[Server-sky] IR filter brings excellent benefits

[Keith Lofstrom]

I recently made a rather amazing discovery for Server Sky,
which may cut cost and launch weight in half.  See the work
in progress at      http://server-sky.com/IRfilter 

Thinsat minimum weight is limited by lightsail thrust - there's
too much of it.  To keep the perturbations down, version 4
thinsats must weigh more than 0.2 kg/m². 

However, if we put a high pass optical filter in front of the PV,
3.5μm cutoff, the infrared emissivity on the front side drops
by 99%, and all the IR emits from the back.  The IR photons
scatter in all directions, but mostly straight back, creating
opposing photonic thrust, compensating as much as 2/3 of the
front surface absorption thrust.

For sail-limited PV systems that dissipate almost all of the
absorbed energy as IR, this is a big win.  We should be able
to cut thinsat mass by as much as 60%.  In addition, we can
do a flip and heat the backside during orbital eclipse, 
reducing lower temperature extremes (the hot extreme isn't
nearly as much of a problem). 

That flip also protects the night sky from light pollution.
Trying to figure out how to encourage companies to perform
the "zero night light pollution" maneuver with their arrays,
which reduces power production in part of the orbit, has
troubled me for years.  Now, the flip protects the thinsats
from ultracold damage, and is necessary to maximize the 
power-to-weight ratio.  Win win!

If a thinsat ever goes out of control, the sail thrust
compensation vanishes, the orbit goes highly elliptical
quickly, and the thinsat deorbits rapidly.  It might take
a few years for perigee to drop to 1000km, but after that
it takes a week to drop to ISS altitude (where the orbit
decays at 50 meters per second), and about 30 minutes to
reenter from there.  Ultralight structures don't stay in
orbit long down in the higher gas density regions.

Not that we want to lose the mass of a dead thinsat if
we can recycle it as ballast instead, but it is nice to
know that there's a failsafe if things go pear shaped.

Thanks to William Mook who's been helping with Éntendue
and filter calculations.

Keith
 
-- 
Keith Lofstrom          keithl at keithl.com