[time-nuts] time-nuts Digest, Vol 101, Issue 152
Jim Lux
jimlux at earthlink.net
Mon Dec 31 16:02:04 UTC 2012
On 12/31/12 12:55 AM, Hal Murray wrote:
>
> jimlux at earthlink.net said:
>> When designing the system to do coherent two way ranging to the Juno
>> spacecraft orbiting Jupiter, we found that the rotating magnetic field
>> (because the spacecraft spins at 2RPM in Jupiter's magnetic field) was
>> enough to modulate the circuits that track the received signal and generate
>> the transmitted signal ( basically a PLL)...
>
> How strong is the magnetic field near Jupiter? I assume big, but maybe the
> orbiters don't get that close.
Big, and pretty close.. Juno has a highly elliptical polar orbit, and
they want to get close because that gives you the best data for all the
sensors. For instance, for radio science, where we're measuring the
speed of the spacecraft to infer gravitational field, the closer you
are, the better the resolution (because you're affected more by what's
closer to you than what's farther away). (GRACE flew a few tens of km
above the Moon's surface for this reason)
I'll see what I can find. it was bigger than Earth's field, I seem to
recall.
>
> How did you test it? I'd guess a big chamber with big Helmholtz coils.
Exactly. We do that anyway for spacecraft equipment: there's
typically magnetometers on the spacecraft to measure the magnetic
fields, so they impose limits on how much field you can make (and what
your static field is). It was unusual for us telecom folks to worry
about magnetic susceptibility. Usually, they're beating up on us about
the permanent magnets in ferrite isolators or the field from the TWTA.
One unusual thing is that most of the test facilities apparently aren't
necessarily set up for a 1/30th Hz (2 RPM) rotating field (which
requires changing the currents in multiple axes at the same time, etc.).
>
> ---------
>
> I just took a fridge magnet to my box of (somewhat?) good oscillators. The 2
> big ones had steel cases. The smaller Vectron had a steel base but
> non-magnetic top.
>
> ----------
>
> jimlux at earthlink.net said:
>> A bigger concern might be if you're locking something you care about to
>> that oscillator. The PLL might be more sensitive (particularly things like
>> the op amp circuitry driving the VCO... there's probably some sort of
>> physical loop in the wiring between the inverting and noninverting inputs
>> of the op amp.
>
> Does that mean we should put that stuff into a steel box? (With junk like
> power supplies outside.)
>
I think that's exactly what the obsessives up the Frequency and Timing
Lab do. Soft iron or mu metal, not steel, also..
For stuff I have to do: I'll watch the layout of the PWB to minimize the
loop area. It's good practice anyway to reduce coupling in or out.
More information about the time-nuts
mailing list