[time-nuts] WWVB D-PSK document
magnus at rubidium.dyndns.org
Sun Mar 31 21:23:30 EDT 2013
True. The station itself doesn't swing +/- 100 ppm, but the channel you
recieve it from can alter delay quick enough to annoy you. But the
phase-change needs to put in context with the time it takes to make the
phase-change in order to come up with a frequency error.
However, if it shifts a cycle in say 60 s, then it will be 1/(60000*60)
about 2.77E-7 in relative frequency, which is small enough that an
unregulated XO variations become a larger issue.
On 04/01/2013 02:29 AM, Bob Camp wrote:
> But can you ignore the WWVB signal swing? The beast flips a cycle (or more) as you hit sunrise or sunset. A cycle at 60 KHz is a lot of ppm. Weather this happens fast enough to actually turn it into a tuning issue is the question.
> On Mar 31, 2013, at 7:36 PM, Magnus Danielson<magnus at rubidium.dyndns.org> wrote:
>> Uhm, yes. You want your pulling-range to handle all the drift and tempco you reasonably can assume, and also that of the application and signal. Since the signal is WWVB in this case, we can ignore that factor. Also, you don't want to be completely de-tuned from start either. I'm sure John knows this already.
More information about the time-nuts