I think that’s exactly it. NASA is very good at shooting rockets between planets and hitting the bullseye. Their record of non-landing space probes is very, very good. The ESA managed to do it last week too.
But robotically landing on another world is a different class of problem. That NASA has been as successful at it as it has is to the agency’s great credit.
Even landing on an airless world like the Moon is probably easier because it can be a more precise affair, and retro rockets are relatively small. But on a planet like Mars or Venus, the thing has to make it through an atmospheric re-entry and survive being buffeted by high winds and slammed to a stop. Or you can send a big enough craft to land with Retro rockets the way the Viking landers did, but they were really expensive.
This page has a good overview of the missions of the space program, and the first thing that jumps out at you is just how many of them fail.
No, that was the Mars Climate Orbiter. The major problem with the Hubble Telescope (there were several) was that a miscalculation caused the mirror to have the wrong curve in microgravity, causing badly out-of-focus images. (The mirrors they construct for big telescopes are ground to some truly astounding tolerance levels.)
I just watched the Nova episode, Mars: Life or Death, and it provides insight into the engineering and design challenges of the Mars Spirit mission. The problems of non-destructive testing, the unexpected problems with the parachute and balloon systems, and the risks involved in the landing. The graphics movies of the landing process were awesome. I really recommend this program.
Poking around Sam’s pages makes one appreciate the iffiness of this endeavor…
The Soviets had some successes landing on Venus, although due to the high temperatures involved, the probes lasted less than a couple of hours on the surface.
Well, in all fairness, a lot of failed Mars probes were Russian.
Don’t know why they had such good luck on Venus, though…maybe they “over engineered” those probes to survive Venusian conditions, hammering out a lot of potential flaws in the process?
But not perfect. The Mars Observer spacecraft, launched in 1992, was lost before it reached Mars.
That incident is a good example of why satellites are unreliable. As I understand it, the satellite was lost because of fuel leaking past a check valve, causing inadvertent propellant/oxidizer mixing (read: explosion). The valve had been used successfully on previous satellites; however it was only used for earth orbit satellites shortly after launch, not after a 11-month interplanetary flight. Even a flight proven component can fail if used in a different manner, and if you don’t test it rigorously enough.
You’d think people would learn from such a mistake. When the Japanese designed their Mars probe (Nozomi), they put a redundant valve to prevent the same problem. But if you put two valves in series to prevent accidental leakage, you double the chance that they fail to open when they are supposed to. And that’s exactly what happened; one of the valves opened only partially at the crucial moment of a powered flyby (engines fired at closest approach for maximum effect, sort of like carrying a big rock downhill and throwing it out before the next climb). The spacecraft didn’t gain enough speed for the planned 1-year journey to Mars. It eventually reached Mars after 5 years, but it was no longer functioning by that time.
