FWIW, the SkyTraq receivers have the notion of a “UTC reference date” which 
allows the 1024 week window to roll forward indefinitely as long as you update 
that value once in a while. I’ve written code in my GPS Clock to update the 
reference date once a year as well as update the default GPS-UTC delta whenever 
it’s wrong (this fixes the “off by 2 seconds for a minute or two” problem). 
This latter issue is an interesting one - If you supply SkyTraq modules with 
backup power, then on *power up* it will copy the GPS-UTC delta to flash, as 
long as the memory of what it was at shut down remains preserved.

The UTC reference date also allows the clock to workaround the Y2K issue in the 
NMEA sentences - you can query the reference date and use a 100 year rolling 
window similarly for that. As long as you power the clock up and give it 
reception at least once every 19 years or so, it should calculate the correct 
date as long as GPS remains available (this only really matters for the clock 
because it’s how DST decisions are made - it doesn’t actually display the date).

I haven’t done this for my GPSDOs because the controller isn’t wired to speak 
to the GPS receiver. But you can use the diagnostic port and SkyTraq’s GNSS 
viewer software to do this if you care to. The only use the GPSDO makes of the 
time/date is time-stamping the logs.

> On Jul 27, 2017, at 7:01 PM, Tom Van Baak <t...@leapsecond.com> wrote:
> 
>> Well quite an unpleasant surprise. So after the 30th do the TBolts stop
> 
> Paul,
> 
> This topic has been covered a number of times over the years. Some time-nuts 
> have even run TBolt's under GPS simulators to verify that the 10 MHz and 1PPS 
> outputs will be fine. So apparently the only effect is that the date & time 
> (in binary TSIP messages) are off by 1024 weeks. This rollover-related effect 
> is by now a "common" issue with many GPS receivers.
> 
> The current version of Mark's Lady Heather program has code to detect this 
> and fix it so you're good to go for the next 19.6 years.
> 
> /tvb
> 
> ----- Original Message ----- 
> From: "paul swed" <paulsw...@gmail.com>
> To: "Discussion of precise time and frequency measurement" 
> <time-nuts@febo.com>
> Sent: Thursday, July 27, 2017 5:36 PM
> Subject: Re: [time-nuts] Trimble Thunderbolt no longer determines thecorrect 
> date
> 
> 
>> Well quite an unpleasant surprise. So after the 30th do the TBolts stop
>> working or is it a case of just the wrong date? I know my Hp3801s been
>> working just fine and its old. Is the TBolt the same issue. Wrong date but
>> still locks thats all I care about actually.
>> With to respect of some sort of a hack I can see that being fairly
>> difficult.
>> Regards
>> Paul
>> WB8TSL
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to