YAY CAREFUL AND CONSIDERED ENGINEERING!
Thank you to all the people who have worked so hard on this.
NASA’s beloved Hubble Space Telescope is able to snap the heavens again after overcoming a hardware issue that had plagued it for more than a month. Its onboard payload computer, which controls its instruments, mysteriously froze, forcing the main computer to put the orbiting observatory's sensors into an inactive safe mode. …
I was just thinking how wonderful it would be if we had a system - say, a reusable spacecraft designed to repair things. It could have a large area for things to be taken up or to provide a base for repairs, and maybe a powered arm to hold things or provide a platform for astronauts to work from. Specialists could be selected for each mission, flown up by dedicated flight-crew. At the end of the mission, the craft could reenter the atmosphere and land on a runway. Of course, that's just science fiction, isn't it?
This is actually why I was interested in the approach that Virgin has taken. If they can scale this up to higher orbits and make some sort of engineering vehicle it could eventually make satellite repair more affordable, also because you wouldn't need super healthy people that can take several Gs during takeof. Still need to train them for space, though, but I like having options.
Hubble is very similar to the dozen or so KH-11 spy satellites the NRO has operated (which have been replaced by the newer KH-12 which is probably of similar size and orbital altitude at least) and those have presumably been deorbited without issue.
At least you'd think it would make the news if a spy satellite crashed down in someone's backyard.
Hmm. I'm not sure about the similarity to the Keyholes. Apart from the fact that the latter point towards Earth instead of away, they were not co-designed with Hubble at all. I heard that somebody from the dark side sat in on the Hubble design team, taking notes but basically saying nothing, occasionally nodding and smiling when they happened to re-invent the same things as the early KHs.
The Hubble used the KH-11 platform - the Keyhole program came first and without it leveraging that technology heavily there likely would not have been a Hubble. The focal length for Hubble was a lot longer, but the 2.4m main mirror assembly, location of instrument bay, etc. is identical to KH-11.
The NRO even gave NASA two "leftover" KH-11s about 10 years ago, with the idea they could be retrofitted as space telescopes. One has been outfitted to fulfill the WFIRST (now called Nancy Grace Roman Space Telescope) mission and is set to launch in a few years. Because the KH-11 has a shorter focal length its field of view is 100x larger than Hubble, but for the WFIRST mission that was considered an advantage.
The other KH-11 remains in storage.
While the Perkin-Elmer designed Hubble camera is said to be very similar to those used in reconnaissance satellites, I'm not sure that the rest of the platforms are all that similar. Hubble's orbit is higher presumably because it isn't trying to capture high resolution images of Earth bound objects. And the inclinations are quite different -- 28 degrees vs (mostly) 97 degrees. The probable reasons for the high inclination are interesting, but more complicated than I care to get into. They wouldn't apply to Hubble whose orbit parameters likely depend more on the latitude of the Cape Canaveral launch facility than anything else. Inclination is important as it strongly influences what latitude band the deorbiting satellite can come down in.
Actually, I doubt the spooks care all that much if their 12 tonne (or more) spy satellite lands on someone, but I'm guessing that that they very much care if it crashes in Russia, China, or some place else where (potentially) hostile intelligence folks can collect and analyze whatever pieces survive. So deorbiting is presumably a significant part of spy satellite mission planning.
For Hubble on the other hand, deorbiting seems to have been left as a problem for future generations. At least that's what a quick internet search leads me to believe.
The rest of the platform (other than the contents of the instrument bay and focal length) is basically identical to KH-11. Without the KH-11, there probably would never have been a Hubble, since it heavily leveraged that technology. The camera is the main difference - the NRO keeps that classified and even removed all the CCDs and other imaging gear from the two KH-11 they donated to NASA about 10 years ago.
...which means, of course, that they have a better replacement :-)
IIRC, one of the (formerly) top secret advances in the KH-11, was a deformable mirror to compensate for atmospheric distortion. How they figured out what the distortion was, in order to counteract it, is probably another big secret.
BRB...someone at the door...
"but I'm guessing that that they very much care if it crashes in Russia, China, or some place else where (potentially) hostile intelligence folks can collect and analyze whatever pieces survive"
I'm guessing that ensuring the spy satellite flies over the areas they want to observe is far more of a concern than where it eventually deorbits. Unfortunately, I suspect that orbits that let you observe Russia and China also mean you could eventually crash in Russia or China.
When NASA wanted to test the Hubble main and secondary mirrors together they did not have a collimator big enough, so they asked if they could use the NSA's. The NSA replied 'yes, but only if we get to point it down some of the time.' As it was an international venture NASA could not agree to that, so Hubble was launched into space without a collimation test and the faulty optics revealed.
So I'm guessing that although the NSA had some excellent spy satellites at the time, none was quite as powerful as Hubble. Of course that might have changed, as the DoD had their own space shuttle and did quite a few launches of their own...
I see. They had this special big collimator but they'd just made it for fun, not because they wanted to use it on mirrors.
And they had a secret extra space shuttle (no, not the X-37 which did not exist then and is anyway too small), and that's not at all a made up story from West Wing.
Not quite as basic.
They changed the power supply.
Which is indeed a very common replacement (on cheaply sourced pc in my old job - no, I'm not suggesting the Hubble parts were sub standard, our power supplies failed after only a couple of years, I think just outside warranty).
Great job though!
My Dyson lamp* started to switch itself off at inconvenient times. They sent me a new power supply unit, it works fine now (well, so far). Though this came through the post, so not as tricky as fixing Hubble.
*Yes, I am easily seduced by shiny tech toys, I admit it. Is there a tech equivalent of Alcoholics Anonymous?
Nah, I'm not into test equipment, it is the full-blown fanboi shiny shiny stuff I like, although I have managed to avoid the latest iMacs on the basis that a) I cannot decide what colour it should be, and 2) they don't have an SD card slot. (Mainly 'a', of course, but '2' is the one I tell the salespeople.)
Easy to check out before spending any money if you have access to a bench power supply . When my Blu-ray player failed it was pretty easy to point the finger at the PSU, since the top of one of the chips in that module had literally blown off. I found the remains rattling around the case, and could just about read the number. But a replacement was no longer available and a new board was more than the €12.50 I paid second hand for the whole player. Still can't chuck it out though, there must be some parts useable for future projects?
[ I've just moved house and it needed two trips to the dump to dispose of the "some parts useable" stash that I had before. Now just starting again... ]
It's always either the 1) O2 sensor, 2) the DNS, or 3) the power supply.
No O2 sensor in space. No name-service neither. Ergo must be power supply.
But like any commodity tech support, you have to follow the script, pick-up every stitch, boffins are out to make it rich, redundant rabbits runnin' in the ditch, must be the season of the twitch.
I think it is amazing what can be done from the ground to "fix" aging hardware in space that has outlived its expected life by a decade. This shows the value of designing redundancy into a device properly and using this redundancy properly. All of this through a narrow radio channel!
I am incredibly impressed.
It is indeed impressive that they were/are able to do this, thanks to careful forethought and disaster recovery planning, particularly so long after its design lifetime.
However, any bean counter worth his salt must be so distraught at all these extra beans that have been unnecessarily to make it possible.
I find it distressing that the bean counter attitude is still so prevalent... as above, so below.
 in the view of the counter of beans.
That in an odd way reminds me of a problem (years) ago when I had to go to an MOD site to figure out why a page printer (LED not laser) was producing garbled printouts from a mainframe it was connected to via RS-232.
The slight complexity was there was no separate "flow control" on the host systems serial interface, so it was all running via XON/XOFF.
Off I went down to the site in the South West - one of those with soldiers on the gates with real guns loaded with real bullets! - complete with an old "luggable" RS-232 protocol analyser with built in 5" CRT. Everything seemed ok, except a lot of the XOFFs which should have been sent by the printer to tell the host system to shut up for a bit, seemed to be being ignored or even missing.
One of our field engineers had been down to it twice and changed most of the controller boards - the serial interface board had been replaced 3 times. All to no effect.
Bit of thinking then out came the multi-meter and onto the PSU. The +12V and -12V outputs were somewhat low (less than +8V and -8V respectively). Turns out that those voltages, only used by the serial interface, (standard RS-232 signalling levels) were just too low for it signal reliably to the host system. So it just wasn't "hearing" the XON/XOFFs.
New PSU, with the correct outputs and normal service resumed.
And I got to "chaperone" one of our other printers on the same site which was going through "naval submarine" validation testing. Basically subjecting it to various shock levels (on a huge shock table) in various orientations to ensure:
1. It worked up to a certain level.
2,. Above that it didn't have to work but nothing had to fly off (simulated battle conditions).
And the last test - see just how stood up to "destruction level shock testing". Basically give it the highest level shock and see what happened. It was prettyy buggered inside, but only the main "smoked plastic" cover flew over 15 feet away, a couple of knows flew off and something cracekd. Otherwise stayed pretty intact.
Fun ain't they getting in with security ranging from the "Just go in" to the third party security guy that would let you in after restating they needed 24 hours notice (Yet again) & the whole checking the underside of the vehicle with mirrors for bombs. I did the ones in Devon, I think one in Cornwall, Whale Island & Norton Fitzwarren. There was also one side trip to the base of the SBS while returning from Whale Island.
Royal Marines if anyone's interested & hasn't worked it out.
I arrived at the RAF site on time. The guard on the gate gets out his wheelie mirror on a stick to view the underside of my 1999 registration Ford Focus motorcar. There is a pause while he manoeuvres it, then:
Me: Very tentatively "Have, have you found something?"
Guard: "No, sir, I've just farted."*
*(My physical presence and demeanour always inspire respect and deference in equal measure.)
Our problem was never the security on the gate, our problem was always the brain dead squaddies/matlots left to guard whatever room we were installing the kit in.
Come back from lunch and its "Cant let you in there , its all secret stuff"
And my team leader would then point out that he'd designed it, bob had built it and I'd installed it and we all knew exactly what it was and what it did... and in any case you were here 30 mins ago and watched us all come out of that room saying we were off for lunch.
I wonder if the military hand out awards for mindlessly following orders...
"I wonder if the military hand out awards for mindlessly following orders..."
Definitely: Awards, promotions, medals, the works. Mindless obedience is the only way to make normal people throw away their lives for some arrogant bigwig's incompetence. The first World War was a striking example of this, hundreds of thousands of soldiers being sent to certain and pointless death, charging with fixed bayonets heavily fortified machine gun positions and being mowed down, assault after assault, just because of their generals' arrogance and stupidity insisted on "trying again". For several years the front only moved by a couple hundred meters, while the soldiers' corpses piled higher and higher on both sides.
All hail our Nazi overlords.
Your post betrays a huge amount of ignorance. Also a complete disrespect for the privileged situation which you so clearly believe that you deserve.
You'd be Trump, with spurs to escape the draft. Before calling all fallen troops 'losers'.
"Bit of thinking then out came the multi-meter and onto the PSU. The +12V and -12V outputs were somewhat low (less than +8V and -8V respectively). Turns out that those voltages, only used by the serial interface, (standard RS-232 signalling levels) were just too low for it signal reliably to the host system. So it just wasn't "hearing" the XON/XOFFs."
Back in the old days of field engineering (MSDOS, so not all that long ago), I came across a similar incident involving a serial printer. I didn't have any of the right test gear you had, other than loop-back pluigs and a copy of CheckIt!. Anyway, there was clearly nothing wrong with the serial port as such and the printer worked on another PC. Out of curiosity as much as anything else, I tested the +/-12v and like you, got about 9v on each. Future visits to errant serial printers involving strange symptoms like missing characters etc, one of the first things was to just plug a new PSU in first since that was actually one of the quickest tests, ie they had leads long enough to not have to remove the old one for testing and I did once come across one where testing the 12v lines didn't show a problem (it was only a voltage drop under certain load conditions)
I don't know how much fuel is left in Hubble, but if there's any it should NOW be used to de-orbit the thing.
I mean, this thing is at least two decades past its designed lifetime and could fail at literally any moment. Having yet another hulk (here's looking at you EnviSat) floating around up there for hundreds of years as a potential collision target is unacceptable.
Hubble doesn't have any fuel, rocket fuel was too dangerous for a payload that has to be approached by a crew.
Pointing is done by reaction wheels and magnetic bars that 'push' against the Earth's field. Its orbit had to be boosted by space shuttle missions.
But you don't have to worry about it staying up for more than another few years, it is too big and too low to remain for long. Although there is no specific plan for a de-orbit other than "thoughts and prayers"
You are absolutely correct about Hubble using reaction wheels to do the essential job of pointing it in the correct direction.
Astronomers, being picky buggers, do like to be able to move their telescope around to look at different things.
However, while safety for crews approaching Hubble during servicing missions was a factor, it wasn't a major one. After all, STS crews would deploy payloads mounted on PAMs (Payload Assist Modules) which are basically a rocket powered platform for firing payloads out of orbit and off elsewhere (to the other planets etc).
The major reason for the reaction wheels? Science....
The alternatives were small maneuvering thrusters which used propellant - typically hydrazine - for moving things about in space. Fine if you are generally moving your spacecraft from Point A to Point B.
If, as is the case with Hubble, you are not making it "go" anywhere, just slewing it around to point in different directions. Then your spacecraft can quickly end up sitting in a orbiting cloud of residue from the hydrazine.
At best this will interfere with all the optical instruments - the whole point of it being in space was to get above most of the atmosphere so the last you want if for your spacecraft to create its own.
At worst it get onto, and seriously contaminate, the surface of the single most critical component that could never be "swapped out - the main mirror.
At some point Hubble (& various other NAS bits of kit) do reach the stage where, despite all the skills of the teams, they can no longer be "revived" due to kit being damaged & alternative fixes exhausted.
But really enjoy seeing the "saves" & kit getting an extended lifespan despite the odds (especially Hubble which has done some great science over the years & maybe even more importantly created some stunning images that have engaged some people previously not that interested in astronomy as public engagement is good as science should be for everyone to access in some form)
Biting the hand that feeds IT © 1998–2021