> -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of WarrenS > Sent: Tuesday, February 23, 2010 1:04 PM > To: Discussion of precise time and frequency measurement > Subject: [time-nuts] Fw: Rb Oscillator - rather fundamental question > > One more try, As often happens when I rant, the rant is discarded in the > posting. > > > > Rick > > Thanks, Interesting but maybe you have missed my too subtle of a point. > > Example: > Lets say the second is redefined in the future to some new super duper thing > that is good to 1 part in e20 > (Which will happen if (when) the super duper thing becomes more available > and proven) > (Maybe based on the time it takes to count all the atoms in the new purposed > 1 Kg sphere OR something like that.) > > Then the CS Osc would not be the BEST primary standard anymore, at least NOT > at the new improved spec it could then be given. > Not because it has changed or is less accurate, but because there is now > something better. > If it is not the primary standard, it does not make it worse, but it does > mean it will now be a second standard at the new higher performance spec, by > definition and need to be then calibrated and checked against the new > primary standard IF one wanted to use it to it's maximum capability as a > cost effective substitute for the supper duper. >
I don't think so.. In that event, the unobtainium and Cs sources would both be primary standards (within their random variation). The Cs+ standard would be less accurate than the Un+ standard, but they're both primary. The key to a primary standard is that if you build it, it requires no calibration against something else. It's like the meter, which is defined in terms of the wavelength of a particular spectral line in vacuum. Doesn't matter how you get the spectral line, but count off some number of wavelengths, and that's a meter. Now, could one argue that a Rb source is also a primary standard, just with low precision? _______________________________________________ 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.
