[sdiy] Re: SV: Re: Saw phase shifter used in frequence shifting.

[Magnus Danielson]

From: karl dalen <dalenkarl at yahoo.se>
Subject: SV: Re: Saw phase shifter used in frequence shifting.
Date: Sat, 17 Sep 2005 03:48:29 +0200 (CEST)
Message-ID: <20050917014829.52936.qmail at web25504.mail.ukl.yahoo.com>

> 
> --- Magnus Danielson <cfmd at bredband.net> skrev:
> 
> > From: karl dalen <dalenkarl at yahoo.se>
> > Subject: Saw phase shifter used in frequence shifting.
> > Date: Sat, 17 Sep 2005 03:29:25 +0200 (CEST)
> > Message-ID: <20050917012925.6623.qmail at web25502.mail.ukl.yahoo.com>
> > 
> > > Harry, Magnus and the rest who migh be interested!
> > > 
> > > I came onto a idea here, can it be so that one could 
> > > use the phased saw as a angle wrecker for an frequency shifter?
> > > SSB, DSB type of shifter? (asuming the saw goes to the basic
> > > saw to tri and then diode or saturated OTA sine shaper).
> > > 
> > > If a PW width of 50 declare 0 degres (saw phase will then start
> > > at 90deg and end at 270 so the variable range would be 180 deg)
> > > and anything above 50 gives positive direction of frequency 
> > > shifts and anything below 50 gives negative frequency shift?
> > > 
> > > You still have the straight saw available as phase reference
> > > 
> > > Many types of quadratues could easily
> > > be made out from one saw phase VCO!
> > > 
> > > Bad idea?
> > 
> > No, I don't see any problem with this. It should work well to get the
> > quadrant
> > variant from the same oscillator core. What is an issue, is the overtones of
> > the shaped sine which I think is more of a concern for a frequency shifter
> > than
> > for normal musical useage. However, beyond that I can't see any problem.
> 
> I figured that to!

Good. I have not pondered about how well the overtones must be suppressed, so
that would be nice to learn. The overtones will cause additional "shadow"
variants, which *may* be of interest, but I think a cleaner version should be
the ideal I guess.

> > So, for the carrier oscillator it is clearly a go, the phase-shifting network
> > for the signal remains an issue, but the polyphase filter is fairly simple
> > and
> > should work well.
> 
> I was talking about the SSB DSB shifters, dont they use just a simple
> LPF filter between each multiplyer! Maybe the allpass type sound better 
> then the SSB? I have no idea since i have never built a Frequency shifter unit!

Sorry, I don't have all the frequency shifter layouts in clear memory, so you
can't expect me to have all the details, pros and cons directly available, now
can you?

What is a common problem however, is the 90-degree angle of the oscillators,
since that will affect the mirror-cancelation which makes it become more than a
expensive form of a ringmodulator.

I have Mike Gingells PhD report online at
http://rubidium.dyndns.org/~magnus/synths/friends/gingell/
and in there you also find a review of different SSB techniques and their pros
and cons. Highly recommened.

Cheers,
Magnus