stereo expander, part 2

[J, David]

I come back from my previous post about this kind of stereo expander.
I am analysing the so called "deluxe version" available at this link:

http://sound.westhost.com/project21.htm

Let's say I'm only interested by applying negative feedback, no positive
feedback. This means I replace both 10K potentiometers with 5K
potentiometers, both in series with a 5K resistor so I can tap only the
negative feedback.
I assume the potentiometer are linear type in these examples. But are they?
When 100% negative feedback is applied, no monophonic in-phase signal got
through. As less negative feedback is applied by slowly turning up the pots,
the monophonic signal will gradually get stronger.

So, should the negative feedback not be controlled by logarithmic
potentiometers instead?

J. David

 

[DaveM]

I come back from my previous post about this kind of stereo expander.
I am analysing the so called "deluxe version" available at this link:

http://sound.westhost.com/project21.htm

Let's say I'm only interested by applying negative feedback, no positive
feedback. This means I replace both 10K potentiometers with 5K
potentiometers, both in series with a 5K resistor so I can tap only the
negative feedback.
I assume the potentiometer are linear type in these examples. But are
they?
When 100% negative feedback is applied, no monophonic in-phase signal got
through. As less negative feedback is applied by slowly turning up the
pots, the monophonic signal will gradually get stronger.

So, should the negative feedback not be controlled by logarithmic
potentiometers instead?

J. David

The pots in this circuit do not affect feedback, neither negative nor
positive. They are introducing a portion of the opposite channel's signal
into the final stage's inputs, which are acting as linear mixers.
In this situation, a linear taper would be used, because the center position
of the pots should result in zero signal going into the opposite channel. A
log taper pot would not give zero output in the center position.

--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in
the address)

Never take a laxative and a sleeping pill at the same time!!

 

[Phil Allison]

"J, David"

I come back from my previous post about this kind of stereo expander.
I am analysing the so called "deluxe version" available at this link:

http://sound.westhost.com/project21.htm

** Just about all this device will *actually* do is REMOVE the deep bass
from your music.

Typical stereo recordings have the deep bass in mono ( ie similar level and
phase in both channels) - pretty much nothing else in the mix is like that.

Better it did not work full range and left the bass frequencies alone.


BTW

The designer is a friend of mine.



......... Phil

 

[J, David]

The pots in this circuit do not affect feedback, neither negative nor
positive. They are introducing a portion of the opposite channel's signal
into the final stage's inputs, which are acting as linear mixers.
In this situation, a linear taper would be used, because the center
position of the pots should result in zero signal going into the opposite
channel. A log taper pot would not give zero output in the center
position.

That's why I replaced the 10K pots to 5K and adds 5 resistors in series.
This way, only the negative feedback (of the adjacent channel, you're right
to spot this) will be used.

 

[J, David]

"J, David"


** Just about all this device will *actually* do is REMOVE the deep
bass from your music.

Typical stereo recordings have the deep bass in mono ( ie similar level
and phase in both channels) - pretty much nothing else in the mix is like
that.

Better it did not work full range and left the bass frequencies alone.

Funny but, a friend of mine just tried a quick and different approach for
that circuit. He picked two small 600 ohm phone line transformers to get the
negative out-of-phase feedback from the adjacent channel. Only passive
parts. Because these transformers have a (very) bad frequency response at
low frequencies as well as high frequencires, bass frequencies are almost
unaffected. Same for treebles.
The end result seems to cancel mostly vocals.

 

[Phil Allison]

"J, Davi"

Funny but, a friend of mine just tried a quick and different approach for
that circuit. He picked two small 600 ohm phone line transformers to get
the negative out-of-phase feedback from the adjacent channel. Only passive
parts. Because these transformers have a (very) bad frequency response at
low frequencies as well as high frequencires, bass frequencies are almost
unaffected. Same for treebles.

** Yawn.

Wota fuckwit.



.......... Phil

 

[J, David]

"J, Davi"




** Yawn.

Wota fuckwit.

Indeed!
Jim


......... Phil

 

[DaveM]

That's why I replaced the 10K pots to 5K and adds 5 resistors in series.
This way, only the negative feedback (of the adjacent channel, you're
right to spot this) will be used.

You keep talking about "negative feedback" in this circuit. This is NOT
negative feedback, or feedback of any kind. It is simply the injection of
in-phase or out-of-phase signal into a mixer stage. The pots simply allow
the user to adjust the amount of signal fed into the mixer. That's it.
Your idea to split the pots into 5K fixed and 5K pots will indeed do what
you want, but your terminology is misleading.
--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in
the address)

Never take a laxative and a sleeping pill at the same time!!

 

[J, David]

Let's say I'm only interested by applying negative feedback, no

You keep talking about "negative feedback" in this circuit. This is NOT
negative feedback, or feedback of any kind. It is simply the injection of
in-phase or out-of-phase signal into a mixer stage. The pots simply allow
the user to adjust the amount of signal fed into the mixer. That's it.
Your idea to split the pots into 5K fixed and 5K pots will indeed do what
you want, but your terminology is misleading.

My choice of words is bad no doubt. But my original question is still
unanswered. Logarithmic or linear pots?
I guess I'll have to experiment...

Regards!
Jim

 

[Ted Edwards]

I come back from my previous post about this kind of stereo expander.

Back in the mid '60s I designed a stereo amp that had this feature.
I've been trying to remember the details without success and the circuit
is burried in a box in the basement. :-( However the principle was
simple and the implementation only required two op-amps, one linear pot
and a few resistors.

The principle is: Use the two op-amps to form a differential amplifier
with variable common mode gain. If the differential and common mode
gain are equal, there is no change in separation. If they are not then
either blending (ultimately producing mono) or increased separation occurs.

I expect one of you geniuses can refresh my memory re what I designed 40
years ago.

Ted

 

[DaveM]

|
My choice of words is bad no doubt. But my original question is still
|
unanswered. Logarithmic or linear pots?
|
I guess I'll have to experiment...
|

|
Regards!
|
Jim
|

|

|

^
Yes I did answer your original question.. If you'll go back to one of my
previous posts, I advised you to use linear pots

--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in
the address)

Never take a laxative and a sleeping pill at the same time!!

 

[J, David]

My choice of words is bad no doubt. But my original question is still

Yes I did answer your original question.. If you'll go back to one of my
previous posts, I advised you to use linear pots

If the circuit is considered as a mixer, using linear pots do make sense of
course.

Jim