High jitter with GPS Clock - comp.protocols.time.ntp

Hi all, I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux 2.4.20-8, ntpd 4.1.1c). However, what I don't understand is the jitter is very high. [root@linux]# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================ == *GPS_PALISADE(0) .GPS. 1 l 30 32 376 0.000 7.369 9.381 +ntp1.ptb.de .PTB. 1 u 44 64 377 18.836 29.007 1.644 When I run ntptrace at the command line I get: ---------------------------------------------------------------- [root@linux]# ntptrace -v server 127.0.0.1, port 123 stratum 2, precision -16, leap 00 refid 127.127.29.0 delay 0.00111, dispersion 0.00000 offset 0.000442 rootdelay 0.00000, rootdispersion 0.00465, synch dist 0.00465 reference time: c35b1f26.0bb352a8 Tue, Nov 11 2003 9:41:10.045 originate timestamp: c35b1f33.fcc36da8 Tue, Nov 11 2003 9:41:23.987 transmit timestamp: c35b1f33.fc7b352a Tue, Nov 11 2003 9:41:23.986 ---------------------------------------------------------------------------- ------ What could be wrong? I have a simple config as follows: ---------------------------------------------------------------------- server 127.127.29.0 prefer # Trimble acutime 2000 fudge 127.127.29.0 stratum 1 fudge 127.127.29.0 time1 0.020 fudge 127.127.29.0 flag2 1 server 192.53.103.103 driftfile /etc/ntp/drift broadcastdelay 0.008 logfile /var/log/ntp # alternate log file ---------------------------------------------------------------------------- --------- And I think the jitter should be very low since I have a local GPS clock. But as you can see above that is not the case. I would be very appreciative for any hints.

"Rickan Lee" <rickan2000@hotmail.com> writes: > Hi all, > > I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux > 2.4.20-8, ntpd 4.1.1c). > > However, what I don't understand is the jitter is very high. PPS or not PPS? AFAIK, the full performance will only be available when you use PPS. Ulrich

No, not PPS. Is PPS also supported by Linux? can PPS signal pass thru an RS422/RS232 converter? "Ulrich Windl" <Ulrich.Windl@RZ.Uni-Regensburg.DE> schrieb im Newsbeitrag news:m3smkvdnr9.fsf@pc5234.klinik.uni-regensburg.de... > "Rickan Lee" <rickan2000@hotmail.com> writes: > > > Hi all, > > > > I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux > > 2.4.20-8, ntpd 4.1.1c). > > > > However, what I don't understand is the jitter is very high. > > PPS or not PPS? AFAIK, the full performance will only be available > when you use PPS. > > Ulrich

On Tue, 11 Nov 2003 16:46:19 +0100, Rickan <rickan2000@hotmail.com> wrote: >No, not PPS. Is PPS also supported by Linux? can PPS signal pass thru an >RS422/RS232 converter? > Speaking of which, the serial driver has a small but varibale delay in fetching characters from the uart. Setting low_latency using setserial should minimise that delay. Villy

In article <3fb1044a$1@news.fhg.de>, Rickan <rickan2000@hotmail.com> wrote: >Is PPS also supported by Linux? Yes. See your favorite kernel mirror, or use Google. -- My real email address is mcintosh ##at## research ##dot## telcordia ##dot## com

Similiar Articles:

High jitter with GPS Clock - comp.protocols.time.ntp
Hi all, I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux 2.4.20-8, ntpd 4.1.1c). However, what I don't understand is the jitter is very high.

NTP High Jitter and Reject Condition - comp.protocols.time.ntp ...
High jitter with GPS Clock - comp.protocols.time.ntp NTP High Jitter and Reject Condition - comp.protocols.time.ntp ... High jitter with GPS Clock - comp.protocols.time ...

GPS clock for Linux - comp.protocols.time.ntp
High jitter with GPS Clock - comp.protocols.time.ntp Hi all, I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux 2.4.20-8, ntpd 4.1.1c).

Strange jitter values - comp.protocols.time.ntp
High jitter with GPS Clock - comp.protocols.time.ntp Strange jitter values - comp ... Strange jitter values - comp.protocols.time.ntp NTP High Jitter and Reject Condition ...

Onboard Local Oscillator Change Improvements - comp.protocols.time ...
... kits using their new CY22180 very low jitter clock ... side is: - which accuracy of local clock can we expect using high ... GPS time does wander in the double-digit ns ...

Linux 2.6 and clock frequency - comp.protocols.time.ntp
Linux NTP Polling - comp.protocols.time.ntp If that frequency is poor, the absolute rate may be close to zero in one ... High jitter with GPS Clock - comp.protocols.time ...

PARSE Clocks and Jitter - comp.protocols.time.ntp
PARSE Clocks and Jitter - comp.protocols.time.ntp Which ntpq command gives me the best idea of my clock ... ... High jitter with GPS Clock - comp.protocols.time.ntp ...

Single best NTP status indicator of clock accuracy? - comp ...
High jitter with GPS Clock - comp.protocols.time.ntp Single best NTP status indicator of clock accuracy? - comp ... PARSE Clocks and Jitter - comp.protocols.time.ntp Which ...

Linux 2.6 + PPS + ntpd? - comp.protocols.time.ntp
High jitter with GPS Clock - comp.protocols.time.ntp (Linux 2.4.20-8, ntpd 4.1.1c). However, what I don't understand is the ... ... Is PPS also supported by Linux? can PPS ...

bc637PCI-U Reference Clock Driver - comp.protocols.time.ntp ...
High jitter with GPS Clock - comp.protocols.time.ntp Hi all, I have connected a GPS-Clock (Acutime 2000 ... converter? > Speaking of which, the serial driver has ...

High jitter with GPS Clock - comp.protocols.time.ntp | Computer Group
Hi all, I have connected a GPS-Clock (Acutime 2000) via phone wire to my PC. (Linux 2.4.20-8, ntpd 4.1.1c). However, what I don't understand is the jitter is very high.

Jitter & Wander Tutorial
When measuring jitter or wander, always be sure what the reference clock is. ... are of importance, high ... Global Positioning Satellite, a source of reference ...

7/22/2012 5:13:33 PM