albertson.chris@gmail.com said:

That "balance" is assuming a classic ADEV type V curve.  That assumption
works better in old non-green systems that burned lots of power in an idle
loop when they didn't have any work to do.  With modern systems, it's easy to
change the temperature significantly by starting/finishing a big chunk of
work.

External temperature changes can also be important.  Every now and then, I
like to repost this link.  This seems like a good opportunity.
NTP temperature compensation
http://www.ijs.si/time/temp-compensation/

--
These are my opinions.  I hate spam.

i've seen that myself, NTP's compensation can almost be used as a way to
measure room temperature.  You can see the AC cycles.  For this reason I
keep the NTP server in a walk in closet that lacks heating duct.  The room
temp cycles slowly.

I've also read but can't remember where about a project to replace the TTL
can oscillator in a PC with a better quality one driven from a OCXO.  That
is a lot of work for small gain.

One thing about that study.  I notice the author had t remove the PPS
reference clock.  It looks like he was unabe to get the graph to move at
all with PPS attached.  I'd like to see the same study done with a good
PPS reference clock let connected.  Likely the results would be
uninteresting, just a flat graph.

On Wed, May 15, 2013 at 12:51 AM, Hal Murray hmurray@megapathdsl.netwrote:

--

Chris Albertson
Redondo Beach, California

Hi!

Thank you for your input.

Yes, I've set minpoll and maxpoll to 4 yet the system corrects the offset
slowly during the calibration interval that starts I believe at 8 seconds
(2^3) and stops at 256 seconds (2^8) even though I've set min and maxpoll.

Even considering this situation I am going crazy with the Trimble Acutime
Gold I installed on my roof.

I connected the port A to a RS422-to-RS232 converter. The PPS+ signal is
directly connected to the DCD pin of this RS232 port. The port B is also
connected to a RS422-to-RS232 converter and I occasionally connect a
Windows computer to check the status of the antenna.

I've reconfigured the antenna to output only NMEA sentences (and I also use
the PPS) and NTP works fine but suddenly without any explanation the NTP
process stops receiving information from the receiver. I really don't
understand what's happening here as the receiver is on the roof and sees
plenty of satellites:

56427 70907.634 127.127.20.0
$GPGGA,194147.0,4055.227,N,00829.618,W,1,07,1.14,00264,M,051,M,,*48
56427 70923.635 127.127.20.0
$GPGGA,194203.0,4055.227,N,00829.617,W,1,07,1.14,00262,M,051,M,,*42
56427 70939.634 127.127.20.0
$GPGGA,194219.0,4055.226,N,00829.616,W,1,06,1.47,00258,M,051,M,,*47
56427 70955.635 127.127.20.0
$GPGGA,194235.0,4055.226,N,00829.615,W,1,07,1.14,00251,M,051,M,,*44
56427 70971.634 127.127.20.0
$GPGGA,194251.0,4055.227,N,00829.616,W,1,07,1.13,00255,M,051,M,,*47

Can anyone help me please?

Cheers,
Miguel

On 15 May 2013 18:33, Chris Albertson albertson.chris@gmail.com wrote:

Le 15 mai 2013 à 21:43, Miguel Barbosa Gonçalves a écrit :

these look ok

The NMEA driver is very stable so my guess would be the 422/232 converter. As you have 2 converters you could try switching them to see if the problem moves.

On Wed, May 15, 2013 at 12:43 PM, Miguel Barbosa Gonçalves m@mbg.pt wrote:

Even if there were zero satellites the GPS receiver should report zero, not
just stop and die.

Are you certain that both PPS and NMEA data both stop?  If just one of
them, then I'd suspect a bad connection, a loose cable or something.  If
both then I'd suspect the receiver or its power supply or the rs422 to
rs232 converter or its power supply.  Power is the easiest thing to check.

Chris Albertson
Redondo Beach, California

On 16 May 2013 04:14, Chris Albertson albertson.chris@gmail.com wrote:

Hi Chris!

Thanks for your help. The strange thing is that is works and suddenly it
stops.

I changed the offset of the NMEA data in NTP and increased the mindist to
0.250.
It is running since 21:12 UTC without problems. I will let it run for 48
hours before
entering production.

Kind regards,
Miguel

On 16 May 2013 11:12, Miguel Barbosa Gonçalves m@mbg.pt wrote:

I was monitoring this server closely and here's what happened

56428 27303.926 0.000006800 158.256 0.000000396 0.005945 4
56428 27319.926 0.000006325 158.256 0.000000552 0.005561 4
56428 27335.926 0.000005382 158.256 0.000000673 0.005202 4
56428 27351.926 0.000005656 158.256 0.000000786 0.004866 4
56428 27367.926 0.000004605 158.256 0.000000498 0.004552 4
56428 27383.926 0.000004386 158.256 0.000000644 0.004258 4
56428 27399.926 0.000003148 158.256 0.000000315 0.003983 4
56428 27415.926 0.000002788 158.256 0.000000485 0.003725 4
56428 27431.926 0.000002173 158.256 0.000000569 0.003485 4
56428 27447.926 0.000001682 158.256 0.000000543 0.003260 4

56428 27463.926 0.000000393 158.256 0.000001338 0.003049 4
56428 27479.926 -0.000695347 258.256 0.000174991 35.355339 4

56428 27495.925 -0.002219770 258.256 0.000186611 33.071891 4
56428 27511.926 -0.003650615 258.256 0.000175434 30.935922 4
56428 27527.926 -0.004993533 258.256 0.000164605 28.937905 4
56428 27543.925 -0.006255032 258.256 0.000154519 27.068932 4
56428 27559.925 -0.007438843 258.256 0.000145242 25.320667 4
56428 27575.925 -0.008550950 258.256 0.000136390 23.685315 4
56428 27591.926 -0.009594694 258.256 0.000128180 22.155584 4
56428 27607.925 -0.010574400 258.256 0.000120117 20.724651 4
56428 27623.925 -0.011495082 258.256 0.000112889 19.386136 4
56428 27639.925 -0.012358729 258.256 0.000105844 18.134069 4

56428 27303.926 127.127.20.0
$GPGGA,073503.0,4055.226,N,00829.616,W,1,10,0.91,00257,M,051,M,,*41
56428 27319.926 127.127.20.0
$GPGGA,073519.0,4055.228,N,00829.619,W,1,08,1.15,00269,M,051,M,,*42
56428 27335.926 127.127.20.0
$GPGGA,073535.0,4055.228,N,00829.619,W,1,08,1.15,00270,M,051,M,,*44
56428 27351.926 127.127.20.0
$GPGGA,073551.0,4055.228,N,00829.619,W,1,08,1.15,00269,M,051,M,,*4E
56428 27367.926 127.127.20.0
$GPGGA,073607.0,4055.228,N,00829.618,W,1,08,1.15,00270,M,051,M,,*47
56428 27383.926 127.127.20.0
$GPGGA,073623.0,4055.228,N,00829.620,W,1,09,1.04,00275,M,051,M,,*4E
56428 27399.926 127.127.20.0
$GPGGA,073639.0,4055.228,N,00829.620,W,1,09,1.04,00277,M,051,M,,*47
56428 27415.926 127.127.20.0
$GPGGA,073655.0,4055.227,N,00829.620,W,1,09,1.04,00275,M,051,M,,*40
56428 27431.926 127.127.20.0
$GPGGA,073711.0,4055.228,N,00829.617,W,1,09,1.04,00262,M,051,M,,*4C
56428 27447.926 127.127.20.0
$GPGGA,073727.0,4055.228,N,00829.617,W,1,09,1.04,00262,M,051,M,,*49

56428 27463.926 127.127.20.0
$GPGGA,073743.0,4055.229,N,00829.618,W,1,09,1.04,00268,M,051,M,,*4F
56428 27479.926 127.127.20.0
$GPGGA,073759.0,4055.228,N,00829.619,W,1,09,1.05,00273,M,051,M,,*4F

56428 27495.925 127.127.20.0
$GPGGA,073815.0,4055.228,N,00829.619,W,1,10,0.91,00268,M,051,M,,*46
56428 27511.926 127.127.20.0
$GPGGA,073831.0,4055.228,N,00829.617,W,1,10,0.91,00262,M,051,M,,*44
56428 27527.926 127.127.20.0
$GPGGA,073847.0,4055.229,N,00829.617,W,1,10,0.91,00263,M,051,M,,*45
56428 27543.925 127.127.20.0
$GPGGA,073903.0,4055.229,N,00829.617,W,1,10,0.92,00264,M,051,M,,*40
56428 27559.925 127.127.20.0
$GPGGA,073919.0,4055.229,N,00829.617,W,1,10,0.92,00265,M,051,M,,*4A
56428 27575.926 127.127.20.0
$GPGGA,073935.0,4055.228,N,00829.617,W,1,10,0.92,00265,M,051,M,,*45
56428 27591.926 127.127.20.0
$GPGGA,073951.0,4055.229,N,00829.617,W,1,10,0.92,00264,M,051,M,,*47
56428 27607.926 127.127.20.0
$GPGGA,074007.0,4055.229,N,00829.618,W,1,10,0.92,00266,M,051,M,,*47
56428 27623.925 127.127.20.0
$GPGGA,074023.0,4055.229,N,00829.618,W,1,10,0.92,00270,M,051,M,,*46
56428 27639.925 127.127.20.0
$GPGGA,074039.0,4055.229,N,00829.618,W,1,10,0.92,00272,M,051,M,,*4F

Nothing out of the ordinary happened physically around the server at
27479.926 for this huge jump in frequency... I don't understand this...

It is running fine and suddenly a huge jump in frequency.

I am running NTP 4.2.6p5 and here's my config file

server 127.127.20.0 prefer mode 18 minpoll 4 maxpoll 4
fudge 127.127.20.0 time2 0.215 flag1 1 flag3 1

server ntp3 iburst burst minpoll 4 maxpoll 4
server ntp4 iburst burst minpoll 4 maxpoll 4

tos mindist 0.250

statistics loopstats peerstats clockstats
statsdir /var/log/ntp/
filegen clockstats file clock.ntp2 type day link enable
filegen peerstats file peers.ntp2 type day link enable
filegen loopstats file loop.ntp2 type day link enable

restrict 127.0.0.1
restrict 10.0.2.3 nomodify notrap nopeer
restrict 10.0.2.4 nomodify notrap nopeer
restrict default nomodify notrap nopeer noquery noserve

How can I debug this problem?

Any help will be appreciated!

Cheers,
Miguel