[time-nuts] How to get 32.768KHz from 10MHz.

[Bruce Griffiths]

Bruce Griffiths wrote:
> An oscillator can be injection locked to at frequency that is a rational 
> number (M/N where M, N are integers ) multiplier of the injection frequency.
> Thus, in principle, a 32768Hz oscillator can be injection locked 
> directly to a 10MHz signal (32768Hz = (256/78125)*10MHz) without 
> requiring any external dividers or multipliers.
> In practice this will only be practical with a 32768Hz crystal 
> oscillator whose frequency determining crystal is tuned sufficiently 
> closely to 32768Hz.
> Note this is not quite the same as tuning the free running oscillator 
> frequency.
> The sensitivity to the injected signal can be optimised if the 32768Hz 
> oscillator is suitably designed.
>
> The 32768Hz injection locked oscillator output can be used to injection 
> lock another 32768Hz crystal oscillator whose design cannot be optimised 
> for direct injection locking to a 10MHz signal.
>
> Bruce
>
>   
The major problem with this approach is the extremely small (<<1ppm) 
locking range achievable.
Thus it isnt likely to be practical with standard 32768Hz crystals, as 
even a relatively small temperature change will perturb the tuning 
sufficiently to make locking to the desired frequency impossible.

Bruce