Message	Date	From	Subject
13737	2017-03-25 02:43:43	Hans Summers	Re: Digest Number 2865
Hi Randy following your explanation the phase offset parameter might also be used to compensate phase errors introduced in the mixers/detectors?? I don't think so. As you pointed out, at 14MHz the phase offset resolution is 1.7 degrees. As I recall in phasing method unwanted sideband to get 40dB cancellation requires at least 1dB accuracy. So the kind of coarse adjustment provided by the Si5351A wouldn't be good enough for phase error cancellation. Better to use an active all-pass network with phase adjustment and amplitude adjustment. Alternatively a passive RC polyphase network works very well, too. It inherently averages out phase errors. You can get good unwanted sideband rejection without any adjustment at all. But choice of components can be tricky, if you want flat response, and it's not so easy to provide adjustments. So I prefer polyphase audio phase shift networks where you want it to be adjustment-free; an active all-pass phase shift circuit can perhaps provide better unwanted sideband cancellation if phase/amplitude conpensating adjustments are provided and are very well adjusted. Either way - I think the Si5351A is great for providing an accurate 90-degree quadrature LO but there isn't enough fine resolution in that adjustment to provide phase compensation overall for the audio phase shift. 73 Hans G0UPL http://qrp-labs.com

Hi Randy

following your explanation the phase offset parameter might also be used to compensate phase errors introduced in the mixers/detectors??

I don't think so. As you pointed out, at 14MHz the phase offset resolution is 1.7 degrees.

As I recall in phasing method unwanted sideband to get 40dB cancellation requires at least 1dB accuracy. So the kind of coarse adjustment provided by the Si5351A wouldn't be good enough for phase error cancellation.

Better to use an active all-pass network with phase adjustment and amplitude adjustment.

Alternatively a passive RC polyphase network works very well, too. It inherently averages out phase errors. You can get good unwanted sideband rejection without any adjustment at all. But choice of components can be tricky, if you want flat response, and it's not so easy to provide adjustments.

So I prefer polyphase audio phase shift networks where you want it to be adjustment-free; an active all-pass phase shift circuit can perhaps provide better unwanted sideband cancellation if phase/amplitude conpensating adjustments are provided and are very well adjusted.

Either way - I think the Si5351A is great for providing an accurate 90-degree quadrature LO but there isn't enough fine resolution in that adjustment to provide phase compensation overall for the audio phase shift.

73 Hans G0UPL

http://qrp-labs.com