/tvb Nice work, I was afraid you had dropped the subject.
There is a chance that the Allen noise ripple using the 100Hz can be reduced or eliminated if you have a processor in the system. The short of it is that every once in a while the 100 Hz syncs up for a short time to the 1 Hz which itself has synced up to something else and is at the wrong value. The sawtooth correction knows that the 1 Hz is at the wrong value and reports it as such so this does not effect the anything if using the Sawtooth correction. When this does happen the 100 Hz is at the wrong value also and becomes no better than using the one Hz signal during this short period, and so during this time it reverts back to the basic performance of the 1 Hz. You can see this sync effect happening on your 100 Hz phase data at the 9 seconds This condition is very easy to detect and correct for, and if the processor did compensate for it, I have found that it helps a lot with fast loops, probable no big deal with the slower loops. One other thing, If someone where using a really crappy OSC with noise so bad that it that needed an update rate at was greater than 1 Hz, then the 100 Hz will give far better performance than either the 1 sec or the corrected sawtooth. Of course no one in there right mind would use that poor of an osc, that is unless they had some weird and different reason for doing it. IT would be interesting to see how much improvement could be obtained in some GPSDO, if the phase detectors errors where eliminated. Have Fun, You sure do nice work and make nice plots. WarrenS ******************* ----- Original Message ----- From: "Tom Van Baak" <[EMAIL PROTECTED]> To: "Discussion of precise time and frequency measurement" <[email protected]> Sent: Friday, December 05, 2008 12:32 AM Subject: [time-nuts] M12 100 Hz vs. 1PPS measurements >I know a number of you have been waiting for the results of > this M12 experiment. I compared the performance of a 100 Hz > output vs. a 1PPS output vs. a 1PPS with sawtooth correction. > > http://www.leapsecond.com/pages/m12-adev/ > > I think it came out quite well and it may surprise some of you. > > /tvb > > > > > _______________________________________________ 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.
