[sdiy] Weaver frequency shifter

Eric Brombaugh ebrombaugh at gmail.com
Wed Apr 27 17:43:56 CEST 2022


I've done a fair bit of work in this area, both in synths and in 
communications for the day job. It's interesting to note that most 
quadrature tuning approaches are mathematically equivalent - the main 
differences being that you're just re-arranging the order of operations. 
Weavers method isn't doing anything fundamentally different from the 
Hilbert or phase-shift but it does put a tuner at the beginning of the 
process which has to be removed at the end.

The advantage of doing this by the Hilbert approach is that a lot of the 
operations become constants you can roll into the Hilbert filter 
coefficients. The result is collapsing the number of DSP operations 
required and thus reducing the CPU load over what you'd need to do with 
Weavers.

Eric

On 4/27/22 03:00, rburnett at richieburnett.co.uk wrote:
> Has anyone tried building a Frequency shifter using Weaver's method? 
> Either analogue or digital.
> 
> I'm asking out of interest, rather than any immediate intention to build 
> anything.
> 
> All of the synth related designs I've seen seem to be based around the 
> phasing method, using either an analogue dome filter or digital Hilbert 
> transform to impart the 90 degree phase difference.  I was looking at 
> Weaver's or "the third method" of SSB generation relating to an SDR 
> discussion at work the other day, and it looks like a good candidate for 
> a modern DSP implementation.
> 
> -Richie,
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy
> Selling or trading? Use marketplace at synth-diy.org



More information about the Synth-diy mailing list