> The phase plots on tvb's web site > http://www.leapsecond.com/pages/tadd-1/3phase2.gif show that the device's > temperature was never stabilized. The closest to being stable is at the 29 > minute point on that plot. > > Particularly when the phase value goes up (not sure if it's when temperature > goes up or down), because the phase was still changing at a pretty high rate > when the compressor changed state. I think Tom should rerun that test with a > much slower cycling, to make sure the phase was stable before the compressor
Agreed. Actually, I don't have that A/C anymore so future tempco tests won't have that rapid cycling. > was turned on (or off). Since the phase excursion was not complete, the > 150pS/degree is probably underestimated. > > My guess is that the period should be at least doubled. Yeah, I think so. The main conclusion I had when I made those plots was that the input filter was a bad idea, at least if the TADD was to be used for precision work. > Didier KO4BB /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.
