[time-nuts] Disciplining Rubidium

[Bruce Griffiths]

John Ackermann N8UR wrote:
> John Miles wrote:
>   
>> Phase noise generally gets better with the higher-frequency OCXOs, though.
>> I think the best of all possible worlds would be a 5-MHz OCXO like the one
>> you describe, being used to discipline a 10 MHz or higher-frequency part.
>>
>> -- john, KE5FX
>>     
>
> I'm not sure about that -- at least, the Wenzel ULNs show better noise 
> at small offsets for the 5 MHz than the 10 MHz versions (though the 
> floor is the same).
>
>   
John

Surely the fact that the product of quartz resonator Q and its resonant 
frequency is approximately constant has something to do with this via 
the Leeson effect?
Higher frequency crystals tend to have lower Q and hence lower close in 
phase noise at a given offset this being somewhat exacerbated by the 
fact that the higher frequency crystal tends to have a wider resonator 
bandwidth.
The way to improve this is to find a substance with a larger Q at a 
given frequency.

Bruce