[hpsdr] CVCSO-914-1000

[Grant Hodgson]

The CVSO-914-1000 is certainly a good part, but being SAW based means 
that the close-in phase noise is not as good as a well-designed crystal 
oscillator such as the ones that Chris GW4DGU has recently reported.

My quick calculations suggest that using the CVSO would worsen the phase 
noise by about 8-9dB or so, compared to using a crystal oscillator at 
116MHz and then using the 9912s on-chip PLL clock multiplier.  If an 
external x8 multiplier were used, the resultant phase noise would be 
slightly better, but for now I believe that the working assumption for 
Phoenix is to use a VHF crystal oscillator with the 9912 PLL.  This is 
felt to be a good compromise between performance/cost/complexitiy/design 
effort etc., and should result in an LO that will outperform the Orion 
on all bands at all carrier offsets, with the exception of 28MHz at 
offsets closer than about 1kHz or so - and even then Phoenix should be 
better than the SDR-1000 and hopefully better than the Flex 5K.

My personal belief is that it is the close-in phase noise that 
determines the 'quality' of an LO; it can be a particularly difficult 
parameter to optimise, and can also be quite difficult to measure.  But 
the effects are real - poor close-in phase noise on either transmit or 
receive can cause interference with only a single, strong signal on the 
band.  So it's worth spending some time and effort aiming for a very 
clean LO; Phoenix should in theory be able to do this.

regards

Grant   G8UBN


> Subject:
> [hpsdr] CVCSO-914-1000
> From:
> "Philip Covington" <p.covington at gmail.com>
> Date:
> Sun, 24 Feb 2008 15:46:50 -0500
> To:
> "HPSDR Reflector" <hpsdr at hpsdr.org>
> 
> To:
> "HPSDR Reflector" <hpsdr at hpsdr.org>
> 
> 
> Maybe a candidate for use with the AD9912 DDS?
> 
> <http://www.crystekcrystals.com/crystal/spec-sheets/vcxo/CVCSO-914-1000.pdf>
> 
> It is not cheap (~$116 ea) but is in stock at Mouser for one...
> 
> 73 Phil N8VB
> 
> 

 1203940224.0