The "KISS Amp WE300B SE Ultrafi" (1) has been finalized. This is a
project that was designed step by step on rec.audio.tubes. The
amplifier is single-ended, zero negative feedback, built with Western
Electric signal and power tubes and a Mullard tube rectifier. There are
(counting by the Gaincard method which leaves the attenuator and power
supply out of the count!) six components in the signal path: 3
resistors, a cap and two tubes. It is a truly silent amplifier (It
sacrifices half the power for that silence) and in particular the
output is free of any odd harmonics. The circuit with a single WE 417A
tube doing both input and driver duty is at:
http://members.lycos.co.uk/fiultra/T...trafi-crct.jpg
The companion, somewhat simpler and less expensive "standard good" 300B
SE amplifier in the ultrafi set, the 300B SE amplifier "Populaire" with
two stages of 6SN7 tubes, has been previously published at
http://members.lycos.co.uk/fiultra/T44bis-'Populaire'-crct.jpg
Both amplfiers are described in The KISS Amp section of Jute on Amps
http://members.lycos.co.uk/fiultra/T...mp%20INDEX.htm
From the introduction: "THE KISS AMP 300B project is an attempt to take
the tube amp designer by the hand and lead him through all the highways
and byways of designing an ultrafi amp, including the thought
processes, the math and the development. It is half-engineering course,
half metaphysics, half bloodyminded prejudice and alltogether
infuriatingly complex because the simplicity of KISS has a very high
price. It has its own index page for the text and another for the
illustrations. You will find the majority of the most useful core
articles in the old Jute on Amps site in The KISS Amp 300B files,
rearranged and rewritten to make the interdependence of tube design
decisions manifest."
Suitable speakers for these amplifiers are also shown, the expensive
first
http://members.lycos.co.uk/fiultra/K...20T91HWAF3.jpg
and then the almost ridiculously cheap
http://members.lycos.co.uk/fiultra/K...Impresario.jpg
More of my amplifiers, both solid state and tube (valve) are at Jute on
Amps
http://members.lycos.co.uk/fiultra/JUTE%20ON%20AMPS.htm
You are also invited to visit my overall netsite which has information
on a couple of of my professions, as a novelist and as a typographere,
and some of my other interests: music both live and recorded which I
reviewed for many years for papers around the world, watches,
bicycling, cooking, etc.
http://members.lycos.co.uk/fiultra/

Many helped along the way by a cheery "Thanks!", or a private note
telling me not to succumb before the assaults of the silicon slime and
the other useless egomaniacs who create nothing except flamewars on the
very few remaining productive members of RAT. (For proof, read on in
the rest of this thread.)

Among those who provided relevant technical advice I must in particular
and in alphabetical order thank Steve Bench, John Byrns, Doug Bannard
and Patrick Turner. Even more admirable than their knowledge is their
patience! Controversial opinions, errors and omissions are mine, of
course.

Pay due respect to high voltage and live long!

Andre Jute
Sauvitor in modo, fortiter in res
Visit Jute on Amps at http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review

Andre Jute wrote:
The "KISS Amp WE300B SE Ultrafi" (1) has been finalized.

OK... I can accept that. Now, has it been built? More importantly, does
it work?

Given the overall accuracy and credibility of the OP, I would believe
the latter two only if independently witnessed and photographed
together with a copy of a recent newspaper headline. And even then I
would be skeptical.

Peter Wieck
Wyncote, PA

In article .com,
"Andre Jute" wrote:

The "KISS Amp WE300B SE Ultrafi" (1) has been finalized. This is a
project that was designed step by step on rec.audio.tubes. The
amplifier is single-ended, zero negative feedback, built with Western
Electric signal and power tubes and a Mullard tube rectifier. There are
(counting by the Gaincard method which leaves the attenuator and power
supply out of the count!) six components in the signal path: 3
resistors, a cap and two tubes.

Hi Andre,

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

In any case, six components give or take for the signal to pass through
sounds like too many for good sonics. In my "7119 PP Potato 2 Minimal"
amp the signal passes through only two components depending on exactly
how you count.

It is a truly silent amplifier (It
sacrifices half the power for that silence) and in particular the
output is free of any odd harmonics.

What is the power output of your "KISS Amp WE300B SE Ultrafi"?

The circuit with a single WE 417A
tube doing both input and driver duty is at:
http://members.lycos.co.uk/fiultra/T...trafi-crct.jpg


Regards,

John Byrns

--
Surf my web pages at, http://fmamradios.com/

On Sun, 14 Jan 2007 13:23:13 -0600, John Byrns
wrote:

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

A much more interesting number is those components that can do
something to the signal. Those are either the ones in the direct
signal path or those that have a signal voltage across them. Some
components in the signal path have essentially zero signal voltage
across them and are thereby incapable of altering the signal.

Now, counting properly, components in the direct signal path are

Input pot
4 by 220R grid stopper
First valve
Coupling cap
220R grid stopper
Second valve
transformer
47k feedback r
200uf B1 decoupler

That's 12

The signal handling components with first order effects are

Battery
10 anode load
200uF B2 decoupler
47k grid leak
56uf cathode decoupler
1k cathode load

That makes 18 total.

Buggered if I can see the relevance of the number though. Any signal
that hits this amp has already been through a few hundred other
components. Of course the whole thing makes a bit of sense when you
consider that the components in this amp are doing a
disproportionately huge amount of damage to the signal. Which they
are. Ultrafi is an interestingly ironic name, don't you think?

d
--
Pearce Consulting
http://www.pearce.uk.com

On Sun, 14 Jan 2007 20:07:06 GMT, (Don Pearce)
wrote:

On Sun, 14 Jan 2007 13:23:13 -0600, John Byrns
wrote:

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

A much more interesting number is those components that can do
something to the signal. Those are either the ones in the direct
signal path or those that have a signal voltage across them. Some
components in the signal path have essentially zero signal voltage
across them and are thereby incapable of altering the signal.

Now, counting properly, components in the direct signal path are

Input pot
4 by 220R grid stopper
First valve
Coupling cap
220R grid stopper
Second valve
transformer
47k feedback r
200uf B1 decoupler

That's 12

The signal handling components with first order effects are

Battery
10 anode load
200uF B2 decoupler
47k grid leak
56uf cathode decoupler
1k cathode load

That makes 18 total.

Buggered if I can see the relevance of the number though. Any signal
that hits this amp has already been through a few hundred other
components. Of course the whole thing makes a bit of sense when you
consider that the components in this amp are doing a
disproportionately huge amount of damage to the signal. Which they
are. Ultrafi is an interestingly ironic name, don't you think?

d

Sorry, miscounted. I hadn't spotted the two 100 ohm resistors and 100
ohm pot forming the series feedback network on the output valve were
carrying signal. Hang - this amp is not meant to have any feedback -
what are they doing there?

d

--
Pearce Consulting
http://www.pearce.uk.com

Don Pearce wrote:

Some components in the signal path have essentially zero signal voltage
across them and are thereby incapable of altering the signal.

Like coupling caps - LOL ! Who mentioned Teflon caps ?

Graham

Andre Jute wrote:
The "KISS Amp WE300B SE Ultrafi" (1) has been finalized. This is a
project that was designed step by step on rec.audio.tubes. The

This may have been covered, in which case, sorry, but why is the signal
ground point taken back at the rectifier? Doesn't that make the output
stage signal loop include the choke and snubbers in the supply?

--
Nick

Don Pearce wrote:

On Sun, 14 Jan 2007 20:07:06 GMT, (Don Pearce)
wrote:

On Sun, 14 Jan 2007 13:23:13 -0600, John Byrns
wrote:

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

A much more interesting number is those components that can do
something to the signal. Those are either the ones in the direct
signal path or those that have a signal voltage across them. Some
components in the signal path have essentially zero signal voltage
across them and are thereby incapable of altering the signal.

Now, counting properly, components in the direct signal path are

Input pot
4 by 220R grid stopper
First valve
Coupling cap
220R grid stopper
Second valve
transformer
47k feedback r
200uf B1 decoupler

That's 12

The signal handling components with first order effects are

Battery
10 anode load
200uF B2 decoupler
47k grid leak
56uf cathode decoupler
1k cathode load

That makes 18 total.

Buggered if I can see the relevance of the number though. Any signal
that hits this amp has already been through a few hundred other
components. Of course the whole thing makes a bit of sense when you
consider that the components in this amp are doing a
disproportionately huge amount of damage to the signal. Which they
are. Ultrafi is an interestingly ironic name, don't you think?

d

Sorry, miscounted. I hadn't spotted the two 100 ohm resistors and 100
ohm pot forming the series feedback network on the output valve were
carrying signal. Hang - this amp is not meant to have any feedback -
what are they doing there?

LOL !

You nearly had me going there. That's the 'hum trim' control.

Graham

On Sun, 14 Jan 2007 21:15:23 +0000, Eeyore
wrote:



Don Pearce wrote:

On Sun, 14 Jan 2007 20:07:06 GMT, (Don Pearce)
wrote:

On Sun, 14 Jan 2007 13:23:13 -0600, John Byrns
wrote:

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

A much more interesting number is those components that can do
something to the signal. Those are either the ones in the direct
signal path or those that have a signal voltage across them. Some
components in the signal path have essentially zero signal voltage
across them and are thereby incapable of altering the signal.

Now, counting properly, components in the direct signal path are

Input pot
4 by 220R grid stopper
First valve
Coupling cap
220R grid stopper
Second valve
transformer
47k feedback r
200uf B1 decoupler

That's 12

The signal handling components with first order effects are

Battery
10 anode load
200uF B2 decoupler
47k grid leak
56uf cathode decoupler
1k cathode load

That makes 18 total.

Buggered if I can see the relevance of the number though. Any signal
that hits this amp has already been through a few hundred other
components. Of course the whole thing makes a bit of sense when you
consider that the components in this amp are doing a
disproportionately huge amount of damage to the signal. Which they
are. Ultrafi is an interestingly ironic name, don't you think?

d

Sorry, miscounted. I hadn't spotted the two 100 ohm resistors and 100
ohm pot forming the series feedback network on the output valve were
carrying signal. Hang - this amp is not meant to have any feedback -
what are they doing there?

LOL !

You nearly had me going there. That's the 'hum trim' control.

Graham

It is also negative feedback, whether or not that is the prime
purpose.

d

--
Pearce Consulting
http://www.pearce.uk.com

John Byrns wrote:
In article .com,
"Andre Jute" wrote:

The "KISS Amp WE300B SE Ultrafi" (1) has been finalized. This is a
project that was designed step by step on rec.audio.tubes. The
amplifier is single-ended, zero negative feedback, built with Western
Electric signal and power tubes and a Mullard tube rectifier. There are
(counting by the Gaincard method which leaves the attenuator and power
supply out of the count!) six components in the signal path: 3
resistors, a cap and two tubes.

Hi Andre,

Can you explain in more detail the "Gaincard method" of counting
components in the signal path? Ignoring the fact that you don't seem to
be including the output transformer in your count, I still can't
duplicate your count for the "KISS Amp WE300B SE Ultrafi"

Let's take Don Pearce's list, which, whatever his motives, seems
reasonable enough to me:

Input pot
4 by 220R grid stopper
First valve
Coupling cap
220R grid stopper
Second valve
transformer
47k feedback r (Pearce must be referring to the grid leak resistor)
200uf B1 decoupler

Now, the Gaincard method, which I checked into when I designed a
minimum silicon amp
http://members.lycos.co.uk/fiultra/K...dre%20Jute.htm
http://members.lycos.co.uk/fiultra/K...dre%20Jute.htm
while I waited for you and Patrick to finish educating Pinkerton so
that he could enter the design contest in which he disgraced himself,
counts only components directly in the signal line from input to
output, and ignores the attenuator and the power supply. That would
leave one with:

1 grid stopper (4x 220R grid stoppers in parallel, counted as one as
the gainclone boys do)
First valve
Coupling cap
220R grid stopper
Second valve
transformer

Hmm. That's six components by the Gaincard method; I must have
subliminally revolted against such wishful thinking and defiantly
included the attenuator as well. Still, even if I include the
attenuator, that's two less than the Gancard tally of nine. One might
argue that a grid stopper soldered to the socket itself is a continuous
part of a component which is not counted, and thus not count it. The
truth is that I don't really care how many components I use, as long as
they are the right number and quality for the sound I want; didn't
Einstein say a thing should be as simple as is necessary but no
simpler. The key thing is to know when too much has arrived and to step
back from it.

In any case, six components give or take for the signal to pass through
sounds like too many for good sonics. In my "7119 PP Potato 2 Minimal"
amp the signal passes through only two components depending on exactly
how you count.

Showoff! Reeling in shock at such parsimony, I went looking for the
circuit on your site. When will you be publishing it? It sounds
fascinating.

It is a truly silent amplifier (It
sacrifices half the power for that silence) and in particular the
output is free of any odd harmonics.

What is the power output of your "KISS Amp WE300B SE Ultrafi"?

3.8W. I know, it doesn't look like very much from a 300B, but that is
the price of silence, of loading the plate with a 5K6 primary impedance
to get the noise down and the excellent harmonic distribution which you
first pointed out back when you compared the output stage of an earlier
version of the more affordable 6SN76SN7300B version of this amp, the
T44 Populaire,
http://members.lycos.co.uk/fiultra/T44bis-'Populaire'-crct.jpg
with the amp entered against it in an earlier design competition, the
Bubbaland 300B.

The circuit with a single WE 417A
tube doing both input and driver duty is at:
http://members.lycos.co.uk/fiultra/T...trafi-crct.jpg

Regards,

John Byrns
--
Surf my web pages at, http://fmamradios.com/

Nice to see you back, John. I was thinking of discussing my next
project (a completely differential amp to drive electrostatic
earphones, what Stax calls "earspeakers on another group now that the
mouthfoamers have driven out so many of the capable RATs, but if you're
back perhaps we can inject some relevant interest. Mind you, my
dissection partner, yeah, all those years ago, had a macabre sense of
humour; he once told a female student, "Pass up that footpump so I can
inflate this cadaver and have sex with it." RAT's a bit like that these
days.

Andre Jute
Visit Jute on Amps at http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review