Dome Filter Question

[Martin Czech]

> Ok, I have seen the question on diy when I was at the office and had my
> FS shifter documents not in front of me. So now I will answer with a cc
> to synth-diy, as I have the EN article in front of me now.
> 
> I used two branches of 6 opamp based all pass filters in series. The
> RC products for each stage are:
> 
SNIP

I'm just curious: did you simulate the phase difference between the two
branches? I mean how good is the quadrature approximation, or how far
can you get with six stages (theoretically)?  You know, there's also
this passive circuit (polyphase network), it needs only matching caps
, no absolute accuracy.  OTOH is has considerable, frequency depended
damping, and for a wide band it needs a lot of stages (~15).

And if one builds such a circuit, how can he check that everything
is working? It would be easy to measure the time delay with 
an two beam scope, if the signals where not 90 DEG out of phase.

Maybe one could use a good multiplier:

sin(wt)*sin(wt+p)=0.5*cos(-p)+0.5*cos(2wt+p)

The first cos is a dc component (0 for p=pi/2), the second has double
frequency and is shiftet by p (which should be PI/2).  The zero crossings
of sin(wt) and cos(2wt+p) should be at the same spot if the phase network
works correctly (ac coupled signals at oscilloscope). A phase error will
also result in a dc component of the multiplier output. If the multiplier
is good enough, it could be measured with a simple integrating voltmeter
(maybe add passive lowpass).



m.c.