What Pax said. Most COTS parts will last 10 years in LEO. One year is
easy, with caveats. Avoid power silicon NFETs if you can. One lucky
cosmic ray can permanently rupture the gate. If you can't avoid them,
derate the voltage by at least 4:1. PFETs, GaNFETs and bipolar are more
rugged. Use automotive rated parts whenever available.
If possible, shut down during passage through the South Atlantic Magnetic
Anomaly. Count on and plan for single event upsets at any time. A
hardware power cycling watchdog is a must.
Derate everything. Derate capacitor voltage at least 2:1, 3:1 is better.
Avoid high localized power densities. Use multiple ground planes to
conduct away heat. Derate resistor power dissipation 4:1. Derate
microprocessor clock rates 2:1. This not only reduces local heating but
gives the internal timing constraints more margin at the temperature
extremes.
NO electrolytic capacitors. Avoid tantalum if you can get enough
capacitance from ceramic.
Use leaded solder. No-lead solder will grow tin-whiskers eventually.
Avoid ball grid array packages. You'll have thousands of thermal cycles
and the solder balls will crack first, sometimes in less than a year.
Besides, the balls are almost always no-lead nowadays.
Conformal coat with acrylic or urethane. This helps with off-gassing,
on-gassing, FOD, tin whiskers, and arcing.
Check all materials on the Nasa out gassing database, not just plastics.
NO cadmium! Lots of connectors are cadmium plated. It will do a wonderful
job of off gassing and then vacuum plating your electronics. Passivate
aluminum with alodine.
Li-ion batteries need heaters. Have enough capacity so you don't discharge
below 40%.
Read the Nasa Cubesat 101 document. Mostly about bureaucratic compliance
which will drive you batty if you don't get in front of it early.
-Allen
On Thu, Dec 5, 2019 at 11:48 AM Pax L <rocketpax@xxxxxxxxx> wrote:
So most LEO cubesats are not radiation hardened...at all. That said,
most LEO missions are pretty short duration comparatively(less than a
year). There is some component selection where some types of equipment
are better than others , but typically not much is done for extra
shielding. If you look at some COTS cubesat parts like radios they
will be in their own aluminum shells or whatever, but that is about
it.
The electronics and software are all designed so that if there are
some bit flips or something the system will or can be easily rebooted
or deal with dead memory blocks.
I think the duration for most of the lunar and mars missions have been
pretty short as well. NASA may do a but more there in electronics
design and fault handling, but still I don't think you see much
physical shielding.
Also, be really careful about plastics selection regarding outgassing
properties. You don't want to poop on other spacecraft
electronics(especially the primes on a launch) or any chambers you
test in.
On Thu, Dec 5, 2019 at 9:03 AM Ian M. Garcia <ian@xxxxxxxxxxxxx> wrote:
electronics in LEO? I’m starting to look at this and comparing rad hard vs
Does anybody here have experience with Cubesat (or any satellite)
COTS with and without shielding. Rad hard options are crazy expensive and
low performance compare to COTS, but I am worried about radiation for a
satellite that could be out there for a while.
a lead block and flight that, but now I am reading that lead is actually
Our joke used to be that in the worst case we’d just drop the boards in
not used that much because it’s too soft, so instead they use tantalum,
tungsten, or just thicker aluminum. But then I continue reading and it
turns out that materials high in hydrogen are better, so they are starting
to recommend polyethylene (HDPE) instead of metals? And then the latest is
that HDPE combined with a metal liner is actually even better because the
metal captures secondary radiation. What the hell? This is all very
confusing.
Cheers,
Ian
--
Ian M Garcia