[time-nuts] Low noise frequency multiplication

Grant Hodgson grant at ghengineering.co.uk
Thu Mar 1 09:20:08 EST 2007


 >It seems there are indeed many ways to kill a cat.

More ways than you could imagine...

IMHO, a 100MHz PLL is likely to give the best results, and is very easy 
to implement.  A 100MHz crystal oscillator should give very low phase 
noise and spur levels; crystals are readily available and there are 
quite a few circuits around.

Crystal oscillators running at 100MHz or so are widely used by radio 
hams for multiplication up to microwave frequencies so the phase noise 
and spur levels has to be low to start with, which a VCXO can achieve. 
Crystal oscillators can easily be disciplined with a varactor diode. 
Various PLL configuations exist to do this and the loop bandwidth is 
very narrow so the phase noise will not be degraded significantly by the 

There are other ways as well, but my vote would be for the VCXO as above.



More information about the time-nuts mailing list