It's worth noting that telescopes that have suffered from severe issues with gyroscopes or reaction wheels (e.g. Compton, Hubble, Kepler, Swift) have typically done only after operating long past their nominal planned lifespans:

• Hubble's planned lifespan was 15 years; it's now 34.
• Compton was supposed to operate for about half a decade; it was deorbited after 9 years.
• Swift's nominal mission lasted 2 years; it's now 20.
• Kepler's initial mission was for 3.5 years; it had reaction wheel issues after 3 and was shut down after 9 because it ran out of fuel.

Now, it's reasonable to assume that a world-class telescope has a good chance of having its mission extended. All the same, it's not worth spending time, money and weight to deal with problems that would likely arise far beyond that lifetime. Hubble was completed well over its initial budget; given the difficulties in funding the dang thing, nobody could justify adding to the bottom line to throw in another backup part.

Compton, for instance, was given the hardware to last five years and hopefully then some; it's awesome that it made it to nine. A mission ending when gyroscopes or reaction wheels fail is usually a great problem to have.

It's worth noting that telescopes that have suffered from severe issues with gyroscopes or reaction wheels (e.g. Compton, Hubble, Kepler, Swift) have typically done so only after operating long past their nominal planned lifespans:

• Hubble's planned lifespan was 15 years; it recently turned 34.
• Compton was supposed to operate for about half a decade; it was deorbited after 9 years.
• Swift's nominal mission lasted 2 years; it's now 20.
• Kepler's initial mission was for 3.5 years; it had reaction wheel issues after 3 and was shut down after 9 because it ran out of fuel (not from further reaction wheel problems).

Now, it's reasonable to assume that a world-class telescope has a good chance of having its mission extended. All the same, it's not worth spending time, money and weight to deal with problems that would likely arise far beyond that lifetime. Hubble was completed well over its initial budget; given the difficulties in funding the dang thing, nobody could justify adding to the bottom line to throw in another backup part.

Compton, for instance, was given the hardware to last five years and hopefully then some; it's awesome that it made it to nine. A mission ending when gyroscopes or reaction wheels fail is usually a great problem to have.

Finally, while problems with these parts have led to the end of a number of missions, they're usually not the reason space telescopes have to be shut down. Funding cuts (GALEX), cooling issues (Herschel, Spitzer, Planck), fuel shortages (Kepler) and computer failures (CoRoT) have all taken out premier observatories -- again, typically at or well past the end of their nominal missions.