[time-nuts] NMEA time off

Said Jackson saidjack at aol.com
Thu Mar 3 12:11:15 EST 2011

Hi Hal,

The almanac is helpful, but not required to track sats. Only the ephemeris info is needed, which is repeated every 30+ seconds or so if I remember correctly, which is why modern receivers can be locked in as little as 36 seconds or less from cold start without almanac or assistance from a network.

The gps to utc offset is for some crazy reason not transmitted in the ephemeris, but only in the almanac, so on average it will take about 6 minutes to receive this info and the nmea time may jump a second or more at that point. It may take a user until the next leap second happens to notice that their system has this error, and is causing problems..


Sent from my iPad

On Mar 3, 2011, at 3:24, Hal Murray <hmurray at megapathdsl.net> wrote:

>> I suspected it might have something to do with the almanac. The Tripmate has
>> been on now for 100 minutes, but has not been able to track four satellites
>> during that entire time. Does the GPS need to track at least four satellites
>> continuously during the  12.5 minutes it takes to download the almanac? If
>> so, then I'll probably need to place this thing outside tomorrow. 
> I'm pretty sure it only needs one satellite to get the almanac.  The whole 
> idea is that the almanac contains all the details about where the satellites 
> are and you need that to compute the Doppler so you know what frequency to 
> listen to.  It's sort of a chicken and egg problem.  If you start from 
> scratch, you just hunt (in frequency) until you get lucky and find a 
> satellite, then you get the almanac from it, then you can find the others.
>> I discovered two things about the Tripmate. First, the sensitivity sucks
>> indoors. It is currently showing 13 satellites visible, but only tracking
>> three.
> If you are indoors, I'm not surprised that an older unit is having troubles.
>> Second, the time data from the GPRMC sentence is about two seconds behind
>> UTC (using my calibrated eyeball for measuring).
> Some units are off-by-one.  As long as it is consistent, you can fix it up in 
> software.  I don't think I've seen any that are off by two, but it wouldn't 
> surprise me.
> Here is a hack I wrote when I was chasing that sort of glitch:
>  http://www.megapathdsl.net/~hmurray/ntp/ntpd/timing.c
> The first column is HHMMSS.sss (UTC from your local system) so it should be 
> easy to see if your system time matches the time from a NMEA device.
> -- 
> These are my opinions, not necessarily my employer's.  I hate spam.
> _______________________________________________
> 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 mailing list