[Beowulf] Server Sky, thermal radiation

Eugen Leitl eugen at leitl.org
Mon Jun 24 06:57:38 PDT 2013


----- Forwarded message from Keith Lofstrom <keithl at gate.kl-ic.com> -----

Date: Sun, 23 Jun 2013 10:21:45 -0700
From: Keith Lofstrom <keithl at gate.kl-ic.com>
To: Eugen Leitl <eugen at leitl.org>
Cc: server-sky at server-sky.com, astro at postbiota.org, tt at postbiota.org
Subject: Server Sky, thermal radiation
Reply-To: keithl at keithl.com
User-Agent: Mutt/1.4.2.2i

Some notes on thermal radiation:

With the corner electrochromic thrusters, thinsats have plenty of torque
to stay exactly sun-oriented, so they get 1360W/m².  I am assuming 900nm
graded-junction semi-amorphous InP cells with 15% efficiency on sunside. 
Higher efficiency is possible, but thin InP cells can be super radiation
resistant.  Some sunlight "wasted", but there's more where that came from!

Server Sky orbits at one radii out, 12789km radius, 6411km altitude
above the equator.  At that altitude, the 250K earth is 60 degrees
across and 6.7% of the sky.  That's 6000 times the area of the moon,
but 1/7th the solid angle of the earth seen from LEO.  

The hottest thing in the sky is the cloud of other thinsats.  They
must be spaced far enough apart for light to penetrate the array,
and for heat to leave.  That means longer delays for intra-array
communication, and precludes some kinds of highly-coupled parallel
computations. 

At noon, most of the sky is still deep space, and a good heat sink.
At midnight thinsats get pretty damned cold, no thermal inertia to
speak of.  Lots of thermal cycling stress, but we know how to design
planar systems for that.  Lots of identical cells in a grid array,
so we can route around damage.

Thinsats will get hot, but not hotter than the CPU in your computer. 
Because of the lousy lateral thermal conductivity, we can anneal some
of the radiation damage segment by segment.  We can connect all the PV
segments together, heat one of the segments enough to induce thermal
runaway, and cook it at 250C.  That will be especially useful to
mobilize trapped charges in the insulators of the backside CMOS chips.

The key to server sky is thin and stackable.  They weigh 0.2kg/m², and
can weigh a lot less if we can add extra ballast mass in orbit (that's
what you do with obsolete thinsats).  They stack into cylinders for
launch, with the mechanical properties of a solid slab of aluminum,
very robust.  A heck of a lot of design is needed, but the cost of
design is divided by millions and someday trillions of thinsats.  

Lots more disorganized stuff at the website, http://server-sky.com .
Use text search to find keywords, look at RecentChanges to see what
I've been working on.  Lately, preparing presentations, and trying
to figure out OpenCL on an nVidia GPU to compute radio emission
patterns of whole arrays (and later, array thermal properties). 
After that, studying thinsats as Rutherford scatterers.

I love looking at problems from new and unexplored angles.  I'm
discovering lots of stuff, some of it world-changing (good and
bad), and some of it $$$+ .  I'll never set foot on another planet,
but I do get to boldly go where no man has gone before.  Fun!

Keith

P.S.  I present Server Sky to DC-L5 in a week, and will visit family in
MD the week after.  If anyone in the area wants to meet F2F, contact me.

-- 
Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://ativel.com http://postbiota.org
AC894EC5: 38A5 5F46 A4FF 59B8 336B  47EE F46E 3489 AC89 4EC5


More information about the Beowulf mailing list