[aprssig] Looking for info on GPS clock synchronization

[scott at opentrac.org]

I've looked for this myself, and I think it's going to be somewhat
device-specific.  The only one I've ever found that actually lists this kind
of information is the Garmin GPS 18:

http://www.garmin.com/manuals/425_TechnicalSpecification.pdf

By default, the sequence starts with GPRMC and references the 1pps output
immediately preceeding it.  It doesn't say specifically how the timing of
the serial transmission relates to that pulse.

Good luck getting that sort of accuracy from serial alone - you're still
limited by the local RS-232 clock on the receiving end.  My OpenTrackers,
for example, get their serial clock from a 20ppm crystal.  Plus, being an
asynchronous protocol the clocks aren't synced anyway.  The UART's just
sampling at some multiple of the baud rate and doesn't care about start bit
edges.  At least, that's true of the devices I've used.

Even if you could get accuracy to 1/8 bit time, that's like 26 ppm at 4800
baud.

Try timing the interval from one $GPRMC sentence to the next - could be
there's NO specific timing relationship, other than that it's sent sometime
during the second in question.

If you need 1-3ppm accuracy, you're just going to have to get a unit with a
1pps output.  Deluo's units don't have one anymore, so I recommend the GPS
18.

Scott
N1VG

> -----Original Message-----
> From: aprssig-bounces at lists.tapr.org 
> [mailto:aprssig-bounces at lists.tapr.org] On Behalf Of Ben Jackson
> Sent: Sunday, July 23, 2006 4:05 PM
> To: TAPR APRS Mailing List
> Subject: [aprssig] Looking for info on GPS clock synchronization
> 
> I'm looking for information about what clock information can 
> be extracted
> from a standard NMEA string.  For example, which moment 
> during sentence
> transmission is associated with the timestamp(s), whether GPS 
> units that
> only output 1s time resolution typically send sentences aligned to the
> second, whether there's any spec for the baud timing accuracy, or the
> start intervals of sentence groups.
> 
> Basically anything to can do with a GPS that doesn't have PPS output
> to get into the 1-3ppm accuracy range.
> 
> -- 
> Ben Jackson AD7GD
> <ben at ben.com>
> http://www.ben.com/
> 
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig
> 
>