Re: [time-nuts] Garmin 18x GPS WNRO, revisited

2019-11-30 Thread Rich Wales
My GPS has been powered down from time to time over the past 10 years.
Its recent WNRO issue was not associated with any power-cycling event.

I am using ntpsec's "SHM" driver (*not* the type-20 NMEA driver), with
the GPS data captured by gpsd.  Gpsd is sensing the PPS signal from the
GPS without difficulty.

I bought a new Garmin 18x LVC recently, btw, and it is running fine on
another computer at my home without any WNRO problems.  I assume the
firmware on the new unit is newer than on my other, 10-year-old GPS, and
that this is why the new GPS is giving good time without my needing to
fudge it in ntpsec.

Rich Wales
ri...@richw.org

On 2019-11-29 10:27, David J Taylor via time-nuts wrote:

Thanks for reporting that, Rich.  I haven't seen that issue with either
the GPS18 LVC, or the GPS18x LVC, but neither have been powered down. 
Were yours running continually?  Anyway, I love the fix!

I'm running reference NTP here, and noted that with a type 22 PPS
driver, the "prefer" must include the type 20 NMEA driver.  With the
problem pending, I had hoped that using an e.g. Internet source for
coarse time might have been enough, and had commented out the NMEA
source, but it appears not to be the case.  The PPS was never used (but
it appeared in the billboard).

Cheers,
David

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.


Re: [time-nuts] Garmin 18x GPS WNRO, revisited

2019-11-29 Thread Fiorenzo Cattaneo
The Garmin 18x I have still works, and I verified it works after a power
cycle. I'm running firmware version 4.40.

Apparently all the 4.XX firmware versions are related to GPS week rollover
fixes. Earlier versions also have fixes for "improved" time keeping. Here
is the download link for 18X:

https://www8.garmin.com/support/download_details.jsp?id=4055

These are all the downloads for the 18x, including the Garmin tool
SNSRXCFG:

https://www8.garmin.com/support/collection.jsp?product=010-00321-31


This link from David might also be of interest:

https://www.satsignal.eu/ntp/Garmin-GSP18x-LVC-firmware-issue.htm


I've found the upgrade process to be a bit complicated, the upgrade
instructions didn't work for me. So this is what I did:
* set receiver at 9600 baud, power cycle
* with garmin utility, set receiver in garmin mode, power cycle
* run upgrade utility, power cycle,
After the upgrade I couldn't get the receiver back to NMEA mode, so I just
did it programmatically:

/* fd is the file descriptor pointing to the open file to the tty device
in RAW mode @ 9600 baud */
/* send_byte is a wrapper code to write() which handles timeouts and write
errors */
void garmin_set_nmea(int fd)
{
printf("==> garmin_set_nmea\n");
/*
 * garmin binary command to switch to NMEA output
 */
int i;
unsigned char buffer[] = { 0x10, 0x0A, 0x02, 0x26, 0x00, 0xCE,
0x10, 0x03 };
for (i = 0; i < sizeof (buffer); i++)
send_byte(fd, buffer[i]);
}


-- Fio Cattaneo

Universal AC, can Entropy be reversed? -- "THERE IS AS YET INSUFFICIENT
DATA FOR A MEANINGFUL ANSWER."

-Original Message-
From: time-nuts  On Behalf Of David J
Taylor via time-nuts
Sent: Friday, 29 November, 2019 10:28
To: Discussion of precise time and frequency measurement

Cc: David J Taylor 
Subject: Re: [time-nuts] Garmin 18x GPS WNRO, revisited

From: Rich Wales

I may have found the answer to my own question.

I added "time1 619315200" to the "refclock" line in my ntp.conf file, and
restarted ntpsec, and it appears to be sane once again.

(And, of course, 619315200 = 7,168 days = 1,024 weeks.)

Rich Wales
ri...@richw.org


Thanks for reporting that, Rich.  I haven't seen that issue with either
the
GPS18 LVC, or the GPS18x LVC, but neither have been powered down.  Were
yours running continually?  Anyway, I love the fix!

I'm running reference NTP here, and noted that with a type 22 PPS driver,
the "prefer" must include the type 20 NMEA driver.  With the problem
pending, I had hoped that using an e.g. Internet source for coarse time
might have been enough, and had commented out the NMEA source, but it
appears not to be the case.  The PPS was never used (but it appeared in
the billboard).

Cheers,
David
--
SatSignal Software - Quality software for you
Web: http://www.satsignal.eu
Email: david-tay...@blueyonder.co.uk
Twitter: @gm8arv


___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.


Re: [time-nuts] Garmin 18x GPS WNRO, revisited

2019-11-29 Thread David J Taylor via time-nuts

From: Rich Wales

I may have found the answer to my own question.

I added "time1 619315200" to the "refclock" line in my ntp.conf file,
and restarted ntpsec, and it appears to be sane once again.

(And, of course, 619315200 = 7,168 days = 1,024 weeks.)

Rich Wales
ri...@richw.org


Thanks for reporting that, Rich.  I haven't seen that issue with either the 
GPS18 LVC, or the GPS18x LVC, but neither have been powered down.  Were 
yours running continually?  Anyway, I love the fix!


I'm running reference NTP here, and noted that with a type 22 PPS driver, 
the "prefer" must include the type 20 NMEA driver.  With the problem 
pending, I had hoped that using an e.g. Internet source for coarse time 
might have been enough, and had commented out the NMEA source, but it 
appears not to be the case.  The PPS was never used (but it appeared in the 
billboard).


Cheers,
David
--
SatSignal Software - Quality software for you
Web: http://www.satsignal.eu
Email: david-tay...@blueyonder.co.uk
Twitter: @gm8arv 



___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.


Re: [time-nuts] Garmin 18x GPS WNRO, revisited

2019-11-29 Thread Rich Wales
I may have found the answer to my own question.

I added "time1 619315200" to the "refclock" line in my ntp.conf file,
and restarted ntpsec, and it appears to be sane once again.

(And, of course, 619315200 = 7,168 days = 1,024 weeks.)

Rich Wales
ri...@richw.org

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.