True, but there is always a probability that they all happen to be off one way.
Quite a small probability, but not impossible.
Sorry for the disturbing thought.
In the case of the 732A, and the early 732B, this is in fact the case! Fluke mentioned, that all the 732A and the early 732B containing the Motorola reference had a systematic positive drift, whereas the later 732B, LTZ1000 based, showed a negative drift. Ref.: fluke.ae/comx/applications/deaver_msc01.pdf Therefore, for an ensemble of references, it is important to have a very good instant calibration (referred to a Josephson effect based standard), plus a determination of the annual drift rate by repeated calibration. Only then, statistical improvement by an ensemble makes sense. Btw.: Many artifact references cannot/define/ the VOLT: Such a definition by Weston ensembles drifted apart 10ppm between the different National Standard Institutes during the years. Only the introduction of the 'Josephson-Volt' in 1970 lead to the reproducibilty of the VOLT within 1e-9 worldwide. But the VOLT still is uncertain by 1e-7 in the SI system.. Perhaps next year this will change by redefinition of the SI-System (h, e, k, kg,..). Frank _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.