Hal wrote:
It would be interesting to measure the propagation delay over a day or week, and watch the PLL error voltage over a scale of seconds or minutes.
Somewhere I probably still have miles and miles of paper tape that came out of a WWVB phase comparator for many years BGE (before the GNSS era). In addition to the slow tracks, which were recorded 24/7, there were also fast tracks taken periodically. At the time, it was the only practical way to maintain NIST traceability of time and frequency measurements.
The take-home points from this data (at least for those of us who are a half-continent or more distant from Ft. Collins) are (i) that the propagation delay varies quite a lot -- 10s of ppm normally, more during periods of unsettled space weather; (ii) that phenomena too numerous to count contribute to the variability, so the resultant error is a complex epicyclic function of phenomena with periods from hours to years overlaid with random noise that can be much greater than the cyclic variations; and (iii) that the amplitudes of many of the lesser cyclic errors do not fall off rapidly compared to the amplitudes of the stronger ones, so simple approximations of the epicyclic features do not allow predicting the propagation delay more closely than about tenths of a ppm (parts in e-7).
Been there, done that -- for many years -- and I'm very glad we're rid of the need for it. Sometimes, progress is ... well ... progress.
Best regards, Charles _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
