[time-nuts] Using GPS for space-based instrument

michael taylor mctylr at gmail.com
Mon Nov 10 19:01:00 EST 2008

On Mon, Nov 10, 2008 at 5:27 PM, Strauss, Karl F
<karl.f.strauss at jpl.nasa.gov> wrote:
> I've been tasked (or was it I was volunteered?) to do some basic design & definition work on an ultra-stable master frequency system for a proposed instrument that is currently planned to be in an Earth-trailing orbit.   Given the first order accuracy requirement of 1 part in 1E-10, my first thought was to grab the GPS timing signal.

Depending on your application requirements, I wonder if an OCXO (Oven
Controlled Crystal Oscillators) in a ruggedised packaging might be
suitable for your needs. For example one well-known frequency source
vendor, Symmetricom offers a number of OCXO packagings that may be
suitable for satellites.


Another option would be a Rubidium (Rb) frequency standard.

A lot depends on your "accuracy" definition and requirements. Does it
need to be accurate relative to UTC / UT / TAI? Depending on your
application, dealing with details like leap seconds may not be worth
staying in sync with UTC / GPS. Or do you need a stable local clock
(within the SV)? Are you concerned with accuracy / stability over a
short period of time (e.g. 1 second) or a longer period of time (e.g.
1 day / month / etc.)?

More information about the time-nuts mailing list