[time-nuts] FE-.5680A trimming resolution

[Magnus Danielson]

On 02/07/2012 03:12 AM, Hal Murray wrote:
>
> magnus at rubidium.dyndns.org said:
>> Not quite right. If you lock up the clock, you do not lock to the birds,
>> but to GPS time or UTC as received over GPS. The observed time of the  birds
>> would be a bad solution since you can't see a particular bird  continously
>> unless you is in geosync orbit.
>
> It's even more complicated than that.
>
> Suppose you decide to lock onto a bird that you can see.  You have to correct
> for Doppler so you can't use a simple PLL.

I considered writing that, but realizing that just not sweeing the same 
bird continuously 24x7 should be a great hint that you have done 
something wrong. The doppler is the second killer and then even if you 
compensat that, each bird has their own frequency error and drift, which 
is compensated by the nav message, but then you no long follow the bird, 
but GPS time.

You can however use a simple PLL, the PLL in the tracking loop isn't 
particularly advanced or special. It's not even aided except for initial 
track-in. It's when you no longer wishes to track the doppler shifts 
that you need to handle it, but you can compensate that out if you know 
your position and even produce an analogue signal predicing and 
cancelling the effect, in which case only the residue between predicted 
and actual doppler will remain visible.

Cheers,
Magnus