Order Tray | Contact Us | Home | SIG Lists

[time-freq] Re: Frequency Standards - Which to believe?

Alberto I2PHD i2phd at weaksignals.com
Sun Jul 2 13:25:53 UTC 2006


    very good, you have now your two standards aligned to each other, with a frequency difference of 1 / (3600 * 10^7)
which roughly corresponds to less than 3 parts over 10^11. That is the accuracy that more or less a Z3801A is capable
of, so there is no much point to go further.

73  Alberto  i2PHD

Arthur Shulman wrote:
> Directed to Eric & Alberto who were kind enough to answer my question, this is
> the result:
> Following the advice of both gentlemen, which was essentially the same, I
> proceeded to access the C-Field adjustment on the Rubidium standard, and with
> some care and experimentation, was able to adjust the period of disagreement
> (one complete Lissajou cycle) to about five minutes. This adjustment was very
> broad and repeatable, but I thought I might be able to do better.
> I next took the outputs of both standards and read the phase difference at 10
> mhz with an HP8405A Vector Voltmeter. It was necessary to use the appropriate HP
> attenuators to avoid damaging the voltmeter sampling probes. This indeed was a
> sensitive and stable indicator, and it quickly became apparent that much
> improvement was possible!
> I monitored the phase difference for about a day, and finally stopped adjusting
> when it seemed to me that I could no longer tell which standard was getting out
> of phase, the Rubidium, or the GPS one! At this point, I estimated that that the
> period for one complete 360 phase shift was in excess of 45 minutes, and
> probably actually exceeded an hour!
> Taking refuge in deference to the Law of Diminishing Returns, I buttoned up both
> standards and put them into service!
> Thank again for your assistance!

More information about the time-freq mailing list