Well done to all !
Top Marks - have a beer on mankind and our fascination with what's out there.
A collaboration between NASA and ESA, and carried to orbit by Space Shuttle Discovery, the Hubble Space Telescope (HST) is entering its fourth decade of service. While the other space observatories in the program – Spitzer, Chandra and Compton – may generate a flicker of recognition among the general public despite their …
I think you have your urban myths crossed.
The mirror was made by PE on the same production line as the keyhole spy satelites - the Hubble being optically a keyhole facing up.
The polishing was 70s technology so if the polishing machine was computer controlled it was only for very small values of "computer". The shape fsck-up as nothing to do with the computer it was pure user error.
Kodak had bid to make a pair of mirrors so to have a backup if anything went wrong in polishing.
Although if they had used the same standard testing technique as PE and made the same mistake any error would have equally applied to both. The polishing technique had nothing to do with the error.
The big failing was PE not to have done a very simple test of the mirror to show any gross error - it relied on having a single super precise test that it did wrong. And then made a super precise mirror to the wrong shape. The lesson is that it's often worth doing a quick and dirty back of the envelope measurement/calculation before trusting the 10decimal place answer from the super precise machine with the flashing lights
They didn't do this test because of time constraints, although ironically Hubble then sat in storage for years after the challenger accident.
Just cos it's Friday, Im bored "working" from home and el'reg readers like to geek out.
The error is interesting.
To precisely test a mirror like this you assemble a bunch of lens to have exactly the same optical specification on a smaller scale (ie. a unit test) called a null corrector.
Obviously vital that you get all the spacings in this precisely correct .
So you make a very precise length metal rod out of Invar, a super hard version of stainless steel that changes very little with temperature (and is a real basterd to machine - I still have scars)
You focus a microscope on the end of the rod and then you know that when you replace the rod with the optics and are in the same focus you are in the same place. This way you never touch the rod.
You don't want some PFY to damage the ends of this rod you put a lens cap over it with a little hole for the microscope to look down. You then paint the cap back so you don't get any stray light.
Unfortunately a speck of paint came off the cap and went down the hole. The microscope got focussed on the resulting area of shiny metal on the top of the cap, a gnat's whatsit above than the end of the rod. The rest of the optics got assembled to this wrong position, the mirror got polished to match the optics and a lot of astronomers got very pissed off - and a bunch of us doing software deconvolution of out of focus images got bought a lot of very nice Sun wokstations. So it all worked out in the end.
PE had qty 2 different test optical systems, one using lenses, one using mirrors. The lens one was f*ed up as you say. They used both, got different results, chose to believe the lens based system. Should have figured out why they produced different results!!
Also, even with this screw up, if they had done an "end to end" test they would have caught it. Sadly NASA "decided" not to do the full test at Lockheed's "Blue Cube" in Sunnyvale where they tested full up KeyHole spy satellites. My belief is that CIA/NRO told NASA they couldn't have access.
That's part of the same faulty thinking though. A full test in a proper test range for this mirror (designed to support itself only in zero g) would have been complicated and expensive. A quick knife edge test with a business card would have shown the error - but you aren't going to do a dumb high school physics test on a $Bn program
The missus and I went down to Florida to watch the launch of Atlantis on STS-125, the final Hubble service mission. It was cool seeing two Shuttles ready for flight. (Endeavour was prepped for a rescue, because Atlantis couldn't reach the ISS from Hubble's orbit if necessary.) The launch was awesome.
Interestingly, STS-125 was the only Shuttle mission ever to launch early. NASA bumped it up a day, because keeping Endeavour ready to launch was expensive. It cost me a fortune to change our travel plans at the last minute to make it to the Cape in time for the launch.
Hubble's reflector was actually quite a hassle even before launch. First the alignment system didn't work and Perkin-Elmer had to subcontract that part of the project to a specialty instrumentation maker that hadn't worked on a telescope since Landsat (my old employer). Then during the years-long launch delay, some of the other contractors lost blueprints that turned out to be important for the repair. (Fortunately we still had those.)
One story I heard from a co-worker who was our main point of contact with Lockheed is that there was a physically separate design office tasked with re-creating the subsystems that couldn't be copied from Lockheed's military satellites, without the benefit of being able to look at blueprints or talk to the engineers at the other office. Sounded like fun...
Hubble captured the public's imagination and also made science appear more accessible. The telescope has been a remarkable success and it's longevity a real tribute to the designers and engineers that built and run it. Crucially the design allowed the telescope to be serviced so that critical components could be replaced in the event of failure and instruments upgraded as technology improved. You then look at the abominations that have followed with the James Webb Telescope and some of the other proposals that are being put forward. None are replacements and I don't see how any are going to deliver the value and public interest Hubble has. The JWST should have been canned years ago, it is just a bottomless pit of money and complexity. It's position at L2 means that it was designed not to be serviced, even if we developed something in the future that could get there.
Currently all the main space telescopes are either retired or are in the later stages of their expected life. The JWST appears to have been a giant ego trip/willy waving exercise by NASA to show how clever they are. The reality is that it is making them look incompetent.
I hope, simply from a money perspective that NASA do get the JWST up and working so that some science comes back but I am not overly optimistic.
Biting the hand that feeds IT © 1998–2020