Slashdot Mirror
Cubesat Launch Ends in Failure
Change writes "The CalPoly Cubesat group's launch yesterday has been a failure. It seems the first stage did not separate from the Dnepr rocket properly, and the vehicle crashed about 25km south of the launch site. More will be known when the debris is recovered and analyzed. A second launch is still in the works, but the loss of the 14 satellites from this launch is an unfortunate end to quite a lot of hard work of many engineering students."
Thou Shalt Make Backups
Failing Rule Number 1...
there goes my chance to see if in space they really can hear you scream
A feeling of having made the same mistake before: Deja Foobar
Doesn't matter how many you fire. Theres no guarantee the next one will work.
I saw a graphic of launch insurance costs for commercial satellites a few months ago, the costs are really spiralling out of control. But until we have a space elevator or anti-gravity, riding an explosion of chemicals to orbit is the best system we got.
"Open the pod by doors, Hal" > "I'm afraid I can't do that, Dave" sudo "Open the pod bay doors, Hal" > alright
Such a subsystem would be a custom design for each payload (engineering design man-hours). It would also have to be able to anticipate or react to any of the hundreds of failure modes of a launch vehicle -- solid booster failure, liquid engine failure in any stage, stage separation failure, guidance system anomalies, guidance computer crashes, gyro alignment errors, and more, requiring exhaustive telemetry of the launch vehicle. It would add a significant weight penalty to every launch (forcing many payloads to move up at least one booster size and eliminating many smaller boosters from carrying any payload at all). It would have to be tested for survivability from the thermal and vibrational launch environments (more engineering man-hours). And last but not least, payloads would have to be redesigned to survive the ride back to earth in that subsystem (mass, power, and structure penalties and engineering man-hours).
You would more than double the cost of your payload hedging against the risk of a 1-in-30 booster failure. For satellite programs so important that a booster failure can't be tolerated, they just build spare payloads.
This has been looked at a few times...but the only launchers that currently have payload recovery capability are manned ones (for obvious reasons).
In general, while losing a payload sucks, it doesn't justify the weight and monetary cost of a payload recovery system and the infrastructure required to go get said payload wherever it lands. The satellite itself would end up needing to be a much more robust design to survive the dynamic environment of an ascent abort.
It's all a numbers game - with the worldwide launch success rate around 97%-98%, it's simpler/cheaper simply to buy the insurance or eat the loss.
Many of these less proven launch systems (such as Dnepr, Falcon) have given very inexpensive rides to orbit to help establish a track record while they work out development issues. The track record is important, because an established launch record helps lower the insurance premium, which is a very large fraction of launch costs to commercial customers. That's why you see a lot of student projects (which are done on the cheap, and usually are uninsured) blowing up.
Worst...sig...ever!