[sfconsim-l] Collisions in Space

1 posts ยท Dec 22 2002

From: Alan and Carmel Brain <aebrain@w...>

Date: Sun, 22 Dec 2002 12:28:18 +1100

Subject: Re: [sfconsim-l] Collisions in Space

While looking through the data on satellite failures, I came across some
interesting causes.

Sometimes's it's a Fender-Bender, a physical Collision:

http://sat-nd.com/failures/cerise.html
"The debris (actually a part of an Ariane rocket) appears to have impacted the
stabilisation boom, which extends 6 metres from the main body of the
spacecraft, at 50,000 km per hour, or 14 km per second."
See also http://sat-nd.com/failures/kosmos539.html
and http://sat-nd.com/failures/telstar6.html

Sometimes it's Dat Debil Radiation:

http://sat-nd.com/failures/aqua.html
where the spacecraft behaved as designed when severely zapped over the South
Atlantic Anomaly, and worked perfectly afterwards.
Then there's http://sat-nd.com/failures/nozomi.html
"The spacecraft was hit by a coronal mass ejection on 21 April 2002. The burst
of solar energy caused one of the electrical power converters on board to
"latch up," knocking out the main power and freezing the onboard propellant."
Despite this "Scientists at Japan's Institute of Space and Astronautical
Science (ISAS)
have re-established contact with the Nozomi Mars probe and hope that the
mission will be back on track soon."

Sometimes screw-ups on the ground cause problems:

http://sat-nd.com/failures/soho.html
"The first two errors were contained in preprogrammed command sequences
executed on ground system computers, while the last error was a decision to
send a command to the spacecraft in response to unexpected telemetry
readings."
Even then, a swift software change by some really top-notch people can
sometimes work-around the problem:
"To stop the rapid depletion of hydrazine fuel, engineers at ESA and Matra
Marconi Space have designed a software programme to enable the spacecraft to
resume science operations without gyroscopes as from 2 February 1999."

Sometimes its the dreaded "Killer Electrons":

http://sat-nd.com/failures/telstar401.html
"One explanation for the death of Telstar 401 seems to be that a magnetic
cloud caused a massive short in the satellite's circuitry. This is in
compliance with NASA observations made shortly after the failure, which
indicated the bird was still in place but slowly spinning.
Other theories claim that the satellite was zapped by so-called killer
electrons."

And sometimes they just suffer a "Major anomaly", ie Go **BOOM**

http://sat-nd.com/failures/telstar402.html

This whole website has increased my respect for other
satellite-designers
enormously. And made me rather glad of certain very paranoid design decisions
I made early on (like making sure the triplicated data was in 3 widely
seperated physical parts of the satellite, not having just one method of doing
something when 4 different methods could be provided as cheaply, etc)

One consequence for designers of space-combat games: you can be sure
that even the
most shoddily-designed el-cheapo commercial ships will have redundancy