[hpsdr] New clock oscillator for OpenHPSDR?

[Dan Mills]

>
> Please correct me if I am wrong, but the loop filter at the output of the
> phase comparator prior to the control input at the 122.88MHz master
> oscillator effectively attenuates the phase noise from the 10MHz reference
> thus preventing it from affecting system performance.
>

More or less yes, in a PLL the intrinsic oscillator PN dominates above the
loop filter cutoff with the reference source PN dominating below the loop
filter cutoff.

So ideally what is wanted is a 10Mhz source with very good close in phase
noise (Which will pass straight through the filter), but where we don't
much care about phase noise at offsets much greater then the loop filter
bandwidth)  and a 122Mhz part with excellent phase noise performance above
loop filter cutoff but where we are prehaps prepared to compromise the
really close in PN.

I would expect the native phase noise of the 122.88MHz master oscillator to
> dominate the phase noise performance of the system. This would mean that a
> low-phase-noise reference (GPSDO or Rb) is not that necessary.
>
> Am I correct?
>

Yes, provided the loop filter bandwidth is sufficiently narrow, and in fact
there can be advantages to carefully picking the filter to optimize the
transition point to minimize total PN.

I had never really considered Rb or GPS to be a low phase noise source, low
drift yes, but low PN?

73 M0HCN.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20131214/b5d16da5/attachment-0004.htm>