[time-nuts] GPSDO and holdover
bob at evoria.net
Tue Apr 22 23:03:52 EDT 2014
Thanks for confirming that these lost PPS pulses are happening to someone else! I've looked through my interrupt code several times to see if I'm missing something. Please let me know if you find anything for sure. I've been thinking of saving the values for the satellites at each tick to see if it's related to AOS/LOS, but there's been too much else to do.
I've already faced most of the points you make, and yeah, the decision on whether to have ideal time or ideal frequency has been a difficult one. But, there's only so much I can do with a nav receiver. I'll address it when a timing receiver goes in. I think I'll go ahead and loosen up my definition of what holdover is, as a strict interpretation seems to cause more damage than it prevents.
> From: Tom Van Baak <tvb at LeapSecond.com>
>To: Discussion of precise time and frequency measurement <time-nuts at febo.com>
>Sent: Tuesday, April 22, 2014 8:57 PM
>Subject: Re: [time-nuts] GPSDO and holdover
>Hi Bob S,
>I have noticed skipped 1PPS on the Adafruit GPS also. Some days are clean, other days miss a few samples. I have not explained it yet. I plan to run two GPS boards and two counters to narrow down the cause.
>In any event, IMHO, a GPSDO should not go crazy if glitches like this occur. I don't think it should go into holdover for one missed sample, or even a few missed samples. But then you need to define what holdover is. I mean, by some definition a GPSDO is in "holdover" between every second.
>I do not think there is any standard. Just conventions: some documented, some not. It's attention to a dozen little details like this that separate a quick hack GPSDO from a quality one. You'll also face a number of design issues at startup, and coming out of holdover. Lastly, you get to choose between it being an ideal time standard vs. an ideal frequency standard.
>p.s. Please fix your address book. The correct email for the list is time-nuts at febo.com
>----- Original Message -----
>From: Bob Stewart
>To: time-nuts-owner at febo.com
>Sent: Tuesday, April 22, 2014 6:03 PM
>Subject: GPSDO and holdover
>I hope I haven't asked this before, but is there a standard way of deciding to go into holdover mode? I'm still wrapping up code for this Adafruit, and as I've posted before: every now and then it skips a PPS. I'm trying to decide whether to allow a free pass (if not followed by another skip within some timeframe) or to immediately stop processing any further PPS pulses until I decide based on some criteria that they're reliable.
>Bob - AE6RV
>time-nuts mailing list -- time-nuts at febo.com
>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>and follow the instructions there.
More information about the time-nuts