7 ms·

My grandfather worked on the Hubble's mirror - the part with all the controversy. It was, simultaneously, both what he was most proud of and most ashamed. One

My grandfather worked on the Hubble's mirror - the part with all the controversy. It was, simultaneously, both what he was most proud of and most ashamed. One thing he would point out, that I never see discussed, was that at the time they built the mirror, there was an expectation that it may not be perfect, on account of how the Earth's gravity would slightly deform the mirror during manufacturing. As he tells the story, they'd tried to factor that into the design, but had less confidence in it. This, of course, was not the reason for the eventual problems, but to illustrate that this kind of thing was understood. The big picture of the failure relates to the Shuttle program as well. At the time Hubble was being developed, its mission was supposed to allow for a lot more around the "shuttle" aspect: not just that it could be reused, but that it could be reused in an actual shuttling capacity, like an orbital pickup truck. So it was supposed to be possible to bring the telescope up into orbit (away from gravitational stresses), see how it worked, and if necessary, bring it back down for corrections. Since the end result of the Shuttle design cut down significantly in its capabilities in this regard, the Hubble program was left without its "Plan B".

walrus01 5y ago
In the end it turned out that for the life of the shuttle program, each launch ended up costing between 700 million and 1.2 billion, so the 1970s goal of an economically reusable spacecraft was definitely not achieved.
dr_orpheus 5y ago
It still amazes me that one, the mirrors are sensitive enough to the deformation of Earth's gravity to have an effect and two, that we are now very good at compensating for this now. Also consider that the mirrors on James Webb space telescope need to account for both deformation from gravity when they are manufactured but also that they are manufactured at room temperature and will be operated at cryo temperatures. But JWST is also special since they have actuators to put stress on the mirror segments to be able to reshape them slightly on-orbit to focus.
- downrightmike 5y ago
  To be fair, Kodak made a sister mirror that wasn't chosen to go up, and when the issue on Hubble was found out, Kodak's mirror was found to be perfect.
  - eckmLJE 5y ago
    Great tidbit -- found this in the NYT archive (likely paywall): https://www.nytimes.com/1990/07/18/us/hubble-has-backup-mirror-unused.html https://www.nytimes.com/1990/07/18/us/hubble-has-backup-mirr...
  - jcims 5y ago
    It's fun to think that there is a bullet list of directions that would take me from where I'm typing this comment to standing in front of that exact mirror.
  - chiefgeek 5y ago
    I worked in the Commercial and Govt Systems group at Kodak in the mid 90s and took a class on image compression from two guys who were part of the team that figured out how to mathematically correct the early, flawed Hubble images. Was really cool!
- CWuestefeld 5y ago
  My same grandpa that worked on the Hubble mirror also worked on anti-ICBM lasers back in the 70s. Apparently these devices had a tendency to melt their own mirrors, and that led to the development on self-deforming mirrors to deflect heat from hotspots. (He also worked on the TEAL AMBER and TEAL BLUE satellite surveillance systems, and probably others that he never even talked about.)
  - 23B1 5y ago
    Is there a good book I can read about all of this cold-war era tech? I find it endlessly fascinating the number of programs and technology advancements that happened during this period.
jessaustin 5y ago
This, of course, was not the reason for the eventual problems, but to illustrate that this kind of thing was understood. For some reason I had thought this (gravity stuff) was the problem. What was the actual problem?
- azernik 5y ago
  A device for measuring the final grinding had been miscalibrated (by a known, constant amount). The result was a spherical aberration, of a kind that is familiar to a lot of users of cheaper lenses and mirrors but on a much smaller scale.
  - krisoft 5y ago
    Exactly as azernik says. If you are interested in a lot more detail this reports spills the tea: https://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf https://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf
spullara 5y ago
Fascinating story about how the null corrector that was supposed to ensure that the mirror was correct was flawed: https://en.wikipedia.org/wiki/Null_corrector https://en.wikipedia.org/wiki/Null_corrector
petertodd 5y ago
> My grandfather worked on the Hubble's mirror - the part with all the controversy. It was, simultaneously, both what he was most proud of and most ashamed. I think a simple way to explain what your grandfather should feel proud of, is they documented the construction process so thoroughly that they were able to _precisely_ figure out what they did wrong, and fix it, on their first try!
- CWuestefeld 5y ago
  That's a really positive way to frame it. Thanks for the perspective.
- chasd00 5y ago
  From what I recall the mirror was built to spec, it was the spec that was wrong not the mirror. Do I have that backwards?
  - abestic9 5y ago
    The mirror was precisely polished "to spec" with custom-built equipment that was assembled incorrectly. They correctly saw the distortion issue on the ground with the conventional kit, but rejected the result as erroneous because it tested fine with the "better" testing equipment. https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Origin_of_the_problem https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Origin_...
  - NickNameNick 5y ago
    I thought that the instrument used to measure the curvature of the mirror was mis-calibrated.
    - ArnoVW 5y ago
      Yes. A tiny area of paint was chipped. Unfortunately that meant that the distance was increased by the distance of the layer of paint.
Abishek_Muthian 5y ago
Thanks for sharing. I wonder how often retired NASA engineers are consulted. Imagine as an Engineer having to troubleshoot something which launched into space before you were born. Of course there must be perfect documentation, SOP along with replicas on land.
deviledeggs 5y ago
The Hubble was also a redesigned spy satellite. The main mirror design was chosen so it could be built with existing mirror tech made for military imaging satellites. There's probably a dozen satellites out there right now with optics an order of magnitude better than Hubble. They're just pointed down

walrus01 5y ago

In the end it turned out that for the life of the shuttle program, each launch ended up costing between 700 million and 1.2 billion, so the 1970s goal of an economically reusable spacecraft was definitely not achieved.

dr_orpheus 5y ago

It still amazes me that one, the mirrors are sensitive enough to the deformation of Earth's gravity to have an effect and two, that we are now very good at compensating for this now. Also consider that the mirrors on James Webb space telescope need to account for both deformation from gravity when they are manufactured but also that they are manufactured at room temperature and will be operated at cryo temperatures. But JWST is also special since they have actuators to put stress on the mirror segments to be able to reshape them slightly on-orbit to focus.

downrightmike 5y ago
To be fair, Kodak made a sister mirror that wasn't chosen to go up, and when the issue on Hubble was found out, Kodak's mirror was found to be perfect.
- eckmLJE 5y ago
  Great tidbit -- found this in the NYT archive (likely paywall): https://www.nytimes.com/1990/07/18/us/hubble-has-backup-mirror-unused.html https://www.nytimes.com/1990/07/18/us/hubble-has-backup-mirr...
- jcims 5y ago
  It's fun to think that there is a bullet list of directions that would take me from where I'm typing this comment to standing in front of that exact mirror.
- chiefgeek 5y ago
  I worked in the Commercial and Govt Systems group at Kodak in the mid 90s and took a class on image compression from two guys who were part of the team that figured out how to mathematically correct the early, flawed Hubble images. Was really cool!
CWuestefeld 5y ago
My same grandpa that worked on the Hubble mirror also worked on anti-ICBM lasers back in the 70s. Apparently these devices had a tendency to melt their own mirrors, and that led to the development on self-deforming mirrors to deflect heat from hotspots. (He also worked on the TEAL AMBER and TEAL BLUE satellite surveillance systems, and probably others that he never even talked about.)
- 23B1 5y ago
  Is there a good book I can read about all of this cold-war era tech? I find it endlessly fascinating the number of programs and technology advancements that happened during this period.

jessaustin 5y ago

This, of course, was not the reason for the eventual problems, but to illustrate that this kind of thing was understood. For some reason I had thought this (gravity stuff) was the problem. What was the actual problem?

azernik 5y ago
A device for measuring the final grinding had been miscalibrated (by a known, constant amount). The result was a spherical aberration, of a kind that is familiar to a lot of users of cheaper lenses and mirrors but on a much smaller scale.
- krisoft 5y ago
  Exactly as azernik says. If you are interested in a lot more detail this reports spills the tea: https://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf https://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf

spullara 5y ago

Fascinating story about how the null corrector that was supposed to ensure that the mirror was correct was flawed: https://en.wikipedia.org/wiki/Null_corrector https://en.wikipedia.org/wiki/Null_corrector

petertodd 5y ago

> My grandfather worked on the Hubble's mirror - the part with all the controversy. It was, simultaneously, both what he was most proud of and most ashamed. I think a simple way to explain what your grandfather should feel proud of, is they documented the construction process so thoroughly that they were able to _precisely_ figure out what they did wrong, and fix it, on their first try!

CWuestefeld 5y ago
That's a really positive way to frame it. Thanks for the perspective.
chasd00 5y ago
From what I recall the mirror was built to spec, it was the spec that was wrong not the mirror. Do I have that backwards?
- abestic9 5y ago
  The mirror was precisely polished "to spec" with custom-built equipment that was assembled incorrectly. They correctly saw the distortion issue on the ground with the conventional kit, but rejected the result as erroneous because it tested fine with the "better" testing equipment. https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Origin_of_the_problem https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Origin_...
- NickNameNick 5y ago
  I thought that the instrument used to measure the curvature of the mirror was mis-calibrated.
  - ArnoVW 5y ago
    Yes. A tiny area of paint was chipped. Unfortunately that meant that the distance was increased by the distance of the layer of paint.

Abishek_Muthian 5y ago

Thanks for sharing. I wonder how often retired NASA engineers are consulted. Imagine as an Engineer having to troubleshoot something which launched into space before you were born. Of course there must be perfect documentation, SOP along with replicas on land.

deviledeggs 5y ago

The Hubble was also a redesigned spy satellite. The main mirror design was chosen so it could be built with existing mirror tech made for military imaging satellites. There's probably a dozen satellites out there right now with optics an order of magnitude better than Hubble. They're just pointed down