[time-nuts] My Garmin 18x, Ver 3.50, currently 1 second slow to UTC

[Cook Mike]

Le 29/12/2010 08:32, David J Taylor a écrit :
> Calling any Garmin 18x users,
>
> I noticed this morning that my Garmin 18x was giving the time 1 second 
> slow to UTC - it has been running in sync with UTC for about 2 years now.
> []
> So I wondered, any other folk out there with a Garmin 18x - is it 1 
> second slow to UTC ?
>
> Regards, Kiwi Geoff (Christchurch, New Zealand).
>
> ________________________________________
>
>
> Geoff,
>
> I see the same effect on my GPS 18x LVC with the 3.50 firmware.  Have 
> you reported this to Garmin?
>
> To me it's unacceptable that the NMEA output is delayed into the next 
> PPS signal, making the unit much less useful.  I measure the NMEA 
> output and it's not even a consistent distance delayed from the PPS, 
> varying quite a bit.
>
It is possible that this has been reported as I do not see 3.50 
available on garmin.com. 3.10 is the latest. Could not find anything in 
faq's or alerts though. It would certainly be useful to have a sentence 
transmission offset from UTC in the NMEA messages (or at least a new 
message heading/tailing the list with that in). The NTP driver has a 
fudge factor for time offset to end of NMEA stream which can be used if 
you can measure it. You could scope it, (I am checking a MG1613S at the 
moment) but that is not that easy if you are selecting just one of the 
many messages being output. An easier method might be to just have NTP 
tell you what it thinks the offset is when measured against other 
references, then use that as your fudge factor.(use noselect directive 
for the NMEA server or set its stratum high so it doesn't get selected).

> [New to this mailing list, so I had to fabricate this message as a 
> reply to at least try and keep it "on thread"]
>
> Cheers,
> David