See patent US5923618 and also http://www.leapsecond.com/notes/gpswnro.htm for details.
Based on what happened to the TS2100 (with Trimble ACE receiver) it looks like they tried to guess which 1024-week GPS epoch it was based on how high the leap second count was. And although this sounds good in theory, the earth is not cooperating with that plan. Not only are leap seconds unpredictable but the rate of leap seconds is even less predictable. In the 1970's there was an average of one leapsecond a year. In the 1980's and 1990's this dropped to about 1.5 years. And since 2000 it's down to one leap second every 4 years. That is, the earth has been gradually speeding back up, their hardcoded algorithm no longer well matches reality, and so they mispredicted which 1024 epoch GPS is in. I would recommend people use the Heol Design replacement receiver, which doesn't have this problem. /tvb ----- Original Message ----- From: "Björn" <[email protected]> To: "Discussion of precise time and frequency measurement" <[email protected]> Sent: Saturday, July 04, 2015 1:30 PM Subject: Re: [time-nuts] End of the World, or beginning of a new one? "Yes the Tymserve and other similar DATUM / Symmetricom products which use the Trimble ACE board will fail again in 2016. This is due to the way the GPS epoch is calculated inside the 2100 : the DATUM engineers estimated the future leap second insertions to determine GPS epochs, which was not a so good idea..." That was Trimble engineers... anyone with the actual algorithm they used? /Björn _______________________________________________ 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. _______________________________________________ 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.
