S.E. otl amp
Adam Stouffer
Has anyone had good results using the russain 6C33C-B in a single ended
otl amplifier?
Gus
Mike Glantz
Adam Stouffer wrote:
>
> Has anyone had good results using the russain 6C33C-B in a single ended
> otl amplifier?
How does one design a single-ended amp without an output transformer?
What mechanism removes the DC current which a single-ended circuit
requires to flow in the load?
--
Mike Glantz Rank Xerox Research Centre/Grenoble
rri...@aol.com
<<How does one design a single-ended amp without an output transformer?
What mechanism removes the DC current which a single-ended circuit
requires to flow in the load?>>
Use the transformer just as a dc current load, but don't hook the speaker
up to the secondary. That is, just use it as an inductor, which will
remove the dc current in a dual manner to a series capacitor removing the
dc voltage.
It could probably work, it's an interesting question.
Mark and Jan Gilmore
So where do you hook up the speaker?
mark
Brendan Biever
In article <328964...@pitt.edu>, te...@mail.sgi.net says...
>
>Has anyone had good results using the russain 6C33C-B in a single ended
>otl amplifier?
>
GO to this URL to see an OTL schematic for the 6C33:
http://www.mic.go.kr/~jhlee/6cotl.gif
Regards,
Brendan Biever
Mike Glantz
<<How does one design a single-ended amp without an output transformer?
What mechanism removes the DC current which a single-ended circuit
requires to flow in the load?>>
Use the transformer just as a dc current load, but don't hook the speaker
up to the secondary. That is, just use it as an inductor, which will
remove the dc current in a dual manner to a series capacitor removing the
dc voltage.
It could probably work, it's an interesting question.
Yes, it will work, but it's not an "output transformerless" amp in
the intended sense of the term. The reason is that one wants to
eliminate the xfmr to avoid the non-linearities which they usually
introduce (due to stray capacitance, leakage inductance, core
saturation, etc). If you use any inductor, you get most of these
problems, *especially* in a single-ended output stage, where
core saturation is a big issue.
So I'm still curious: how does one design a single-ended output
stage without a transformer or inductor?
(hint: I know at least one way this can be done, but you'd get almost
no output power and almost all heat, and you'd still have a big capacitor
in the signal path)
rri...@aol.com
<<Yes, it will work, but it's not an "output transformerless" amp in
the intended sense of the term. The reason is that one wants to
eliminate the xfmr to avoid the non-linearities which they usually
introduce (due to stray capacitance, leakage inductance, core
saturation, etc). If you use any inductor, you get most of these
problems, *especially* in a single-ended output stage, where
core saturation is a big issue.>>
True, but you do get rid of several hundred feet of wire in series with
the signal, so it should be substantially better. Also, since you're not
using the secondary, you can put more primary turns on and stay away from
saturation problems better, and many of the conventional parasitic
problems of a transformer won't affect the signal. (leakage, skin effect,
proximity)
Not that it's a great way to go. Seems like a lot of work for not much
power.
And since you can't get much power anyway, maybe it would be OK to run
some dc bias current through the speaker?
mark gilmore
Perhaps I wasn't as clear as I'd hoped. Where and how do you hook up
the speaker? It seems that a direct coupled approach is out of the
question. This may not be the design goal of all OTL aficionados, but I
think one of the greatest advantages of OTL topology is served by direct
coupling to the load.
mark
Mike
>>Use the transformer just as a dc current load, but don't hook the speaker
>>up to the secondary. That is, just use it as an inductor, which will
>>remove the dc current in a dual manner to a series capacitor removing the
>>dc voltage.
>>It could probably work,
It will work. LC coupling is as old as the hills.
When B+ actually came from a battery, and therefore was rather expensive, LC
coupling was king at least for interstage coupling. HF chokes and LF chokes
occupied a good deal of the real estate on early breadboard radios. RC
coupling is obviously much less efficient on B+ current, but far cheaper to
execute.
However, the key must surely be that LC coupling to a loudspeaker would not
give any form of impedance matching and therefore power transfer would be
lousy. Even with a low impedance triode like the 6C33C-B you might be lucky
to get 3 watts into 8 ohms. However, if you needed a 250 Watt amplifier, that
doubled as a room heater, then 10 6C33's in parallel would match very nicely
into 8 ohms!
Mike
Jaehong Lee
Adam Stouffer <te...@mail.sgi.net> wrote in article
<328964...@pitt.edu>...
> Has anyone had good results using the russain 6C33C-B in a single ended
> otl amplifier?
>
> Gus
>
yes cHECK HTTP://MIC.GO.KR/~JHLEE/ FOR DETAILS.
SEE PHOTO,CIRCUITSETC FOR 6C33C-B OTL
mark gilmore
I think I understand the problem now. We've failed to adequately define
the concepts of "direct coupled" and "single ended". If we're willing
to ignore these requirements, the field of possibilities are wide open.
mark
Kurt Strain
Mike Glantz (mike....@grenoble.rxrc.xerox.com) wrote:
: Brendan Biever wrote:
: > GO to this URL to see an OTL schematic for the 6C33:
: uh, Brendan, this circuit isn't a single-ended design, it's a
: classic Futterman "push-push".
: --
: Mike Glantz Rank Xerox Research Centre/Grenoble
You can replace V5 of this circuit with a resistor for single-ended output.
You can then use that tube in parallel with V6 for more reasonable power.
The circuit would then have to operate class A, and that would mean the
total current would be about 1A. The load resistor would then need to be
about 180 ohms, 200W. A more reasonable thing to do would be to lower the
+180V to about +40V and then the resistor would be 40 ohms, 50W. The output
power would then be about 4W into an 8 ohm load. The damping factor with
20 dB feedback would be about 2. The power consumption would be about
220 W, about 1.8% efficiency.
Also, when the tube drifts or shorts or opens, expect DC to possibly
destroy the woofer before the fuse blows (maybe). You will need a delayed
turn-on relay switch to the speakers for warm-up.
You can see the limitations of SE OTL at this point.
Kurt
Mike Glantz
Mike Glantz
Jaehong Lee wrote:
> yes cHECK HTTP://MIC.GO.KR/~JHLEE/ FOR DETAILS.
> SEE PHOTO,CIRCUITSETC FOR 6C33C-B OTL
Well, this page can't be accessed, so we can't say for sure,
but I'll be very surprised if the circuit is a *single-ended*
OTL amp.
Mike Glantz
mark gilmore wrote:
> I think I understand the problem now. We've failed to adequately define
> the concepts of "direct coupled" and "single ended". If we're willing
> to ignore these requirements, the field of possibilities are wide open.
Correct. If I redefine "transistor" to be "tube", "push-pull" to be
"single-ended", "moby transformer" to be "OTL", and "Aiwa" to be
"Ongaku", then virtually all audiophiles will consider my system to be
utterly state-of-the-art. If I define my income to be $10,000,000/year,
the IRS will come a-knockin'.
But if we accept the more commonly used meanings of "SE" and "OTL",
we will see that, as Kurt described, the idea of "SE OTL" is fantasy.
David Lawrence
Why not use a big capacitor to block the D.C. as has
often been done with push-pull? Capacitors of course
have disadvantages, but maybe it would be worth
trying.
Mike Glantz
yes, they do have some disadvantages, but they at least
come closer to satisfying the intent of OTL. Just one
question: what's the static load on the output tube?
The speaker will not be a part of the DC load.
mark gilmore
> Mike Glantz Rank Xerox Research Centre/Grenoble
Very well said, Mike. There is intelligent life on the the planet OTL.
mark
jby...@kiwi.pyrotechnics.com
In article <56d7el$6...@canyon.sr.hp.com>,
ku...@sr.hp.com (Kurt Strain) wrote:
> You can replace V5 of this circuit with a resistor for single-ended output.
> You can then use that tube in parallel with V6 for more reasonable power.
> The circuit would then have to operate class A, and that would mean the
> total current would be about 1A. The load resistor would then need to be
> about 180 ohms, 200W. A more reasonable thing to do would be to lower the
> +180V to about +40V and then the resistor would be 40 ohms, 50W. The output
> power would then be about 4W into an 8 ohm load. The damping factor with
> 20 dB feedback would be about 2. The power consumption would be about
> 220 W, about 1.8% efficiency.
Hi Kurt,
Your calculated damping factor of 2 seems way too high. By my math, the
damping factor with 20 dB feedback would be 11, not 2. I am assuming that
"20 dB feedback" means a 20 dB gain reduction when the feedback loop is
closed. I am further assuming that an 8 Ohm load is connected to the output
of the amplifier while the gain reduction is being measured. I think this
connection of the load is what is causing our numbers to differ, it looks
like you measured 20 dB feedback with no load connected. I am not sure which
is the correct way to measure the feedback, comments anyone?
Regards,
John Byrns
-----------------------------------------------------------------------
This article was posted to Usenet via the Posting Service at Deja News:
http://www.dejanews.com/ [Search, Post, and Read Usenet News]
C.M.
Mike Glantz wrote:
>
> Jaehong Lee wrote:
> > yes cHECK HTTP://MIC.GO.KR/~JHLEE/ FOR DETAILS.
> > SEE PHOTO,CIRCUITSETC FOR 6C33C-B OTL
>
> Well, this page can't be accessed, so we can't say for sure,
> but I'll be very surprised if the circuit is a *single-ended*
> OTL amp.
> Mike Glantz Rank Xerox Research Centre/Grenoble
What is it you would like to know??????
C.M.
--
C.M.
Frank Harrison
Mike Glantz <mike....@grenoble.rxrc.xerox.com> wrote:
>
> Adam Stouffer wrote:
> >
> > Has anyone had good results using the russain 6C33C-B in a single ended
> > otl amplifier?
>
> What mechanism removes the DC current which a single-ended circuit
> requires to flow in the load?
> Mike Glantz Rank Xerox Research Centre/Grenoble
I don't know what these other guys are doing, but one possibility
would be to use RC coupling. Not very efficient, but it should work.
Frank
Frank Harrison
Mike Glantz <mike....@grenoble.rxrc.xerox.com> wrote:
>
> Adam Stouffer wrote:
> >
> > Has anyone had good results using the russain 6C33C-B in a single ended
> > otl amplifier?
>
> How does one design a single-ended amp without an output transformer?
> What mechanism removes the DC current which a single-ended circuit
> requires to flow in the load?
> --
> Mike Glantz Rank Xerox Research Centre/Grenoble
I don't know what these other guys are doing, but one possibility
Mike Glantz
Frank Harrison wrote:
> Mike Glantz <mike....@grenoble.rxrc.xerox.com> wrote:
> > What mechanism removes the DC current which a single-ended circuit
> > requires to flow in the load?
>
Correct. Kurt Strain suggested this as one possibility. In fact, he
suggested running the resistor to a negative rail so that the cathode
would be approximately at zero, thus eliminating the need for the cap.
As both you and Kurt pointed out "not very efficient". In fact, this
is an understatement. The efficiency is so pathetic that it's no mystery
why you *never* see a resistive load in an output circuit. You either
see
an inductor (or xfmr) or direct coupling to the speaker.
Neil brown
In article <56hsnd$4...@solaris.cc.vt.edu>, Frank Harrison
<rama...@vt.edu> wrote:
> Mike Glantz <mike....@grenoble.rxrc.xerox.com> wrote:
> >
> > Adam Stouffer wrote:
> > >
> > > Has anyone had good results using the russain 6C33C-B in a single ended
> > > otl amplifier?
> >
> > How does one design a single-ended amp without an output transformer?
> > What mechanism removes the DC current which a single-ended circuit
> > requires to flow in the load?
> > --
> > Mike Glantz Rank Xerox Research Centre/Grenoble
>
> I don't know what these other guys are doing, but one possibility
> would be to use RC coupling. Not very efficient, but it should work.
>
> Frank
Could you bridge it and use no output capacitor or transformer?
Mike Glantz
> I don't know what these other guys are doing, but one possibility
> would be to use RC coupling. Not very efficient, but it should work.
Could you bridge it and use no output capacitor or transformer?
Bridge what? What would this circuit look like?
Frank Harrison
nbr...@whispa.co.nz (Neil brown) wrote:
>
> > Could you bridge it and use no output capacitor or transformer?
> >
> > Bridge what? What would this circuit look like?
>
> (lots in parallel) and bridge 2 channels using no DC blocking device. I
> can't come up with a good way to track tube warm up though so the outputs
> don't drift differently.
>
> regards
> NEil
If you bridge it, it's not single ended, anymore, it's push-pull.
Frank
Mike Glantz
> > > Could you bridge it and use no output capacitor or transformer?
> > > Bridge what? What would this circuit look like?
> > Couldn't you in theory take something like a mu follower using the el86
> > (lots in parallel) and bridge 2 channels using no DC blocking device. I
> > can't come up with a good way to track tube warm up though so the outputs
> > don't drift differently.
> If you bridge it, it's not single ended, anymore, it's push-pull.
What he said. In fact, the resulting circuit is very similar to the
Wiggins Circlotron currently used in the Atma-Sphere amps. A very
elegant direct-coupled push-pull output stage.
Neil brown
In article <urvib4v...@pelee.grenoble.rxrc.xerox.com>, Mike Glantz
<Mike....@grenoble.rxrc.xerox.com> wrote:
> In article <nbrown-1711...@wh4-14.wave.co.nz>
nbr...@whispa.co.nz (Neil brown) writes:
>
> > I don't know what these other guys are doing, but one possibility
> > would be to use RC coupling. Not very efficient, but it should work.
>
> Could you bridge it and use no output capacitor or transformer?
>
> Bridge what? What would this circuit look like?
Couldn't you in theory take something like a mu follower using the el86
(lots in parallel) and bridge 2 channels using no DC blocking device. I
can't come up with a good way to track tube warm up though so the outputs
don't drift differently.
regards
NEil
Andrea Ciuffoli
I am a home made audio builder and at moment I have start some
test on a OTL with 6C33C-B.
Now I am using this tube as cathode follower (2 tubes for channel) with
125v dc and 4.7 + 65 ohm on cathode and the result it's incredible also
if some my simulation with spice3f4 give a output resistenca of 5 ohm
and a max power of about 5w.
for any question contact me.
--
email: ciuf...@saritel.it
Andrea Ciuffoli
via degli Zingari 50
00184 Rome ITALY
Andrea Ciuffoli
Adam Stouffer wrote:
>
> Has anyone had good results using the russain 6C33C-B in a single ended
> otl amplifier?
>
I am working on 6C33C-B power cathode follower (3 tubes for channel)
with 125v dc and 4.7 + 65 ohm on cathode and the result it's incredible
also if some my simulation with spice3f4 give a output resistenca of 4
ohm and a max power of about 5w.
This is one of the very few OTL single ended.
contact me for any information.
Scott Dorsey
In article <32CB8B33...@saritel.it> Andrea Ciuffoli <ciuf...@saritel.it> writes:
>Adam Stouffer wrote:
>>
>> Has anyone had good results using the russain 6C33C-B in a single ended
>> otl amplifier?
>>
>> Gus
>
>I am working on 6C33C-B power cathode follower (3 tubes for channel)
>with 125v dc and 4.7 + 65 ohm on cathode and the result it's incredible
>also if some my simulation with spice3f4 give a output resistenca of 4
>ohm and a max power of about 5w.
Ought to be a maximum power transfer Z of 24 ohms with three parallelled
6C33Cs.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Mike Glantz
>I am working on 6C33C-B power cathode follower (3 tubes for channel)
>with 125v dc and 4.7 + 65 ohm on cathode and the result it's incredible
>also if some my simulation with spice3f4 give a output resistenca of 4
>ohm and a max power of about 5w.
Interesting ... I thought we ran through this before, but I guess
somebody has come up with a solution: how do you keep DC from flowing
through the speaker in a SE OTL amp?
Scott Dorsey
Through a whomping big capacitor. Take a look at the Nelson Pass single
ended (well, SRPP really) transistor amp designs.
Kurt Strain
Mike Glantz (gla...@pelee.grenoble.rxrc.xerox.com) wrote:
: >I am working on 6C33C-B power cathode follower (3 tubes for channel)
: >with 125v dc and 4.7 + 65 ohm on cathode and the result it's incredible
: >also if some my simulation with spice3f4 give a output resistenca of 4
: >ohm and a max power of about 5w.
: Interesting ... I thought we ran through this before, but I guess
: somebody has come up with a solution: how do you keep DC from flowing
: through the speaker in a SE OTL amp?
1. Capacitor coupled.
2. Servo DC coupled.
3. Bridge circuit - Circlotron.
However:
In most these so-called "Single-Ended" amps, they are really "totem
pole" outputs. In the case of the circlotron that I've seen,
there are two "single-ended" cathode followers arranged in a floating
bridge to remove the need for a DC blocking cap. They are in
opposite phase - and that's Push-Pull.
In my mind, a true "SE amp" is Single Ended in it's simplest form;
i.e., there's one active amplifying device - the other is a load
(active or passive). With the White Cathode Follower, the device
is a totem pole "push-pull" output stage. It's just that the input
of the White Cathode Follower is Single Ended to start and the
output is Single Ended to finish. So some will call this a single
ended amp. Yet the totem pole stage serves to drive the output from
the top and from the bottom simultaneously - push from above and
pull from below or pull from above and push from below - like in
Push-Pull.
It is better to be specific. Most SE amps are tubed common
cathode output stages. Even the PASS Zen amp has a common source
output stage. But some amps are SE amps when they are FET common
drain output stages or tubes that are common anode. So there are
a lot of people using the SE label, with various actual topologies.
The White Cathode Follower can be part of this variation on a theme,
but that just shows that "SE" is not some stamp of approval by a
a bunch of "SE enthusiasts". Futterman amps had the White
Cathode follower output stage, I believe, and Julius Futterman
wasn't calling his OTL amps "SE-OTL" amps back then.
It's my opinion the core "SE movement" is basically derived in
general from using a directly heated triode in a common cathode
output stage with an air-gapped output transformer and little or
no global feedback. The more you deviate from that, the more
you stick your neck out when prominantly claiming "it's an SE amp."
My response to these designs that have this marketing label,
"Yeah? Do you mean it sounds like a typical directly heated triode
in a common cathode output stage with an air-gapped output
transformer and little or no global feedback, or is it just lacking
balanced inputs, or what?"
Kurt
obse...@rec.tubes
Is this really a signature?:
In article 47DB...@saritel.it:
>I am working on 6C33C-B power cathode follower (3 tubes for channel)
>with 125v dc and 4.7 + 65 ohm on cathode and the result it's incredible
>also if some my simulation with spice3f4 give a output resistenca of 4
>ohm and a max power of about 5w.
>
Dave Cigna
Kurt Strain <ku...@sr.hp.com> wrote:
[snip]
> In the case of the circlotron that I've seen,
> there are two "single-ended" cathode followers arranged in a floating
> bridge to remove the need for a DC blocking cap. They are in
> opposite phase - and that's Push-Pull.
No argument that this is push-pull. But I'm curious how this is coupled
to the speaker. Is this a pair of ordinary cathode followers, i.e. is the
load coupled through a big-assed cathode load resistor? Seems like it
would be necessary to dissipate a lot of power there. The Futterman design
(White cathode follower) does not have the elegance of symmetry, but it
doesn't require heating a pair of big power resistors.
Anybody care to try a pair of White cathode followers in a bridge
arrangement?
-- Dave Cigna
SVETENGR
Four years ago I tried this very scheme, using 2 6C33s in a parallel
cathode follower, capacitor coupled to the speaker.
It worked; I got 3 watts into 8 ohms at 5% distortion. The main problem
was that the bias point would drift up and down. 6C33s are great tubes but
don't seem to like fixed bias arrangements.
(Too bad the Ulyanovsk factory doesn't like to sell to Americans...)