NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Possible GPS Date Rollover problems week of April 6, 2019
From: Brad Morris
Date: 2019 Mar 27, 10:38 -0400
From: Brad Morris
Date: 2019 Mar 27, 10:38 -0400
Lu
In support of your assertion, I retrieved my 20 year old GPS receiver and powered it on. During satellite acquisition, it displayed what it thought was the correct date. The date displayed was not Jan 1980 or week zero. My thought was that it likely stored the date displayed in nonvolatile memory along with the week number. For this issue to affect my 20 year old device, it would have to remain powered off and stored without batteries for 2^10 weeks or 19.7 years, such that it would miss an entire 10 bit count or cycle of weeks.
I have now removed the batteries and stored it in anticipation of the roll over event. I predict no issues will occur, as the date it displays should not be before the date it has stored. In other words, when the rollover occurs, the week number from the satellites will be zero, implying a date in 1980. The device should then compare the stored and verified date against the date it has just computed. At which point, the device will correct for the rollover. This assumes that the firmware is coded correctly.
We will see.
Brad
On Wed, Mar 27, 2019, 10:18 AM Lu Abel <NoReply_LuAbel@navlist.net> wrote:
I would guess that the "GPS Rollover problem ... not seeming to get much press" is because all modern receivers (ie, anything manufactured in perhaps the last 20 years) handle it okay -- and that's the vast majority of GPS receivers (almost all cellphones have GPS receivers - and "smart" cellphones with GPSs didn't exist 20 years ago) Ditto for on-board GPS-equipped computers in automobiles. And, in fact, how many people even care if their GPSs display the correct date?? Most users simply want latitude and longitude.
