Re: How do you measure the inaccuracy of the most accurate clock?

Hi Gil,

Caesium fountain clocks (atomic clocks) are so accurate that we now define our time by them; one second is defined as 9,192,631,770 oscillations of radiation from an atomic transition of Caesium 133. The physical principle used here is that a particular atomic transition within an atom of a particular element always gives off radiation of exactly the same frequency. What limits the accuracy of our clocks, then, is the accuracy to which we can measure that frequency.

In a caesium fountain, the frequency of a microwave beam is tuned until it matches, as closely as possible, the frequency at which the caesium atoms absorb and emit radiation. It seems to me, then, that the accuracy of the clock is determined by the accuracy with which the two frequencies can be matched. It's not that there's a more accurate clock somewhere, which we use as a standard, but rather that there's a limit to the accuracy of our frequency measurements. Inside an atomic clock, these could be introduced by collisions between the caesium atoms, or motions of the atoms within the microwave cavity of the clock. Both of these processes introduce very slight variations into the measured frequency of the radiation emitted and absorbed by the atoms. In addition, quantum mechanics introduces a fundamental limit to the accuracy with which we can measure the frequency of an atomic transition. This latter effect is something know as the uncertainty principle. You can find a somewhat longer discussion of atomic clock accuracy at The Why? Files.

For more information, try the following links:

• Clockworks: atomic clocks - an introduction to timekeeping from Britannica Online.
• NIST - the time standard for the USA.
• National Physical Laboratory: Time & Frequency - NIST's equivalent in the UK.
• www.time.gov - accurate time from NIST and the US Naval Observatory.

Jim O'Donnell, jim@penfold0.demon.co.uk