[Andy Evans]

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

[John Byrns]

In article om,
"Andy Evans" wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

IIRC there was at least one published in "Sound Practices" magazine way
back when, unfortunately I can't quote the issue number at this point.


Regards,

John Byrns

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

[Eeyore]

Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

[Patrick Turner]

Eeyore wrote:

Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

Yes Graham, but he wants a complete schematic all worked out.
Allen Wright has a few.

A long tail pair with a pair of 2SK369 each side with 2k2 RL and CCS
tail
will give differential gain of about 88 with 5mA in each j-fet.
Source resistors can reduce this gain and the THD/IMD if required.

But then the folowing RIAA filter has to be considered, and whether one
wants to use NFB for the RIAA eq,
or passive filtering, and whether the gain is 60dB over all at 1 kHz for
MC, or 40dB for MM.
With phono there are numerous things to consider.

I found balanced to be difficult to get to be as quiet as ppl said I
should be able to get it,
and I settled for single ended cascode input, passive RIAA, then a
µ-follower
and I ended up with the Rocket at
http://www.turneraudio.com.au/preamp...hono-2005.html


Patrick Turner.

[Patrick Turner]

Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Try Allen Wright's website at
http://www.vacuumstate.com

Patrick Turner.

[Eeyore]

Patrick Turner wrote:

Eeyore wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

Yes Graham, but he wants a complete schematic all worked out.
Allen Wright has a few.

A long tail pair with a pair of 2SK369 each side with 2k2 RL and CCS
tail will give differential gain of about 88 with 5mA in each j-fet.
Source resistors can reduce this gain and the THD/IMD if required.

Yes, I was thinking this would be a good case for CCSs. In fact, if the objection to
semiconductors isn't to great,a current mirror at the top end would be brilliant.


But then the folowing RIAA filter has to be considered, and whether one
wants to use NFB for the RIAA eq,
or passive filtering, and whether the gain is 60dB over all at 1 kHz for
MC, or 40dB for MM.
With phono there are numerous things to consider.

I like the idea of doing RIAA part passive and part active actually.


I found balanced to be difficult to get to be as quiet as ppl said I
should be able to get it,

Balanced will always be ~3 dB noisier in theory, all other things being equal. I'm
unclear of the reason for it here.

Graham

[Patrick Turner]

Eeyore wrote:

Patrick Turner wrote:

Eeyore wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

Yes Graham, but he wants a complete schematic all worked out.
Allen Wright has a few.

A long tail pair with a pair of 2SK369 each side with 2k2 RL and CCS
tail will give differential gain of about 88 with 5mA in each j-fet.
Source resistors can reduce this gain and the THD/IMD if required.

Yes, I was thinking this would be a good case for CCSs. In fact, if the objection to
semiconductors isn't to great,a current mirror at the top end would be brilliant.

The idea of a current mirror for the drain loading of each side of the
j-fet
halves increases complexity, and hugely increases gain and output
resistances
which then dramatically increase Miller C if NFB isn't used.

Having 2k2 or about 2k7 for each drain load for 2SK369 gives
a gain of 88 to 108 approx, and an identical following stage
gives the same, so with a total gain of approx 10,000, you will have a
gain of
1,000, or 60dB to allow and MC input 0.3mV at 1 khz to become 300mV at
the output,
allowing for the passive attenuation of the 1kHz in the RIAA filter
between stages to
be -20dB.

The first stage gain can be reduced with source resistors, ie, local
current FB
and the Miller C then reduces OK for higher Z MM carts.
The Miller C won't affect low Z MC carts.

The 2SK369 is a very low noise device and even in SE mode the THD of the
first stage
with full gain and MM is negligible since THD varies with output
voltage.
With an SE device, THD is quite high compared to any tube, about 1%+ per
output volt,
but if the Vo = 10mV, then THD = 0.01% approx, and mainly all 2H,
depending on the load.
With 6.4k ohm load on 2SK369, there is almost no 2H from the fet;
there is a null in 2H like there is with a pentode with a particular
value of RL.
So mainly only 3H etc remain, but it is suprisingly low.
The use of the LTP cancel whatever 2H and other even order products so
having loads of 2k7
will produce much lower overall THD ( and IMD ) than the SE device.

With an LTP expect 0.01% THD at 0.1V out.

The second stage will produce the most THD because the signal
voltage is about 10 times that of the first stage, so the use of an LTP
made with twin triodes
and pair of direct coupled cathode follower buffers makes goos sense to
reduce THD to utte5rly
negligible levels.
The buffers form a really good low Z source to drive a small sized
output transfromer
to allow the output secondary to float completely free, so that it will
suit
the balanced or single ended input of any following amp and should be
hum/noise free since
no direct earth connection made in a OV line.

But then the folowing RIAA filter has to be considered, and whether one
wants to use NFB for the RIAA eq,
or passive filtering, and whether the gain is 60dB over all at 1 kHz for
MC, or 40dB for MM.
With phono there are numerous things to consider.

I like the idea of doing RIAA part passive and part active actually.

I found balanced to be difficult to get to be as quiet as ppl said I
should be able to get it,

Balanced will always be ~3 dB noisier in theory, all other things being equal. I'm
unclear of the reason for it here.

One can use 3 x j-fets per side of the LTP in parallel, which should
make the
input amp stage quieter.
The loading can also be lower, 1k per side is OK and thus the output
noise is low
which is important with low level MC carts.

I recently used 3 j-fets in parallel for an SE input stage with Id =
15mA, and RL = 2k2,
with Rs = 72 ohms, and this was a splendid performer, open loop gain
was 240, closed loop with the local current FB = 27.
The 2k2 is a load close to where the null in 2H occurs ( with 3 fets )
and with the CFB the distortion is utterly negligible.
I have a passive 318uS and 3180 uS filter following this stage, then a
µ-follower
stage with fet + bjt follower, then 75uS passive filter, then µ-follower
using all darlington connected bjts.
Supply is +/-41V and maximum output voltage is 16Vrms and undistorted to
up to 30kHz,
and is slightly better than the tube circuits I have built.
No global FB or local loop FB is used for RIAA, only local current FB in
source or emitter circuits
to firm the gain of each of the 3 stages so that device variations don't
lead to large gain differences
between channels or after replacing a fet or bjt if need be.

When using an accurate reverse RIAA filter at the input and a 1kHz
square wave signal
the output signal remains less distorted, ie, better looking square wave
to a higher
voltage than the fully tubed Rocket design of mine where the RIAA eq is
done in just one filter
between only 2 gain stages.

I will be publishing the schematic at my site when i find time.

Patrick Turner

Graham

[Eeyore]

Patrick Turner wrote:

Eeyore wrote:
Patrick Turner wrote:
Eeyore wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

Yes Graham, but he wants a complete schematic all worked out.
Allen Wright has a few.

A long tail pair with a pair of 2SK369 each side with 2k2 RL and CCS
tail will give differential gain of about 88 with 5mA in each j-fet.
Source resistors can reduce this gain and the THD/IMD if required.

Yes, I was thinking this would be a good case for CCSs. In fact, if the objection to
semiconductors isn't to great,a current mirror at the top end would be brilliant.

The idea of a current mirror for the drain loading of each side of the
j-fet halves increases complexity, and hugely increases gain and output
resistances which then dramatically increase Miller C if NFB isn't used.

Eh ? By providing a near constant anode voltage, the current mirror load all but
eliminates the effect of Miller C !

Graham

[Patrick Turner]

Eeyore wrote:

Patrick Turner wrote:

Eeyore wrote:
Patrick Turner wrote:
Eeyore wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Your best bet IMHO is a long tailed pair input stage with fixed gain followed by
an EQ stage.

Graham

Yes Graham, but he wants a complete schematic all worked out.
Allen Wright has a few.

A long tail pair with a pair of 2SK369 each side with 2k2 RL and CCS
tail will give differential gain of about 88 with 5mA in each j-fet.
Source resistors can reduce this gain and the THD/IMD if required.

Yes, I was thinking this would be a good case for CCSs. In fact, if the objection to
semiconductors isn't to great,a current mirror at the top end would be brilliant.

The idea of a current mirror for the drain loading of each side of the
j-fet halves increases complexity, and hugely increases gain and output
resistances which then dramatically increase Miller C if NFB isn't used.

Eh ? By providing a near constant anode voltage, the current mirror load all but
eliminates the effect of Miller C !

Graham

Are we talking about the same thing?
By current mirror, I meant having a couple of CCS to each LTP j-fet
drain circuit,
so that current change in CCS, even though it be tiny, is mirrored in
the other CCS.

To keep the circuit fully balanced, and have high gain, and have R loads
on each drain, and have low Miller C, then the use of a pair of
cascoded j-fets to each LTP half is called for.
Where you have a cascode pair of j-fets, the top one has the voltage
gain of gm x RL,
but bottom j-fet drain sees the source input Z of the top, ie, 1/gm,
which is low,
and thus the bottom fet has unity gain, because drain gain = gm x RL, so
gm x 1/gm = 1, simple.
But we have used 4 j-fets and we get no better linearity than just two.

Patrick Turner.

[Andy Evans]

On Feb 10, 3:58?pm, Patrick Turner wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Try Allen Wright's website athttp://www.vacuumstate.com

Patrick Turner.

Allen's phono stage is actually single ended if you look at it. One of
the good ones I've dug up is your fellow aussie Mark Kelly:
http://www.members.iinet.net.au/~quiddity/audio.html

Others I've found a
http://www.tubecad.com/july99/page9.html
http://www.rintelen.ch/pdf/DIY/Anastasia.pdf

and ss -
http://www.klaus-boening.de/html/schematics.html#MKIII
http://www.klaus-boening.de/html/phonostage.html
http://www.quadesl.com/pdf/icphono.pdf

What do you guys prefer out of that lot? Andy

[Arny Krueger]

"Andy Evans" wrote in
message
ups.com

Andy Evans wrote:

Anybody know where I can find some schematics for a
balanced phono stage? Fancy trying a build. Andy

http://www.klaus-boening.de/html/schematics.html#MKIII

I seriously doubt that its RIAA equalized.

http://www.klaus-boening.de/html/phonostage.html

Lacks a differential input.

http://www.quadesl.com/pdf/icphono.pdf

The passive equalization probabaly sacrifices dynamic range.

[Patrick Turner]

Andy Evans wrote:

On Feb 10, 3:58?pm, Patrick Turner wrote:
Andy Evans wrote:

Anybody know where I can find some schematics for a balanced phono
stage? Fancy trying a build. Andy

Try Allen Wright's website athttp://www.vacuumstate.com

Patrick Turner.

Allen's phono stage is actually single ended if you look at it. One of
the good ones I've dug up is your fellow aussie Mark Kelly:
http://www.members.iinet.net.au/~quiddity/audio.html

Others I've found a
http://www.tubecad.com/july99/page9.html
http://www.rintelen.ch/pdf/DIY/Anastasia.pdf

and ss -
http://www.klaus-boening.de/html/schematics.html#MKIII
http://www.klaus-boening.de/html/phonostage.html
http://www.quadesl.com/pdf/icphono.pdf

What do you guys prefer out of that lot? Andy

There is much food for thought provided here.

I thought Allen Wright had a couple of balanced input staged phono amps.

Fully balanced means bal in, bal out, and depends how fussy one is...

One could have an OPT slung between cathodes of a pair of CF output
triodes,
then the secondary just floats completely untied to any 0V by any
resistance,
and this is the real McCoy of balanced output adaptable to SE without
any switch or soldered link
or wasting 1/2 the amp output.


And for dessert, with icecream, after the food above for dinner,
one could turn to Erno Borbely,
http://www.borbelyaudio.com/pics/Borbely-jfet1-web.pdf

He's into j-fets in a big way!

Patrick Turner.

[Paul D. Spiegel]

Allen Wright's _The Tube Preamp Cookbook_ describes both single-ended and
balanced phono pre-amps. It's a fun read and has lot's of interesting
ideas.

The Sound Practice project mentioned in a previous post is likley the J.C.
Morrison design in 1993 Vol. 1 #3 he calls the 'Siren Song'. The input
stage is balanced but finishes with a single-ended output.

[Patrick Turner]

"Paul D. Spiegel" wrote:

Allen Wright's _The Tube Preamp Cookbook_ describes both single-ended and
balanced phono pre-amps. It's a fun read and has lot's of interesting
ideas.

The Sound Practice project mentioned in a previous post is likley the J.C.
Morrison design in 1993 Vol. 1 #3 he calls the 'Siren Song'. The input
stage is balanced but finishes with a single-ended output.

And why would this be so?
Probably because we try to minimise hum and noise with a balanced input,
yet accommodate the vast majority of power amps which have se inputs,
and where signal levels are high enough to very easily get a decent SNR.

Balanced input are one thing, but balanced outputs are a complete PITA
and a waste if they are not floating, and building the floating balanced
output without using an OPT is difficult, is it not?
OPTs in preamps as well as IPTs are disfavoured by 99% of commercial
makers because of the cost and difficulty.
Many would say such low level transformers are contributory to poor F
response
and increased distortions, veils upon the face of music,
but then as Allen Wright said in his cook book, many great recordings
have been done
using transformer riddled equipments.

Patrick Turner.

[Andy Evans]

Many would say such low level transformers are contributory to poor F
response
and increased distortions, veils upon the face of music,
but then as Allen Wright said in his cook book, many great recordings
have been done
using transformer riddled equipments.

Patrick Turner.

But we accept all kinds of capacitors in the signal path - even
electrolytics for God's sake. Talk about veils on the face of the
music - get rid of those caps. Transformers are no worse and often
better.

[Eeyore]

Andy Evans wrote:

Many would say such low level transformers are contributory to poor F
response
and increased distortions, veils upon the face of music,
but then as Allen Wright said in his cook book, many great recordings
have been done
using transformer riddled equipments.

Patrick Turner.

But we accept all kinds of capacitors in the signal path - even
electrolytics for God's sake. Talk about veils on the face of the
music - get rid of those caps. Transformers are no worse and often
better.

What makes you think electrolytics are audible when *correctly* used ?

Graham

[Patrick Turner]

Eeyore wrote:

Andy Evans wrote:

Many would say such low level transformers are contributory to poor F
response
and increased distortions, veils upon the face of music,
but then as Allen Wright said in his cook book, many great recordings
have been done
using transformer riddled equipments.

Patrick Turner.

But we accept all kinds of capacitors in the signal path - even
electrolytics for God's sake. Talk about veils on the face of the
music - get rid of those caps. Transformers are no worse and often
better.

What makes you think electrolytics are audible when *correctly* used ?

Graham

Electros are used in just about all audio gear to smooth the PS
and offer a low impedance working supply. Like a battery, they are.
Electro coupling caps are very much hated by the Hifi Cognescenti
and heven help anyone who doesn't use Black Gate PS caps.

Bypass caps are routinely condemned, and even Allen Wright
has stern words about the perils of bypassing, Electro PS caps,
styrene caps, and ceramic caps.
I have tried Auricaps instead of Wima polypropylene and heard no sound
changes.
I have swapped non polarised 3.3uF electros in series with a tweeter
with
3.3uF plastic caps and again heard no change at all.

Electro coupling caps used to be much more common in discrete bjt based
amps of the
1960s and 70s, because base input resistance was so low, and to get good
bass
one needed a high value of C and something smaller than
a beer can, so an electro was the choice.

Now where there may be 5 such couplings in the signal path, and if all
these caps go
dry after 20 years, then there will be a sound change, and it isn't a
good one.

But if we replace ALL the caps in dad's favourite 1970 NAD, then maybe
its sounds well again,
or if it ever did when it was new.

Patrick Turner.