Gentlemen,

here is a link to a corrected schematic and article using that
"phase splitter with no name" (in German, though):

http://www.jogis-roehrenbude.de/Roeh...unktechnik.pdf

The 1M pot, according to the text, indeed is for adjusting the
drive voltages to the output stage to get maximum 2H cancelation.

Tom

--
The opposite of a correct statement is a false statement. But the
opposite of a profound truth may well be another profound truth.
- Niels Bohr

Hi Tom ,

Looks like a Mullard paraphase splitter with an extra C to me ....

Ronald .


"Tom Schlangen" schreef in bericht
...
Gentlemen,

here is a link to a corrected schematic and article using that
"phase splitter with no name" (in German, though):


http://www.jogis-roehrenbude.de/Roeh...unktechnik.pdf

The 1M pot, according to the text, indeed is for adjusting the
drive voltages to the output stage to get maximum 2H cancelation.

Tom

--
The opposite of a correct statement is a false statement. But the
opposite of a profound truth may well be another profound truth.
- Niels Bohr

http://www.jogis-roehrenbude.de/Roeh...unktechnik.pdf

The 1M pot, according to the text, indeed is for adjusting the
drive voltages to the output stage to get maximum 2H cancelation.


You could almost think of the inverter tube as a sort of op-amp
circuit. The 100K resistor from the plate of the first triode
feeds via the coupling cap the "inverting input" (grid) and
continues on thru the 180K and the trimpot to the "op-amp
output" (plate). The load on this plate is around the parallel
value of the plate resistor, the feedback resistor, and the
grid resistor of the output tube. About 25K. The open loop gain
of this stage is around 21, with the cathode resistor of 1200 ohms.
SO the "virtual zero" at the "inverting input" will see about -1/21
of the output signal. The trimpot can boost the gain a bit to
compensate for this, to get a closer inversion of a gain of -1.

On Thu, 10 Feb 2005 05:31:42 +0000, robert casey wrote:


http://www.jogis-roehrenbude.de/Roeh...unktechnik.pdf

The 1M pot, according to the text, indeed is for adjusting the drive
voltages to the output stage to get maximum 2H cancelation.


You could almost think of the inverter tube as a sort of op-amp circuit.
The 100K resistor from the plate of the first triode feeds via the
coupling cap the "inverting input" (grid) and continues on thru the 180K
and the trimpot to the "op-amp output" (plate). The load on this plate is
around the parallel value of the plate resistor, the feedback resistor,
and the grid resistor of the output tube. About 25K. The open loop gain
of this stage is around 21, with the cathode resistor of 1200 ohms. SO the
"virtual zero" at the "inverting input" will see about -1/21 of the output
signal. The trimpot can boost the gain a bit to compensate for this, to
get a closer inversion of a gain of -1.

So, is that why the 100k is there and why 0.5 has been chosen for the
coupling cap? Low input impedence into the "virtual ground"?

--
Mick
(no M$ software on here... :-) )
Web: http://www.nascom.info
Web: http://projectedsound.tk

"mick"


So, is that why the 100k is there and why 0.5 has been chosen for the
coupling cap? Low input impedence into the "virtual ground"?



** The "virtual ground" point is at the conjunction of the 100k, 180k and
1M trim.

The 0.5uF simply cap drives a 1 Mohm load with a -1dB point at 1 Hz.





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

Tom Schlangen wrote:

Gentlemen,

here is a link to a corrected schematic and article using that
"phase splitter with no name" (in German, though):

http://www.jogis-roehrenbude.de/Roeh...unktechnik.pdf

The 1M pot, according to the text, indeed is for adjusting the
drive voltages to the output stage to get maximum 2H cancelation.

The schematic above includes grid coupling cap.

The pot isn't there to get lower 2H from the phase inverter stages themselves.
That is impossible. Whatever 2H is produced by the first half of
the phase inverter is fed into the second stage of the inverter.
This second and entirely separate stage has a shunt FB loop of which the 1M pot is a
part,
and its purpose is to equalise or otherwise adjust the amplitude of the two phases.
But by varying the drive to each PP output tube 2H of the output stage can be nulled to
its low as possible level
and inclusding nulling the 2H from the inverter/input stages.
But you'd need thd meter and CRO to achieve all that, and I bet
trying for maximum 2H cancellation at 1 watt wouldn't mean you'd get good cancelation
at 3 watts, or 20 watts, because the rate of increase in 2H is quite different for
the stages involved.
The idea of driving one output tube harder than the other to cancel 2H isn't
my cup of tea.
A better solution is to use an LTP, with low 2H in the first place, and perhaps better
output
tubes, with lots of class A and 30 watt class A capacity, and UL or CFB or triodes,
and with that sort of kit there is never any thd to worry over at normal levels.

Patrick Turner.





Tom

--
The opposite of a correct statement is a false statement. But the
opposite of a profound truth may well be another profound truth.
- Niels Bohr

"Patrick Turner"

But by varying the drive to each PP output tube 2H of the output stage can
be nulled to
its low as possible level and inclusding nulling the 2H from the
inverter/input stages.
But you'd need thd meter and CRO to achieve all that,



** Have you never done just with a 600 Hz frequency sine wave, a dummy load
and tuned it by ear ??

I mean by listening to the faint OT sing and tuning the drive for least
2H.





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

On Thu, 10 Feb 2005 20:42:16 +1100, Phil Allison wrote:


"mick"


So, is that why the 100k is there and why 0.5 has been chosen for the
coupling cap? Low input impedence into the "virtual ground"?



** The "virtual ground" point is at the conjunction of the 100k, 180k
and 1M trim.

The 0.5uF simply cap drives a 1 Mohm load with a -1dB point at 1 Hz.


Thanks. :-)

--
Mick
(no M$ software on here... :-) )
Web: http://www.nascom.info
Web: http://projectedsound.tk

Phil Allison wrote:

"Patrick Turner"

But by varying the drive to each PP output tube 2H of the output stage can
be nulled to
its low as possible level and inclusding nulling the 2H from the
inverter/input stages.
But you'd need thd meter and CRO to achieve all that,

** Have you never done just with a 600 Hz frequency sine wave, a dummy load
and tuned it by ear ??

I mean by listening to the faint OT sing and tuning the drive for least
2H.

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


Not in hi-fi amps with FB.

I don't claim to be able to hear 2H that is usually below the theshold of my
hearing ability
to detect such things, even with a speaker connected with a lot of series R and
across a dummy load.

But on the odd occassion when farnarkling with some amp and with biasing,
I have had a speaker connected and plainly heard the distortion kick in when
there
has been some frightful fault.

Most amps have NFB and any 2H (or 3H) present that is audible at any level means
that the open loop
2H (or 3H) must be horrendous, and something must be wrong with the amp.
I usually rely on the test gear to tell me all about what's wrong.
Usually, the 3H is the most dominant harmonic in PP amps at all levels but there
is always some
2H in a tube PP amp, unless the circuit is balanced throughout, and has no
SE triode stages, and the balanced tubes, including the outputs, are matched
fairly closely.

In a pair of Quad II amps I worked on last year, I was able to get the thd down
by up to 15 dB just by
swapping positions of the KT66 and EF86.
Finally at about 2 watts, the thd was about 0.05% or a lot less, and one doesn't
hear such levels
of 2H and 3H.
And this was after biasing the tubes for class AB, with separate RC cathode bias
networks.
Full class A would have made only a marginal difference.

Class AB amps can produce considerable 2H if the output tubes are old, or
unmatched,
and especially at high levels, and where the power in the +ve and -ve peaks
differ considerably.

Class A PP triode amps can be routinely made to produce 0.1% thd at 20 watts
with a pair of KT88, and some NFB.
At a watt, they usually measure 0.025%, since thd reduces approximately
with output voltage. That's not something that hits you in the ear very hard.
If it was 7th harmonic, then the amp might be in trouble, and the music.......

Patrick Turner.