Phase shift networks

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
[quote author="StephenGiles"]Look at this:
http://img.photobucket.com/albums/v317/StephenGiles/phaseshift.gif
What possibilities do we have for strong phasing when modulated?
Stephen[/quote]

The handwriting looks familiar. What is this from?

Hard to modulate APFs where the building blocks are not 1st order, btw.

JH.
 
Ah, Orban.
The files are bigger than I can download right now. What function does this box perform?

JH.
 
It's a stereo synthesiser, with all sorts of stuff on 3 pages of circuits. It seems to detect whether a stereo signal is present at the input, and if not, creates one using the phase shifters in whichever channel is necessary.
Stephen
 
Ok, I understand. So this phase shift network doesn't even need high precision, thus the 2pole building blocks. And no need to sweep anything either.

JH.
 
These are the 3 main circuits for your bed time reading until you can download the files!

http://img.photobucket.com/albums/v317/StephenGiles/orban275b.gif
http://img.photobucket.com/albums/v317/StephenGiles/orban275a.gif
http://img.photobucket.com/albums/v317/StephenGiles/orban275c.gif
Stephen
 
[quote author="StephenGiles"]Look at this:
http://img.photobucket.com/albums/v317/StephenGiles/phaseshift.gif
What possibilities do we have for strong phasing when modulated?
Stephen[/quote]
This is second order allpass. Orban used it in his Optimod and
I have analysed it. It is very, very sensitive to gain (r155 / r 156)
if resistors in gain are not matched, it is not allpass.
Orban types ! near resistors, in some schematics.
To change poles or tunning it is absolutely impossible.
One real chance to tune this is if you change condensers to varactors :)
 
[quote author="StephenGiles"]Thanks xvlk, I thought we might be in line for a new phaser sound here!
Stephen[/quote]
The second order allpass from Orban s optimod is here:
http://sweb.cz/vavak3/semestralka/orfc.tif
Orban uses it to improve crest-factor of the speech.

Posibility to modulate second order allpass is this.
If you know reactancy transform, you can all devices in some circuit
with voltage transfer /hard feed, soft load/ multiply with p without change
of transfer. If you do it inthis allpass with group of R and C in the bottom,
you change capacitors to resistors and resistors to inductors.
Note, that resistors are two with the same value.
If you do substitution (p=1/p), the inductors became capacitors
and frequency became inversion, but allpass is still allpass.
This second order allpass can be modulated via paired optocoupler.
Only capacitors must be accurate and you must combined, but
for one piece you can do it, for serial production you must have it wounded.
This is circuit:



_____ ____
---------|_____|--X-|____|--|
| | |
| ---|- |
| | | | |____x_
--------X-----| |-----------X--| | |
| Rf | | Rf | |- |
| _____ ____ | |
|--|_____|-x-|____|-| __|__
| _____
| |
x-----------------------
|
__|__
_____
|
__|__


It is hard to make semigraphics with proportional font, but think, that
I explained it for You. In Orban schematics it is R24,R25,R26 which
changes to capacitors and C6 and C7 which changes to optoresistors
/matched optocouplers/ To compute values of capacitors you must
work on normalised prototype (this is filter with unity frequencies
and unity resistors/. You can read Huelsmann, if you don t know what
normalisation is.
 
[quote author="StephenGiles"]Thanks xvlk, I thought we might be in line for a new phaser sound here!
Stephen[/quote]
Via reactancy transform of second order allpass you can really do
NEW sound, if you will have parametric second order allpas with
Q higer that one /i.e. Q=2/, sound of it can not be done by any cascade
of first order alpases.
Example of characteristic of second order allpases with different Qs
is here:
http://mujweb.cz/www/xvlkxvlk/phdia.gif
your second order allpass is hard to design, I don t know nice
design equation for this type circuit, I had only iteration.
If someone have it, he can give it.
Via iteration I have computed two values of normalized prototype.
Values for Q=1 are very pretty, values for Q=2 are not.
here are:
http://mujweb.cz/www/xvlkxvlk/normal.gif
You can do reactancy transform with this circuit and then generate diferent
still alpass. Examples of this transforms are here:
http://mujweb.cz/www/xvlkxvlk/needit.gif
Only one transform is valued /because it is without coils/
example of use it for 10 kohm optos and 8kHz allpass is here:
http://mujweb.cz/www/xvlkxvlk/odnorm.gif
final circuit is here:
http://mujweb.cz/www/xvlkxvlk/final.gif
If you have lowest frequency 4 kHz at 10k optos, double capacitor values.


It was my semestral work from electronic filter design in year 2001. I m glat that it can someone use it.

xvlk
P.S.: you must cut and paste links to the browser.
if you click on it, only blabla you will see.
 
[quote author="StephenGiles"]Hi xvlk - I can't get the gifs, I just get a page called ATLAS.
Stephen[/quote]
Hmmm... I can just click on the link and the gif comes up in a new browser.
 
[quote author="StephenGiles"]Yes, now I can - I didn't read the post properly. Thanks xvlk. Are you seeing a lot of tourists from western Europe in your city now? We hope to visit sometime.
Stephen[/quote]
I live on the subrub and no tourists are here. Only factories
after bancrots are here and buildings of scientific institutes going to ruins.
I am not studiing tourist-university and I not understand it,
but if someone prosper as hotel owner, why I can not prosper
as an laboratory owner?
/after I finish university/

xvlk
 

Latest posts

Back
Top