teach. So STFU
!

Graham

Pooh Bear wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

Graham

This has been posted for a whole 12 hours already & Bret Ludwig has not
shown us a better way. Could he actually be stymied? Probably not! Watch
for another fog of bull****!

Cheers, John Stewart

"Poopie Bear"


** Graham Stevenson .....


Dead meat.





......... Phil

In article , Pooh Bear
wrote:

Phil Allison wrote:

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain. Inherently
flat frequency response ! Much more linear.


** A power triode inherently has a flat response from DC to hundreds of KHz.

With a *pure resistive* load - yes.... The impedance presented by a
transformer
loaded with a loudspeaker is *anything but* pure resistive however !

With a reactive load in the plate circuit the gain will be approx proportional
to the load Z. Hence any variation in Z leads to a variation in gain and hence
changes in the amplitude/frequncy repsonse.

That's *basic* stuff !

The same is true of transistors, just substitute the word collector for plate.


Regards,

John Byrns


Surf my web pages at, http://users.rcn.com/jbyrns/

Adam Stouffer wrote:

Pooh Bear wrote:


I wasn't meaning an OTL design. Stick the transformer in the cathode
circuit instead.


Its been done. One of the Audio Anthology books has one single ended
6V6. The RDH4 describes one using eight 6V6s in push pull.

Thanks. I'll check out RDH4 for that example

Graham

John Byrns wrote:

In article , Pooh Bear
wrote:

Phil Allison wrote:

"Poopie Bear"

Once again I'm not talking about an OTL design.

Simply the transformer in the cathode circuit. No voltage gain. Inherently
flat frequency response ! Much more linear.


** A power triode inherently has a flat response from DC to hundreds of KHz.

With a *pure resistive* load - yes.... The impedance presented by a
transformer
loaded with a loudspeaker is *anything but* pure resistive however !

With a reactive load in the plate circuit the gain will be approx proportional
to the load Z. Hence any variation in Z leads to a variation in gain and hence
changes in the amplitude/frequncy repsonse.

That's *basic* stuff !

The same is true of transistors, just substitute the word collector for plate.

Regards,

John Byrns

If transistor amps were collector loaded - then yes.

Most audio SS amps today use an emitter follower output. That arrangement simply
doesn't care about the load Z. It's important that the emitter follower has enough
current gain though ( use darlingtons or triples ) or the load might still 'reflect'
back to the previous voltage gain stage and possibly cause stability problems (
usually phase angle related ).

Those that do have a collector load use serious nfb to correct the issue of response
errors.

Graham

"Poopie ****wit Bear"

Those that do have a collector load use serious nfb to correct the issue
of response
errors.


** WRONG.

The use of local and loop NFB reduces non-linear distortion and lowers
output impedance.

A absolute fool and fake like Poopie, who will not admit his errors, is
condemned forever to repeat them.




........... Phil

John Stewart wrote:
Pooh Bear wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

Graham

This has been posted for a whole 12 hours already & Bret Ludwig has not
shown us a better way. Could he actually be stymied? Probably not! Watch
for another fog of bull****!

It has been pretty well described by your favorite guru, Norman
Crowhurst. Just read his (quite good) articles in the Audio
Anthologies.

Phil Allison wrote:

"Poopie ****wit Bear"

Those that do have a collector load use serious nfb to correct the issue
of response
errors.

** WRONG.

The use of local and loop NFB reduces non-linear distortion and lowers
output impedance.

And lowering output impedance reduces response errors too !


Hence I was *CORRECT*.

Why are you playing the court jester ?

Who rattled your damn cage anway ?

Graham

On Mon, 19 Dec 2005 06:46:22 +0000, Pooh Bear
wrote:

The hilarity is that the SET ( and indeed P-P ) output can only begin to have
a *flat frequency response* and also lowish output Z by using serious NFB.

A better way to say this is that no conventional amplifier
can begin to have a flat frequency response and lowish output Z
without serious NFB *except* a properly designed SET (or
a modern class-A MOSFET follower design).

The difference in perspective defines the gulf of perception.

Always the best,

Chris Hornbeck

"Angels are, dreaming of you" - Sonic Youth

Bret Ludwig wrote:

John Stewart wrote:
Pooh Bear wrote:

Has anyone ever used a triode configured as a *cathode follower* for a
speaker output ?

It would need tons of grid drive volts but I see several possible
options to achieve that.

The result would have all the perceived benefits of classic single
ended Class A operation together with a better amplifer-loudspeaker
interface and much lower distortion than the classic plate / anode
loaded circuit.

Graham

This has been posted for a whole 12 hours already & Bret Ludwig has not
shown us a better way. Could he actually be stymied? Probably not! Watch
for another fog of bull****!

It has been pretty well described by your favorite guru, Norman
Crowhurst. Just read his (quite good) articles in the Audio
Anthologies.

What did he have to say ?

Graham

Phil Allison wrote:

"Poopie Bear"

** Graham Stevenson .....

Dead meat.

You're an idiot Allison.

"Chris Hornbeck" wrote in message
...
On Mon, 19 Dec 2005 06:46:22 +0000, Pooh Bear
wrote:

The hilarity is that the SET ( and indeed P-P ) output can only begin to
have
a *flat frequency response* and also lowish output Z by using serious NFB.

A better way to say this is that no conventional amplifier
can begin to have a flat frequency response and lowish output Z
without serious NFB *except* a properly designed SET (or
a modern class-A MOSFET follower design).

**ALL amplifiers use NFB. Every single one.



--
Trevor Wilson
www.rageaudio.com.au

"Pooh Bear"

= Graham Stevenson

= CRIMINAL PSYCHO






.......... Phil

Some Lying, Autistic, CRIMINAL ****WIT called Graham Stevenson


** WRONG.

The use of local and loop NFB reduces non-linear distortion and lowers
output impedance.

A absolute fool and fake like Poopie, who will not admit his errors, is
condemned forever to repeat them.




........... Phil