[time-nuts] List of time synchronization hardware and software

Magnus Danielson cfmd at bredband.net
Tue Jan 17 19:17:46 EST 2006


From: Hal Murray <hmurray at suespammers.org>
Subject: Re: [time-nuts] List of time synchronization hardware and software
Date: Tue, 17 Jan 2006 10:50:19 -0800
Message-ID: <20060117185020.EFFE0BDE2 at ip-64-139-1-69.sjc.megapath.net>

Hal,

I'm gonna babble some here, so the real collectors can correct me and point out
all the nitty gritty details that I don't know.

> > I think a new Caesium-beam cascade is *MUCH* cheaper.
> 
> Can somebody give me a lesson on this area?  Is there a good web page that 
> answers questions like these?

Go to leapsecond to get a clue.

> What does a Caesium/Rubidium box cost?  Any reason to prefer one or the other?

Caesium is far more stable than the Rubidium, but the Rubidium is much cheaper.
I don't have the prices in the back of my head, but I think I recall something
like 50 kUSD and 2-3 kUSD. A hydrogen is somewhere in the 100-150 kUSD up to
400-500 kUSD. Then you can get used gear cheaper naturally.

> What sort of lifetime do they have and/or how much care/fiddling do they take?

Depends. Caesium is some 6-8 years normally, but the high performance tubes are
less (high performance tubes burns more ceasium per time unit). Rubidium have
life-span of 10-20 years. The failure mechanisms are quite different.

> Do they need any calibration or tuning or setup magic, or do I just plug them
> in and get a known frequency out?

If they are in a good condition they phase-lock, but the time to phase-lock may
be fairly long. Unless they are completely out of tune you should not have to
do anything to trim them up.

> Does it make sense for a non-wizard to get something from ebay and if so, 
> what do I look for?  Any obvious models/manufacturers to look for and/or 
> avoid?

If you look at ebay right now there are three Caesium clocks available. Only
one of them phase-locks, and that takes 40 min. If they don't phase-lock, then
stay away unless you know how to fix that model.

OK, that was my very schetchy oversight view. I recommend you to learn some
about the physics in these things, that helps to understand what makes them
tick and what might be wrong.

Cheers,
Magnus - still Cs-less




More information about the time-nuts mailing list