time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Re: [time-nuts] 60hz disciplined watch follow up

HM
Hal Murray
Thu, Apr 21, 2011 7:29 PM

My estimate of the citizen ecodrives automatic rate adjustment occures at 10 day intervals.

That only works if you can track the signal continuously.

Suppose you lose the signal for a while.  How many cycles was that?  If it's
only a short while and your clock is stable enough, you can assume you will
come back on the right cycle and only be off a bit on the phase.  Your clock
stability determines how long a "short" while is.

Now consider that your watch will be out of contact for 8 hours while you
sleep.  If it was accurate to 1 cycle in 8 hours, you wouldn't need to go
through this much effort to calibrate it.

This is similar to the problem of RS-232 clock accuracy.  The receiver needs
to be able to determine the difference between N and N+1 identical bits.

--
These are my opinions, not necessarily my employer's.  I hate spam.

> My estimate of the citizen ecodrives automatic rate adjustment occures at 10 day intervals. That only works if you can track the signal continuously. Suppose you lose the signal for a while. How many cycles was that? If it's only a short while and your clock is stable enough, you can assume you will come back on the right cycle and only be off a bit on the phase. Your clock stability determines how long a "short" while is. Now consider that your watch will be out of contact for 8 hours while you sleep. If it was accurate to 1 cycle in 8 hours, you wouldn't need to go through this much effort to calibrate it. This is similar to the problem of RS-232 clock accuracy. The receiver needs to be able to determine the difference between N and N+1 identical bits. -- These are my opinions, not necessarily my employer's. I hate spam.