[time-nuts] time-nuts Digest, Vol 101, Issue 152
jimlux at earthlink.net
Mon Dec 31 15:41:08 UTC 2012
On 12/30/12 8:28 PM, Bob Camp wrote:
> Indeed, there will always be some EMF into the EFC from some field. You can never really get rid of a loop with some cross section in the EFC circuit. Most of us don't get to worry about 1x10^-16 at 1,000 seconds on our OCXO's….
Yes. I think it's not so much the OCXO as the performance of the overall
system it's in where magnetic fields would be the issue.
I'll have to go look up what the USOs we use have for magnetic field
performance.. For Cassini, Sami Asmar's 1997 paper says 5E-13/Gauss,
and the rest of the missions were in the 1E-12/Gauss range..
that USO was built by APL.. And APL's Greg Weaver and colleagues had a
paper at PTTI in 2004
that describes some of the things (although not magnetic fields).
For superduper accuracy applications like deep space radio science, I
think the issue today might be more about things like hysteresis and one
way shifts (aging, radiation, etc.). In theory, one could calibrate out
the effect of the rotating magnetic field. We already have to calibrate
out the effect of the apparent phase center of the spacecraft antenna
moving as the spacecraft rotates because the phase center isn't
perfectly on the axis of rotation, nor is the spacecraft perfectly
pointed towards earth (at least not at 1E-12 kinds of precision). The
magnetic field effect would presumably be another sinusoidal variation
with the others.
Jensen and Weaver had a paper at PTTI in 2007 talking about the USOs on
New Horizons, accounting for all the observed changes in frequency.
Since New Horizons went by Jupiter on its way to Pluto, it was affected
by the radiation there. I didn't see any mention of magnetic fields, and
since NH is a spinner, it would get the same rotating magnetic field
thing we worried about for Juno.
GRACE (which just finished its mission by crashing into the Moon's
surface) carried a USO on each of the spacecraft. I wonder if they have
any data on magnetic fields. I remember it coming up in the design
reviews, but don't recall if it was considered a big issue, or just one
of those "this is what we always do, and it seems to work well enough".
APL, of course, is striving for ever better quartz clocks for this kind
of application. At JPL, we're looking at flying a trapped Hg ion clock
(aka DSAC- Deep Space Atomic Clock) as a demo in the next few years to
give us several orders of magnitude improvement over quartz.
Both are important when you do "one way ranging".. For the more common
two way ranging (where we send a signal up to the spacecraft and it
sends it back) the quartz oscillator isn't as important as the
"turnaround" performance, where the VCXO control is what's important.
More information about the time-nuts