The two things that come to mind are is the pulse wide enough and is it the correct polarity. Polarity wrong would only add delay but the pulse width would cause the PC to not see it.
Check these links: http://www.tapr.org/kits_fatpps.html http://www.wraith.sf.ca.us/ntp/#gps-receiver ----- Original Message ---- From: Bruce Taylor <[EMAIL PROTECTED]> To: [email protected] Sent: Monday, October 20, 2008 3:09:07 PM Subject: [time-nuts] Getting NTP/Linux to work with GPS-DO? Folks, Has anyone gotten a Jackson Labs 'Fury' gps-do to work as a time/pps source for Linux? Our lab is looking to use both the stable 10MHz clock for various LOs on a radio-telescope, and also get reasonably tight timing for a software correllator to use in discriminating pulsar signals. Our Linux guy has lost most of his hair out trying to get the NTP daemon to receive the PPS signal and thus get NTP to see the GPS-DO as a stratum 0 source. Our configuration is: vanilla Linux from kernel.org (version 2.6.23) running on an intel platform. We've applied the LinuxPPS kernel patch, and we're feeding the serial port with the Fury in NMEA mode. We made the required internal mod on the Fury to feed the PPS signal out on the serial port, and can see it blinking using a scope. Apologies if my description is vague, as I'm just getting into the fun and joy of kernel hacking :-) Many thanks for any advice, - Bruce Taylor _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
