Freq. Independent Phase Shifter

Dang! has my mind gone missing again? It's always wandering off. (shuffles through papers on desk) Ahh here it is.

I thought we were responding to Bill (the OP) and not to Jan. Certainly with DDS and a couple of look up tables you can make sine waves with any phase shift you want. But the phase sequence filter would do what the OP wanted. Perhaps too many parts for your taste?

When I first played with this I air wired the R's and C's together... It had a certain beauty to it...

Opps there goes the mind again,

George H.

No, it doesn't! Just buffer that output and it's good to go. It has output impedance variation, not amplitude.

That'll get phase shifts from 0 to 90 degrees ("0" shift -> sine, "90" shift -> cosine).

sin( w*t + phi) =3D sin(w*t) cos(phi) + cos(w*t) sin(phi)

so your weighted sum of sine and cosine just needs the phase sine and cosine for its coefficients; this in general requires negative coefficients so the potentiometer is joined with an inverting amplifier or something in the way of a coupling transformer.

But it does!

Pb is that with a pot you're mixing proportionally to the pot angle, not the sin and cos of the pot angle.

IOW output = phi sin(w*t) + (1-phi) cos(w*t)

(instead of sin(w*t) cos(phi) + cos(w*t) sin(phi) )

which has a min sqrt(1/2) amplitude for phi=0.5

Oops, got that wrong; there IS amplitude variation, up to about 30 percent; there's another familiar phase-shifter with potentiometer/capacitor that does get the amplitude constant...

Gack! That's MUCH easier. if A =3D sin(wt) and B =3D sin(wt + 120 degrees) then A-B =3D sin(wt + 240 degrees)

So, get a 1:1 audio coupling transformer, wire the primary to the A and B terminals, and ground one end of the secondary.

Drat, my first reply was bogus.

If A =3D sin(wt ), and B=3D sin(wt + 120 degrees), and you want to make C =3D sin(wt + 240 degrees), just note that these add to zero

A+B+C =3D 0 So, C =3D -(A+B)

If you have audio coupling transformers handy, that's a good way to do the summation (no power needed, low noise, high reliability).

...

Maybe "Signal Processing Algorithms and Architectures", Hassan Masud Ahmed, PhD thesis, Stanford University, June 1982. (1981?)

It is frequently cited* but might not be freely available online.

You are probably right. I always thought of Ahmed as a first name, and found it odd to be this guys last name. Not that I know much about the middle east, but that oddity made the name stick.

You need to run the look up table for the coordic a few more bits than word size for the final result if you want the answer good to the last bit. Still, the efficiency of the algorithm is so much better than a simple sine look up table that going a few extra bits is worth it.

ter

ror

from

"The Journey is the reward"

eff.com

If you are making ONE box AND you have the right topology, you can just measure your standard tolerance capacitor to a high degree of accuracy, then buy the proper high accuracy resistors. These filter designs usually use more op amp, but have greater flexibility. Leapfrog ladder filters generally can be built in this manner. To be a clearer here, the filter can't depend on ratios of capacitors.

Obviously for a production unit, it would be cumbersome to pick all the resistors based on measured capacitors. But to build a few circuits on the cheap (if you don't values your labor!), the scheme works.

See my post for a circuit that makes the amplitude nearly constant.

se

d,

quoted text -

I should/could have thought of the loaded pot, that's an easy way to get rid of some of the variation. (I learned that trick from Phil H's book.) But I've never seen postive feedback used like that before. I assume that this gives some gain that changes as the source impedance changes.

Thanks,

George H.

Late at night, by candle light, snipped-for-privacy@protech.com (Bill Murphy) penned this immortal opus:

Try Visual Analyzer from

Late at night, by candle light, Jan Panteltje penned this immortal opus:

VCO,

obvious.

Doesn't take all that much time. The soundcard is usually there already, d/l'ing the sw is another few minutes (see my above post), the time to get used to it is about the same as fiddling with a real generator. Plus the sw comes loaded with a number of other goodies.

- YD.

On a sunny day (Wed, 12 May 2010 01:45:27 -0300) it happened YD wrote in :

Those are good points, but then it will kill the music from your soundcrd. Even in my case with 2 soundcards.. It also requires a free PC, as sound from a soundcards is not always free from hickups if other processes run with high priority. WTF is so difficult about a few chips in a small box? Not even mentioning the low frequency side, my SB live soundcard is already way low at 10 Hz... Did you ever look with the scope at 14 kHz on a 48 kHz samples audio out? Frequency charecterisic is not flat either, even with 'tone control' disabled. A R2R ladder on a EPROM output is 10 x better. But indeed if you need 16 bits dynamic range..... then maybe the soundcard, maybe. output ^ | bit of RC lowpass ^ | R2R DACs x as many as you like ^ ^ | 8 | EPROMS with sine tables + ^ | | 8 ... 16 whatever bits [ ]

On a sunny day (Tue, 11 May 2010 18:27:34 -0700 (PDT)) it happened George Herold wrote in :

Just use AGC.

hase

und,

de quoted text -

Yes the positive feedback causes the gain to rise as the impedance does.

It matches up nicely with the loss so that the amplitude ends up nearly flat.

The two resistors for the phase fixing don't quite match up perfectly but it makes the error hit zero at 5 points along the travel of the pot so it is darn good.

pha=3D

oun=3D

ide=3D

=A0I

AGC is way more complex than just adding a single resistor to the design so why go that way?

On a sunny day (Wed, 12 May 2010 06:28:53 -0700 (PDT)) it happened MooseFET wrote in :

Yes it is more complex, but it sort of get you a 100% correct amplitude. I do not see a flat curve with resistors, that math eludes me.