[time-nuts] Epoch rollover?

[Magnus Danielson]

On 10/19/2011 11:18 PM, Bruce Lane wrote:
> Hi, gang,
>
> 	Did we just have another GPS epoch rollover? My trusty old Odetics 425
> seems to believe the date is March 4th, 1992.

Units can have their internal shifted rollovers. This is well 
illustrated using a GPS date calculator:

http://csrc.ucsd.edu/scripts/convertDate.cgi?time=1992+03+04

http://csrc.ucsd.edu/scripts/convertDate.cgi?time=2011+10+19

If you look at GPS Week you see that it beleives it has GPS week 634 
(which is what the GPS birds transmit since they do modulo 1024) while 
actual GPS week is 1658, which has the expected difference of 1024.

A GPS receiver can operate correctly even if they do modulus 1024 
errors, it is just annoying that the time is off.

> 	I could probably correct it in firmware, if I looked hard and long
> enough, but the ToD is still correct and the frequency standard is
> staying nicely locked. Not sure if recovering the correct Julian date is
> worth the effort.

It would be a very limited part of the firmware in that case. Should be 
an offset date hidden in there somewhere.

Cheers,
Magnus