[time-nuts] New frequency standard, Mercury better than Cesium?

Magnus Danielson cfmd at bredband.net
Mon Jul 17 08:08:53 UTC 2006


From: Hal Murray <hmurray at suespammers.org>
Subject: Re: [time-nuts] New frequency standard, Mercury better than Cesium?
Date: Sun, 16 Jul 2006 10:28:19 -0700
Message-ID: <20060716172820.435B3BDF0 at ip-64-139-1-69.sjc.megapath.net>

> >From the horses mouth:
> 
>   http://www.nist.gov/public_affairs/releases/mercury_atomic_clock.htm
> 
> 
> This brings up a question I've been meaning to ask for a while.
> 
> How do you tell how good your best clock is?  I can figure out how good a 
> not-great clock is by comparing it to a better one.  But what if there isn't 
> a better one?

There are basically two methods that have been in use:

1) Build two clocks and compare them against each other. This is what Ramsey
   et al did for the hydrogen masers. Their phase-noise sould be about equalent
   so you can put down both clocks for the 1/sqrt(2) of the measured phase
   noise (they contribute the same amount of noise energy with the same
   distribution in this assumption).

2) Compare three clocks, all of low phase noise. Make three pairs of
   measurements for clocks 1-2, 1-3 and 2-3. The noise contribution of each
   clock into each measuremnts allow for cancelation and the phase-noise of
   all three clocks may be found.

As an alternative to method 2 you may have a clock with know phase noise, but
then measure your new clock against it and they subtract out the phase noise of
your known clock. Whenever you compare the phase-noise of a better clock with
known clocks, you rarely want to have a phase-noise more than a decade worse
then the clock you are going to measure, since you will run into the precission
of the decimals and that takes averaging time. 

They can often quite accurately predict the phase noise they get. They have a
fair idea of the various sources of errors and it is this understanding which
have led them towards this type of sources. Infact, at one time Thallium was
competing with Cesium to become the standard and it was judged to be more than
2 times more precise, but it was ten judged that microwave design at 24 GHz was
more delicate (and thus harder to stablize and repeat) then down at 9.2 GHz so
it was the wavebreaker then.

I've seen this work going on for a few years now.

Hmm. With ultraviolet lasers you should be able to get the ultimate suntan in
no time. ZAP! :P

Cheers,
Magnus - yes, I am on vacation, but keeps track of fellow time-nuts




More information about the Time-nuts_lists.febo.com mailing list