🚀 go-pugleaf

-From: dhl@mrdog.msl.com (Donald H. Locker)
-Subject: Hubble's mirror
-Date: 3 Nov 92 21:39:06 GMT
-Organization: Chelsea MSL, Inc., Chelsea, MI

-Now that I know a little about mirror-making, I'd like to hear again
-how the Hubble mirror contractor messed up the figure of the main
-mirror.  I understand it has spherical aberration, but wonder how
-[Rockwell?] managed to do that.

The contractor who made the mirror was Perkin Elmer (now Hughes Danbury).
The manufacture specification called for the use of two "null correctors" -
a highly accurate reflective corrector, and a less accurate corrector, which
I believe was based on transmission rather than reflection. The reflective
corrector is designed to be set to a range of figures, so it can be used
for many different mirrors. To make sure it was set correctly for the HST
primary mirror, a calibration rod was provided by NIST. Two or three
Perkin Elmer employees set the adjustment of the corrector, and unfortunately
they used the calibration rod incorrectly, thus giving the reflective
corrector the wrong setting. The reflective null corrector was used in the
manufacture of the mirror and in the most precise tests, so while the final
figure of the mirror matched very precisely the setting of the reflective
corrector, it was the wrong figure. Tests were made (as required in the spec)
using the transmission corrector, and those tests *did* show the error, but
Perkin Elmer made the arbitrary decision that since the reflective
corrector was so much more precise, any discrepancy must be the fault of
the transmission corrector. Furthermore, they didn't bother to inform NASA
of the discrepancy. (NASA only had something like a third of the number of
inspectors that they would normally use for a project of this size and
complexity, attributed to severe budget cuts in the project and heavy
pressure from DOD to keep out of the manufacturing process.)

There was an opportunity to conduct an integrated test of the finished
telescope, but this was not done because of concerns that the mirror
might become contaminated with organic matter and thus useless in the UV
spectrum (perhaps a good thing, because UV astronomy is now one of the
strongest points of HST), and because such a test could not adequately
compensate for gravitational effects, etc., and thus would not have
determined whether the scope was fully within specs. (It would have
detected the large error present, but there was no reason to believe there
would be such an error, assuming that Perkin Elmer had done their job
with the previous testing. Also, I suspect that if the error *had* been
detected after assembly, it might well have led to cancellation of the
entire project - the time to find the error was before assembly.)

Fortunately, there is considerable hope that the WF/PC II installation
(which had been planned even before discovery of the error) and the
addition of COSTAR will compensate for nearly all of the error, at an
incremental cost of only a small fraction of the initial cost of HST.
(The main loss is three years of the best measurements - HST is even now
producing world-class results, but it could have been doing even better
without the error. Computational deconvolution has been helping considerably,
at least for bright objects. The High Speed Photometer instrument will be
sacrificed for the installation of COSTAR, but it's been the least-used
instrument - only employed a few times.)

Other changes planned for the upcoming HST Shuttle service mission:
 - The European supplied solar panels that cause HST to wobble for several
   minutes after each day-night transition will be replaced with new panels
   (also from Europe, but they claim that this time they got the design
   right).
 - Some of the gyroscopes needed for the most precise attitude sensing have
   gone bad, thus reducing the redundancy of the system. The bad ones are
   scheduled to be replaced.
 - There *may* be a plan to replace the current control computer with a more
   powerful one, which would make it easier to handle complex maneuvers
   like compensation for wobbling. (I don't remember whether that's scheduled
   or not.)

All of that is old news - here's something new: NASA is now seriously
considering suing Hughes Danbury for failing to inform them of the errors
they detected.

John Roberts
roberts@cmr.ncsl.nist.gov