[time-nuts] Is oscillator sync always bad?
bruce.griffiths at xtra.co.nz
Thu Dec 18 20:31:36 UTC 2008
Predrag Dukic wrote:
> Hi, Time -Nuts,
> Did anyone try to deliberately allow syncronisation of two
> oscillators, by , for example paralleling outputs of two 10811.
> I expect to see some benefits from the usual statistics: Phase
> noise divided by sqrt of 2 and also decreased amplitude random
> frequency jumps.
> Aging could also be average of the two.....
> Predrag Dukic
First ensure that doing this doesnt overstress the output amplifier cascade.
Then the two oscillators will need to be tuned so that they are very
close in frequency before injection locking can occur.
The higher the reverse isolation of the output amplifier and the higher
the crystal Q the closer the 2 oscillator frequencies will have to be.
The maximum frequency difference tolerable for injection locking to
occur can be estimated from Adlers equation.
More information about the time-nuts