Re: [gpsd-dev] RFC 2783

On Tue, Mar 10, 2015 at 4:17 PM, Hal Murray <address@hidden> wrote:

> It has an external antenna connector that I'm using. It came with a battery
> holder that I never used. It locks quick, even with a long off, cold start.

What's your version of "quick"?

There's an LED that pulses when 1PPS starts. From memory, seemed like a minute or so. The GPS module is an MTK something (I'll have to look for where I stashed the manual)...

I just removed power from the GPS for about 25 minutes. Then I applied power and started watching the serial port. Attached is the output. It probably took me about 3-4 seconds between applying power and starting the "cat" of the serial port..

Without the battery, I'd expect all starts to be cold. There may be enough
capacitance in there to hold for a few seconds.

No battery or capacitors involved. It's currently powered by the USB bus of the PC, via a jumper wire. Between the GPS module & the RS-232 serial port is a level converter (3.3v -> 5v). I pulled the 5v USB jumper to power it off and left it that way for about 25 minutes.

This brings up the opportunity for an interesting bug. One of the things
that gets saved is the offset between GPS time and UTC. If you are running

What saves the offset? GPSd? NTPd? I'm running straight NTPd with this GPS.

in NMEA mode, there is no way for the device to say "I know where I'm located

I am running in NMEA mode. This GPS spits out data once per second, and I think once per minute it spits out additional NMEA statements. When using directly with NTPd, I turn off all sentences except GPGGA & GPRMC. I set NTPd to use the GPGGA sentence to be used for the second because it appears first, closest to the start of the second

After it's been running for a while, here's some of my NTP output:

[2.1-RC0][address@hidden]/root(1): ntpq -p

remote refid st t when poll reach delay offset jitter

==============================================================================

oGPS_NMEA(0) .GPS. 0 l 13 16 377 0.000 0.002 0.001

LOCAL(0) .LOCL. 12 l - 64 0 0.000 0.000 0.000

*tock.usshc.com .GPS. 1 u 13 64 377 28.339 2.381 1.570

+2001:4f8:2:d::1 .GPS. 1 u 36 64 377 67.043 0.228 1.084

+tick.apple.com .GPS. 1 u 55 64 377 69.472 2.118 1.681

-ns4.khresear.ch 192.5.41.40 2 u 64 64 377 37.733 3.506 0.721

+ntp.westbrook.c 204.163.51.41 3 u 16 64 377 68.148 1.479 1.020

-mirage.knovideo 129.6.15.29 2 u 16 64 377 36.957 5.110 2.886

+clock1.albyny.i .CDMA. 1 u 18 64 377 39.948 2.498 1.443

[2.1-RC0][address@hidden]/root(2): ntpdc -c peers

remote local st poll reach delay offset disp

=======================================================================

=tick.apple.com 69.243.176.6 1 64 377 0.06845 0.002231 0.03482

*GPS_NMEA(0) 127.0.0.1 0 16 377 0.00000 0.000002 0.01524

=tock.usshc.com 69.243.176.6 1 64 377 0.02834 0.002381 0.03812

=2001:4f8:2:d::1 2001:558:6033:2 1 64 377 0.06703 0.000228 0.03780

=clock1.albyny.i 69.243.176.6 1 64 377 0.03995 0.002498 0.05157

=LOCAL(0) 127.0.0.1 12 64 0 0.00000 0.000000 3.99217

=ntp.westbrook.c 69.243.176.6 3 64 377 0.06815 0.001479 0.04736

=mirage.knovideo 69.243.176.6 2 64 377 0.03696 0.005110 0.05220

=ns4.khresear.ch 69.243.176.6 2 64 377 0.03772 0.003506 0.05130

[2.1-RC0][address@hidden]/root(9): ntpq -c cv

assID=0 status=0000 clk_okay, last_clk_okay,

device="NMEA GPS Clock",

timecode="$GPGGA,155031.000,3909.1367,N,08543.4938,W,2,10,1.20,222.8,M,-33.5,M,0000,0000*52",

poll=3749, noreply=0, badformat=0, baddata=0, fudgetime1=0.000,

stratum=0, refid=GPS, flags=5

and the GPS time, but I don't know the UTC offset." It can either wait or
give you the time that may be wrong.

My first attempt with one of Gary's PPS over USB units was off by a second.
That would make sense if the firmware had a compiled in default from before
the last leap second.

--
These are my opinions. I hate spam.

From:	Frank Nicholas
Subject:	Re: [gpsd-dev] RFC 2783
Date:	Tue, 10 Mar 2015 18:10:12 -0400