[ntp:questions] Odd offset for PPS DCD w/ Garmin GPS 18x LVC

Dave Hart davehart at gmail.com
Thu Mar 24 04:17:00 UTC 2011

On Thu, Mar 24, 2011 at 02:19 UTC, lellis <larry.ellis at gmail.com> wrote:
> Well, I have used the ever-faithful binary search on the Windows
> builds to try to find where the strange offset introduced itself.

I'm impressed by your patience with the tedium required.

> The first build to show the odd offset (between NMEA and PPS ATOM) is
> 4.2.7p24.
> 4.2.7p23 seems to behave normally.
> The offset difference between PPS and NMEA is about 125 ms  (e.g. NMEA
> -10, PPS +115) on 4.2.7.p24, on p23 the offsets of the two drivers are
> about the same.

4.2.7p24 came out in April 2009.  Staring at the comments for the last
year's changesets:


Nothing between p23 and p24 jumps out at me as likely related. I
wonder if this cset from October 2010 (first included in 4.2.5p237-RC)
is what you were looking for:


You might try configuring your GPS to emit two sentences each second
and configuring NMEA (via ntp.conf server ... mode X) to process the
sentence your GPS emits 2nd each second.  That will eliminate the
windows ntpd user PPS hack which doesn't really fit with the split of
fudge time1/time2 by NMEA.

I use NMEA + PPS myself on ntp.davehart.net, two different ways.  I
have two GPS 18x LVCs wastefully connected to the same machine (for
lack of a better box to use the 2nd GPS with at the moment.)  From

# puck on com1 - user PPS for NMEA, plus separate ATOM driver using PPSAPI
server mode 0 minpoll 4 prefer
fudge refid uPPS flag1 0 flag4 1 time1 -0.0067295 time2 -0.0067295
server minpoll 4
fudge refid kPPS time1 -0.0067295
# puck on com2 - integrated NMEA+PPSAPI
server mode 0 minpoll 4 prefer
fudge flag1 1 flag4 1 time1 -0.0067295 time2 -0.0067295

ntpq -p excerpt with whitespace removed to attempt to avoid wrapping:

*GPS_NMEA(1)  .uPPS.  0 l   13   16  377  0.000  -0.235  0.007
xPPS(1)       .kPPS.  0 l   12   16  377  0.000  -0.025  0.005
oGPS_NMEA(2)  .GPS.   0 l   11   16  377  0.000  -0.019  0.003

The ntpd version is close to 4.2.7p139 (it may have part of p140 or
p141, I tend to update that ntpd using in-progress bits between
releases attempting to exercise changes before they're integrated).

> Again, I am using the older Garmin firmware, v3.20.  I am not
> particulaly keen to move to a different firmware version because this
> version is working well, but I'll be happy to test any new builds on
> this firmware, if helpful.

You're wise to use 3.20 from what I gather.  I have two units
precisely because I hit the widely-reported bug in 3.10 and earlier
(IIRC) that can cause a GPS 18x LVC to go deaf and dumb until sent in
for repair or simply left unpowered for a month.  I decided I liked
the unit enough to want to minimize the wait and ordered a 2nd unit
from the same vendor who wired up my first one (mentioned on this list
repeatedly already).

3.30 was the first to have the much-later-in-the-second NMEA bug/issue
present in all later versions so far.

Dave Hart

More information about the questions mailing list