6AS7 OTL Amp

[Michael Thompson]

Does anyone out there have a reprint of the 6AS7 OTL
amp from the February 1990 issue of Glass Audio they
could email me?
I am thinking about building this amp and have only a
partial schematic. If someone has built one I would
appreciate any comments about it. I was thinking of
using 6 - 6AS7's.

Thanks,
Michael

[BEAR]

Michael Thompson wrote:

More output tubes = lower output Z = more power = better at lower Z loads...

_-_-bear

--
_-_-bearlabs

http://www.bearlabsUSA.com
- Silver Lightning Interconnects -

[John Stewart]

See ABSE for the complete article.
I built one in 1956 designed by Dickie & Macovski.
That circuit is somewhat similar to this one.
It used 3 of 6AS7 & successfully drove a 16 ohm speaker.

Cheers, John Stewart

[Fred Nachbaur]

John Stewart wrote:
> See ABSE for the complete article.
> I built one in 1956 designed by Dickie & Macovski.
> That circuit is somewhat similar to this one.
> It used 3 of 6AS7 & successfully drove a 16 ohm speaker.
>
> Cheers, John Stewart

Good heavens... there really *isn't* anything new under the sun. It's
almost the exact topology that I've got planned for a monoblock using
four 6AV5's. Except that I will be using an OPT.

The other difference is that I'm planning to use an assymetrical
cathodyne phase splitter, to account for the different gains on the two
branches of the SRPP output stage. The Rozenblit design - as near as I
can make out - uses the usual symmetrical cathodyne; I can't see this
giving the best results, since the top of the totem pole is essentially
a cathode follower (common anode), whereas the bottom is a common
cathode arrangement.

Cheers,
Fred
--
+--------------------------------------------+
| Music: http://www3.telus.net/dogstarmusic/ |
| Projects, Vacuum Tubes & other stuff: |
| http://www.dogstar.dantimax.dk |
+--------------------------------------------+

[Jon Yaeger]

Fred,

I'm gathering parts to build the Rozenblitt design. I'd be MOST
interested in your recommendations.

Thanks!

- Jon

In article <3D5C32CA...@netscape.net>, Fred Nachbaur

[Fred Nachbaur]

Unless someone points out the error in my ways, it appears to me that
you'd want the plate resistor of the cathodyne to be on the order of
(mu) times that of the cathode resistor, to equalize the gains between
top and bottom.

This isn't to say that the Rozenblit design wouldn't work, obviously it
does because people have built it. However, it seems to me that the
contribution of the top tube to the output signal will be 1/mu that of
the bottom one.

It'll be awhile before I can actually try this out (too many other
things on the go right now) but once I do play around with it, I'll let
the results be known.

Cheers,
Fred

[Nick Sheldon]

Hi Fred and other RATs

One set of valves has a gain of one and a low output impedance, and the
other set a gain much higher than unity but also a higher output impedance,
cancelling out the gain. My guess is that the contributions are therefore
fairly similar.

Best wishes

Nick Sheldon

Fred Nachbaur wrote in message <3D5DB29C...@netscape.net>...

[Tim Williams]

"Nick Sheldon" <nick.s...@sheldonassociate.demon.co.uk> wrote in message
news:1029569260.15625....@news.demon.co.uk...

> One set of valves has a gain of one and a low output impedance, and the
> other set a gain much higher than unity but also a higher output
impedance,
> cancelling out the gain. My guess is that the contributions are therefore
> fairly similar.

(Ahem, CFs always have less than unity gain..)
I'd think that the total lack of NFB (especially 100% FB as in a CF)
would just beg for distortion.

Besides, when that side starts pulling down hard (I assume it's in
class AB), it'll be carrying it *all*, and with an obscenely low load
such as this, and ZNFB, it'll distort like crazy.

So just swap the balanced drive lines on your Futterman circuit, OK? ;)

Tim

--
"Yeah, and I'm not easily impressed. Whoa, a blue car!"
- Homer Simpson

[John Stewart]

There is both +ve & -ve fb in Rozenblit's circuit.
Referring to Figure One the -ve fb is labeled as that beside
the 1k resisitor R22 & does it's thing in a regular way.
The +ve fb is by way of the output signal applied to the bottom
end of the split load phase inverter. That fixes the unequal drive
requirement problem as ID'ed by Fred N.
Bruce covers that at the bottom right hand corner on page 5
of his article & attributes the fix to Julius Futterman.

Cheers to all, John Stewart

[Fred Nachbaur]

Aha! Thanks, John.

I found it difficult to read the posted version, even component values
on the schematic were largely a matter of guessing. I'd really like to
take this in more fully, would it be possible to repost this in higher
resolution? (Asking a lot, I know... so if you don't care to be
bothered, I'll certainly understand.)

Cheers,
Fred

[John Stewart]

Fred Nachbaur wrote:

> Aha! Thanks, John.
>
> I found it difficult to read the posted version, even component values
> on the schematic were largely a matter of guessing. I'd really like to
> take this in more fully, would it be possible to repost this in higher
> resolution? (Asking a lot, I know... so if you don't care to be
> bothered, I'll certainly understand.)

Done. See ABSE. Cheers, John

[Fred Nachbaur]

John Stewart wrote:
>
> Fred Nachbaur wrote:
>
>
>>Aha! Thanks, John.
>>
>>I found it difficult to read the posted version, even component values
>>on the schematic were largely a matter of guessing. I'd really like to
>>take this in more fully, would it be possible to repost this in higher
>>resolution? (Asking a lot, I know... so if you don't care to be
>>bothered, I'll certainly understand.)
>
>
> Done. See ABSE. Cheers, John

Got 'em! Ah, much easier on the old eyes. Thanks a million!

Cheers,
Fred

[John Stewart]

See ABSE for another version of the OTL Amp.
This is the one I built in 1956. It's long gone!!!
Was originally published in Audio mag, June 1954.
Cheers, John

[Michael Thompson]

What is the reference to "ABSE"? I would like to take
a look at the schematic.

Thanks,
Michael

"John Stewart" <jh.st...@sympatico.ca> wrote in message
news:3D5F9257...@sympatico.ca...

[Fred Nachbaur]

alt.binaries.schematics.electronic

[Nick Sheldon]

I'm really not at all convinced by what has been said about unequal drive
requirements and their "correction".

The following applies to one PP OTL pair of 6AS7s working in class B - we
have to start somewhere :-)

My data lists the 6AS7 as having an ra of 280 ohms and a gm of 7.1mA/v. This
implies a voltage gain of 2 into an infinite impedance, or of unity into 280
ohms. Into 140 ohms it would be around 0.67 and into 31 ohms about 0.2.

In common anode mode, the gain is exactly unity into an infinite impedance,
but as the output impedance is 1/gm = 140 ohms, we'd get 0.5 into 140 ohms
and about 0.18 into 31 ohms.

In short, as the load impedance falls, the two gains converge, and even at
31 ohms the second harmonic distortion this slight mismatch in figures would
generate is less than some valves do working in SE.

Apologies to all RATS concerned if this is an incorrect analysis, but
something makes these common anode + common cathode circuits work!

Regarding the positive feedback, has anyone tried to analyse this
mathematically? Some of Futterman's circuits 1. did not behave as described
and 2. did not work as described.

On an entirely different topic, have any RATs tried 211s in PP/PPP? I would
have thought 6 would *just* make for a kilowatt amp without exceeding
ratings, as long as the biasing/load/HT were correct.

Nick Sheldon

, John Stewart wrote in message <3D5E32D8...@sympatico.ca>...