[ntp:questions] fudge time1 for gps-18x-LVC?

David Lord snews at lordynet.org
Sun Feb 7 12:24:41 UTC 2010


David J Taylor wrote:
> 
> "David Lord" <snews at lordynet.org> wrote in message 
> news:7t6oaaFul7U1 at mid.individual.net...
> []
>> Now NMEA(0) 92u/6u and PPS(0) 75u/2u, 60 minutes after ntpd restarted.
>>
>>  Config:
>>  pps to lpt0
>>  pps to serial dcd
>>  lpt0 at ppbus
>>  pps0 at ppbus
>>  crw-r--r--  1 root    wheel     164,       0 Feb  7 01:47 pps0
>>  server 127.127.20.0 mode 1 prefer minpoll 6   maxpoll 8
>>  server 127.127.22.0 minpoll 4 maxpoll 6
>>
>>  7 Feb 01:46:18 ntpd exiting on signal 15
>>  7 Feb 01:47:38 clock PPS(0) event 'clk_noreply' (0x01)
>>  7 Feb 01:51:58 synchronized to 192.168.59.60, stratum 2
>>  7 Feb 01:51:58 kernel time sync status change
>>                 0x2001<PLL,NANO,MODE=0x0=PLL,CLK=0x0=A>
>>  7 Feb 01:51:59 synchronized to GPS_NMEA(0), stratum 0
>>  7 Feb 01:53:09 synchronized to PPS(0), stratum 0
>>
>>       remote       refid      st t when poll reach delay offset  jitter
>>  ======================================================================
>>  +GPS_NMEA(0) .GPS.            0 l   19   64  377  0.000 -1.680 253.141
>>  oPPS(0)      .PPS.            0 l   10   16  377  0.000 -1.652   0.150
>>  -k6x400      xxx.xxx.xxx.xxx  2 u   26   64  377  0.740  0.822   0.778
>>  +me6000g     .PPS.            1 u   21   64  377  1.767 -0.439   0.309
>>  -p4x2400c    xxx.xxx.xxx.xxx  2 u   22   64  377  1.893  0.611   0.520
> 
> David,
> 
> Thanks for the details.  My (rather limited) understanding was that the 
> atom driver was simply a filter on an existing PPS signal, and therefore 
> you would need a driver which knew about the LPT port and could detect 
> and use its PPS signal.  Which driver does that I don't know, as I've 
> never used one.
> 
>> Now NMEA(0) 92u/6u and PPS(0) 75u/2u, 60 minutes after ntpd restarted.
> 
> Are these jitter/offset figures?  Do I gather that the 253 milliseconds 
> NMEA jitter was at the start, and it's now 6 microseconds?  Why so bad 
> at the start, or was that simply the first reading?  But it can't be the 
> first reading as it shows Atom sync, which was 7 minutes after the 
> start. The "billboard" NMEA jitter figure puzzles me!
> 

offset/jitter for NMEA now at 10u/2u and for PPS at 7u/2u

Downside of not having fudge is first few NMEA offsets are 350ms
until the NMEA driver sorts out its own handling of pps. As soon as 
offset shifts to a few ms or less jitter then rockets to near 350ms
and works down from there fairly quickly.

It's probably worth trying with the fudge time1 in place as that
might just improve startup.
....
  xGPS_NMEA(0) Offset 650ms
I'm sure that indicates it doesn't work.
....
  *GPS_NMEA(0) reach 377 Offset 650.167 jitter 0.305
and servers right now all just been marked rejected
  GPS_NMEA(0)  reach 0
  serv2 .PPSa. reach 0 Offset -649.27

Now back to config without fudge

Cleanest but not fastest startup is with NMEA fudged correct
amount without serial DCD, and using only ATOM driver for
accuracy. That still gives large startup jitter from NMEA
sentences coming in.

These are VIA C3 mini epia me6000 and me6000g motherboards
and I get impression they  don't have good quality clock
oscillators.

David




More information about the questions mailing list