[ntp:questions] timing issue with a HP 58534A

Mark C. Stephens marks at non-stop.com.au
Fri Feb 3 14:45:20 UTC 2012


Hello People,


I seem to have a timing issue with a HP 58534A smart antenna; the NMEA refclock is always about half a second out and never gets selected:

[root at NTP /etc/ntp]# ntpq -p
     remote          		refid     		 st 	t 	when 	poll 	reach   	delay   	offset  	jitter
==============================================================================
 GPS_NMEA(0)     	.GPS.            	0 	l    	6   	16  	377    	0.000 	525.229  0.027
*time.non-stop.c 	203.35.x.x    	2 	u  	28   	64  	377    	0.285	1.158   	0.312

I have tried this on a win NTP server and a freebsd job. Same result.

I wonder if I could bounce this off you, any insight would be appreciated.

According to the manual:
The first character of the first serial data packet is transmitted 400 ms
(± 2 ms) following the rising edge of a 1PPS pulse.
A date/time stamp for a 1PPS pulse is included in the serial data
packets which are transmitted by the 58534A GPS Timing Antenna
directly before the rising edge of each 1PPS pulse.
In a series of data packets, the date/time stamp is normally the last data
packet in the series.

Please see http://www.non-stop.com.au/sites/users/marks/097-58534-01-iss-1.jpg for diagram.

The timing looks wrong to me, and I would have thought the sentence should come on the pulse, not before.

*Is this normal behaviour?

Now, I have just applied fudge 127.127.20.0  time2 -0.5 via ntpdc

After a few seconds:
[root at NTP /etc/ntp]# ntpq -p
     remote           		refid      	st 	t 	when 	poll 	reach   	delay   	offset 	 jitter
==============================================================================
 GPS_NMEA(0)     	.GPS.            	0 	l   	 4   	16  	377    	0.000   	25.272 	377.930
*time.non-stop.c 	223.252.x.x     	2 	u   	26  	128  	377    	0.299    	0.731   	0.076


I will monitor for a few hours and post result.


Many thanks,
Mark





More information about the questions mailing list