6 ms·

Ya. That said, flying a repair mission to install a corrective lens on a space telescope is/was totally in the realm of 'sexy solutions'.

sanoli 12y ago
The problem here is that messing up the mirror on such a project in the first place was not very sound engineering.
- brudgers 12y ago
  Professional engineering practice admits the possiblity of errors and ommissions. Insurance is the technique used to address it. In ordinary practice this is something that is purchased from agents with an associated cost. In the case of Hubble, it was purchased for the cost of making the telescope serviceable. Infallability is not an engineering practice. "Don't screw up" is not an engineering heuristic.
- jeffwass 12y ago
  Fair point, but 'messing it up' to such a high degree of accuracy that it could be corrected for via COSTAR does show soundness of engineering. Ie, mistakes will inevitably happen. But having the process refined enough to compensate, improvise, and enhance on original plans and imperfections does show good design.
- adventured 12y ago
  You're right, and it's a funny kind of right. It's an astounding thing when you're creating a mirror closest to perfection that had ever been made, and missing by 2.2 micrometres ruins it. And given that it was the 1980s and you're dealing with nano scale engineering, it was just that much more incredible.
- icegreentea 12y ago
  I should have made my point slightly more clear. I think that the OP's post is actually a non-sequitur. Both statements are generally true, but neither necessarily follows from the other.
Retric 12y ago
Of course we could have also easily launched 5 redundant space telescopes for the cost of those servicing missions. So, the point still stands sexy engineering is rarely the ideal approach. EX: Consider how much better the space shuttle would have worked without a cargo bay.
- damoncali 12y ago
  This is just not true. I worked on three of those servicing missions as an engineer. They were required to keep HST running and to enhance it's capability. Adding COSTAR to fix the mirror fuckup was one tiny piece of the servicing missions.
  - Retric 12y ago
    There were 6 Hubble missions all of them where manned. I suggest that the savings of using unmanned missions, not making a serviceable telescope, and saving 1 full trip would have paid for construction of 4 additional telescopes with minor design updates for instruments and or fix bugs but no changes to major systems. With the added advantage of not risking astronaut lives on a publicity stunt.
    - damoncali 12y ago
      How would you have gotten even one HST sized telescope into orbit without a manned mission? Do you know what life span of a Hubble would have been if it were not serviceable? Your approach would have cost more, not less. More manned fights, more time consuming, expensive design and construction. Believe it or not, they thought about this stuff.
      - Retric 12y ago
        How would you have gotten even one HST sized telescope into orbit without a manned mission? How do we get similar sized spy satellites into orbit without people? http://www.space.com/16000-spy-satellites-space-telescopes-nasa.html http://www.space.com/16000-spy-satellites-space-telescopes-n... As to lifetime, we don't need a 30 year lifespan. If you average 6 years * 5 satellites you get 30 years and I suspect we could easily average 8-10 years. As to thinking about this stuff. Clearly, but Hubble was specifically designed to make the Shuttle seem more relevant. Just about all of the major decisions where made for political not technical reasons.
CWuestefeld 12y ago
My grandfather worked on the Hubble mirror. It was, at the same time, one of his proudest accomplishments as well as a source of embarrassment. Everybody agrees that the mistakes with the mirror were a real screw-up, something that should never have happened with better management and organization. However, my grandfather also pointed out that there's another direction to point the blame finger, that actually never enters into the discussions. According to gramps, it was understood that they wouldn't be able to get the mirror perfect (admitting, of course, the actual error was worse than any spec allowed for). Apparently they knew they weren't able to control perfectly for the distortions that would arise from moving the mirror from Earth's gravity to weightless, and perhaps other factors. But the original goal for the space shuttle was supposed to make it relatively easy to bring the thing up into space, try it out, and tweak it or even bring it back for repairs. According to gramps, this was always a fallback possibility, and the fact that it was so difficult and expensive in practice reveals as much about the failures of the space shuttle program as it does about the Hubble mirror - but the latter is always the focus of the discussion. Indeed, the shuttle gets credit as the savior, if anything, rather than any blame for how poorly it served its intended mission (including how many years its own program delays resulted in delays for the Hubble itself). Caveat: this is all completely based on the opinion of one guy as communicated to his grandson, and potentially suffers from all the errors that this might imply. EDIT - fixed mis-wording in para 3.

sanoli 12y ago

The problem here is that messing up the mirror on such a project in the first place was not very sound engineering.

brudgers 12y ago
Professional engineering practice admits the possiblity of errors and ommissions. Insurance is the technique used to address it. In ordinary practice this is something that is purchased from agents with an associated cost. In the case of Hubble, it was purchased for the cost of making the telescope serviceable. Infallability is not an engineering practice. "Don't screw up" is not an engineering heuristic.
jeffwass 12y ago
Fair point, but 'messing it up' to such a high degree of accuracy that it could be corrected for via COSTAR does show soundness of engineering. Ie, mistakes will inevitably happen. But having the process refined enough to compensate, improvise, and enhance on original plans and imperfections does show good design.
adventured 12y ago
You're right, and it's a funny kind of right. It's an astounding thing when you're creating a mirror closest to perfection that had ever been made, and missing by 2.2 micrometres ruins it. And given that it was the 1980s and you're dealing with nano scale engineering, it was just that much more incredible.
icegreentea 12y ago
I should have made my point slightly more clear. I think that the OP's post is actually a non-sequitur. Both statements are generally true, but neither necessarily follows from the other.

Retric 12y ago

Of course we could have also easily launched 5 redundant space telescopes for the cost of those servicing missions. So, the point still stands sexy engineering is rarely the ideal approach. EX: Consider how much better the space shuttle would have worked without a cargo bay.

damoncali 12y ago
This is just not true. I worked on three of those servicing missions as an engineer. They were required to keep HST running and to enhance it's capability. Adding COSTAR to fix the mirror fuckup was one tiny piece of the servicing missions.
- Retric 12y ago
  There were 6 Hubble missions all of them where manned. I suggest that the savings of using unmanned missions, not making a serviceable telescope, and saving 1 full trip would have paid for construction of 4 additional telescopes with minor design updates for instruments and or fix bugs but no changes to major systems. With the added advantage of not risking astronaut lives on a publicity stunt.
  - damoncali 12y ago
    How would you have gotten even one HST sized telescope into orbit without a manned mission? Do you know what life span of a Hubble would have been if it were not serviceable? Your approach would have cost more, not less. More manned fights, more time consuming, expensive design and construction. Believe it or not, they thought about this stuff.
    - Retric 12y ago
      How would you have gotten even one HST sized telescope into orbit without a manned mission? How do we get similar sized spy satellites into orbit without people? http://www.space.com/16000-spy-satellites-space-telescopes-nasa.html http://www.space.com/16000-spy-satellites-space-telescopes-n... As to lifetime, we don't need a 30 year lifespan. If you average 6 years * 5 satellites you get 30 years and I suspect we could easily average 8-10 years. As to thinking about this stuff. Clearly, but Hubble was specifically designed to make the Shuttle seem more relevant. Just about all of the major decisions where made for political not technical reasons.

CWuestefeld 12y ago

My grandfather worked on the Hubble mirror. It was, at the same time, one of his proudest accomplishments as well as a source of embarrassment. Everybody agrees that the mistakes with the mirror were a real screw-up, something that should never have happened with better management and organization. However, my grandfather also pointed out that there's another direction to point the blame finger, that actually never enters into the discussions. According to gramps, it was understood that they wouldn't be able to get the mirror perfect (admitting, of course, the actual error was worse than any spec allowed for). Apparently they knew they weren't able to control perfectly for the distortions that would arise from moving the mirror from Earth's gravity to weightless, and perhaps other factors. But the original goal for the space shuttle was supposed to make it relatively easy to bring the thing up into space, try it out, and tweak it or even bring it back for repairs. According to gramps, this was always a fallback possibility, and the fact that it was so difficult and expensive in practice reveals as much about the failures of the space shuttle program as it does about the Hubble mirror - but the latter is always the focus of the discussion. Indeed, the shuttle gets credit as the savior, if anything, rather than any blame for how poorly it served its intended mission (including how many years its own program delays resulted in delays for the Hubble itself). Caveat: this is all completely based on the opinion of one guy as communicated to his grandson, and potentially suffers from all the errors that this might imply. EDIT - fixed mis-wording in para 3.