Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
NASA IT

NASA Can't Figure Out What's Causing Computer Issues On The Telescope (npr.org) 84

Posted by msmash from the closer-look dept.
The storied space telescope that brought you stunning photos of the solar system and enriched our understanding of the cosmos over the past three decades is experiencing a technical glitch. From a report: Scientists at NASA say the Hubble Space Telescope's payload computer, which operates the spacecraft's scientific instruments, went down suddenly on June 13. Without it, the instruments on board meant to snap pictures and collect data are not currently working. Scientists have run a series of tests on the malfunctioning computer system but have yet to figure out what went wrong. "It's just the inefficiency of trying to fix something which is orbiting 400 miles over your head instead of in your laboratory," Paul Hertz, the director of astrophysics for NASA, told NPR. "If this computer were in the lab, we'd be hooking up monitors and testing the inputs and outputs all over the place, and would be really quick to diagnose it," he said. "All we can do is send a command from our limited set of commands and then see what data comes out of the computer and then send that data down and try to analyze it."

At first NASA scientists wondered if a "degrading memory module" on Hubble was to blame. Then on Tuesday the agency said it was investigating whether the computer's Central Processing Module (CPM) or its Standard Interface (STINT) hardware, which helps the CPM communicate with other components, caused the problem. Hertz said the current assumption, though unverified, was that the technical issue was a "random parts failure" somewhere on the computer system, which was built in the 1980s and launched into space in 1990. "They're very primitive computers compared to what's in your cell phone," he said, "but the problem is we can't touch it or see it." Most of Hubble's components have redundant back-ups, so once scientists figure out the specific component that's causing the computer problem, they can remotely switch over to its back-up part.
This discussion has been archived. No new comments can be posted.

NASA Can't Figure Out What's Causing Computer Issues On The Telescope More

NASA Can't Figure Out What's Causing Computer Issues On The Telescope

Comments Filter:

  • Windows 11? (Score:5, Funny)

    by Pascoea ( 968200 ) on Thursday June 24, 2021 @01:36PM (#61517048)
    Did someone accidentally click the "update now" button?
    • What do you mean? Windows automatically updates itself.
    • Noooo, silly! This is a much older version of Windows.
      Back before the current MS idiot leaders screwed things up!

      (Remember: the current MS leaders are from a culture that stems from thinking it is religious to bath, shit, piss, and drink the same water.)

  • It is not attached to the space station? I know it was made earlier and hard to move it in orbit, but don't they still have the instructions how to make another one at NASA? Just make another one!

    • I would guess Hubble occupies a higher orbit (547 km) than the space station (400km) because it needs to be clear of other satellites. It would ruin many readings to have another satellite suddenly block the image. The space station orbit also has to be optimized for resupply.

      • Re:Any reason why... (Score:5, Informative)

        by habig ( 12787 ) on Thursday June 24, 2021 @02:08PM (#61517216) Homepage
        The ISS is in such a low orbit that they routinely have to fire boosters to make up for the atmospheric drag which would otherwise degrade the orbit. Given that there's always supply ships docking, this isn't so much of a problem, and being lower makes it that much easier to reach with all that traffic. The HST is in a high enough orbit to minimize the drag problem, but had to be comparatively low enough to be reachable by the shuttle missions that did maintenance. JWST (HST's replacement) is going to be way the heck out there at the L2 point. Once they decided that service wasn't a constraint for JWST, might as well put it out where the observing is best.

    • Re:Any reason why... (Score:5, Informative)

      by necro81 ( 917438 ) on Thursday June 24, 2021 @01:58PM (#61517176) Journal

      It is not attached to the space station?

      One very good reason why is that they are in totally different orbits. Hubble is at about 540 km altitude, 28.5 deg inclination. The ISS is at 420 km and 51.6 deg. It would take an extraordinary amount of rocket propulsion to change Hubble's orbit to match the ISS' - far more than Hubble presently is capable of.

      Another good reason it is not attached is that, for 99.9% of its life, it is best if Hubble is completely untouched. (The other 0.1% is service and maintenance.) This provides the most stable viewing platform, immune to the many vibrations that would come from the ISS. The ISS maintains a consistent attitude w.r.t. its orbital direction (i.e., "front" is always pointed in the direction of forward movement) and the Earth's surface ("down" is always pointing towards Earth), while its solar panels are always slewing around to match the sun. Hubble is always trying to maintain a fixed orientation relative to the stars (or whatever it is imaging, like a planet). The mismatch in motions means that the Hubble would either 1) not be able to maintain its pointing or 2) would collide with the station once per orbit.

      • Re: (Score:2)

        by Strider- ( 39683 )

        Hubble has no propulsion of its own. It controls its attitude (direction it's pointed) with gyros/mass wheels, and bleeds the excess energy off through magneto torquers. There is far too much risk of it fouling its optics if it had its own thrusters. When the shuttle serviced it, they had to approach it from very specific directions, and also the door was closed and sealed to protect the telescope from the residue from the shuttle's RCS.

    • Re:Any reason why... (Score:5, Informative)

      by hackertourist ( 2202674 ) on Thursday June 24, 2021 @02:22PM (#61517272)

      The main reason is vibration. Any time one of the astronauts in the station moves around, measurable vibrations are generated. Hubble has between 1000 and 1000000 times less vibration than the ISS. It still has some, from thermal expansion and contraction, from repointing the solar arrays and occasionally from thruster firings. Attaching a telescope to the station is pointless.

      Hubble is in a much higher orbit to make sure it stays in orbit without the regular reboosts the ISS needs. At ISS altitude, atmospheric drag is noticeable on the timescale of a few months.

    • Re:Any reason why... (Score:4, Informative)

      by Dan East ( 318230 ) on Thursday June 24, 2021 @03:22PM (#61517442) Journal

      Any reason why... It is not attached to the space station?

      Oh, a long list of reasons.
      - The space station didn't exist when Hubble launched. It came 8 years before the first ISS component was even launched.
      - Hubble's orbit is optimized for its purpose, which is imaging the universe. The ISS's orbit is about half the altitude and intended for hundreds of visits by spacecraft launched from earth.
      - All the various vibrations and motions of the space station (solar panels tracking the sun, people moving around, docking and undocking, etc) would interfere with the stability of the Hubble's images.
      - Politics. ISS is partnered with the Russians. Hubble is not.

  • The Hubble telescope contains a tall hierarchy of fail-safe and recovery mechanisms. I'm certain one of them is a direct hardware connection to the radio that permits extremely low-level interfaces to be reconfigured and tested.

    • I bet NASA is kicking themselves by not getting AAA service when they launched Hubble. Sure it would have been expensive over the last 30 years but it would have been helpful now. Of course the problem now is that Paul would have be floating next to Hubble with his card when the tow truck arrives.
  • If there are redundant spare parts for each potential failure, why not swap out each one (and back) until it works?

    • Re:Fake it until you make it (Score:5, Informative)

      by necro81 ( 917438 ) on Thursday June 24, 2021 @02:02PM (#61517192) Journal
      As the article pointed out: the approach they are taking is lower level diagnostics to determine the problem first, then switch to the backup for just that component. Their motivation is to not change anything that is presently working, for fear of causing new problems, allowing problems to spread to not-yet-damaged modules, or just making things worse and harder to diagnose.
      • Change one thing and observe the output. It may give a hint as to the problem. Take a guess.

        • Re: (Score:2)

          by Pascoea ( 968200 )

          Take a guess.

          Take a guess is possibly OK when you're sitting next to the machine. Not so much when you guess wrong and completely brick Hubble. Hell, one of my few formal reprimands came from troubleshooting a prototype [thing]*. System wasn't working, so I started swapping out parts until it worked. Unfortunately, I managed to fry 3 of the 4 modules I was swapping around because there was a short somewhere. The engineer was PISSED, and rightly so. A methodical analysis would have saved him the trouble of repairin

          • And I troubleshot and repaired massive, complex Postal Service sorting machines festooned with computers and various machine control networks. They were also connected by network locally to back office image processing computers and nationally to various sections of the Postal Service. We were tasked with solving complex problems in real time. Sometimes you just have to alter the problem in some way to get a handle on the problem by getting new, different data. And a failed power supply is usually an obvio

        • Alrighty, what they don't know is that the problem is a shorted power supply that's putting out 48V instead of 5V. So when they switch in the backup image sensor, it goes up in smoke and the whole thing is now trash.

          Sometimes it's good to think before you act.

          • If it were putting out 48v instead of 5v the whole thing would be trash right now and unsavable whatever you did.
      • That's fair, thanks for the answer, I wasnt assuming that NASA couldn't think of basic bruteforce analysis

    • Re: (Score:2)

      by TWX ( 665546 )

      Since there are well-supported rumors that Hubble is a derivative of a common American spy satellite platform [wikipedia.org], it would probably make more sense to have built more Hubble-platform satellite bodies and optics (of course without involving Perkin-Elmer this time) and been ready to construct a proper electronics package with whatever is considered modern space-electronics hardware at the time that there's a need to launch a replacement.

      • Re: (Score:2)

        by cusco ( 717999 )

        NASA has two already Hubble-class telescopes already, former Keyhole satellites from the National Reconnaissance Office. The almost-unknown NRO gets so much money that these two satellites sat unused in a nitrogen-filled warehouse in upstate New York for a decade, apparently as spares. When they were declared obsolete they were dumped on NASA in 2012, who had to struggle to scrape up funding to receive them. The WFIRST infrared sky survey was declared the very top priority for the 2020s of astronomers in

  • and pay the subscription fee.

  • I wonder if they tried (Score:3)

    by RitchCraft ( 6454710 ) on Thursday June 24, 2021 @01:48PM (#61517122)
    FDISK /MBR
  • ...if we had a reusable launch vehicle--a space-plane perhaps--that could shuttle parts and astronauts to and from the telescope safely, thereby letting them do repairs? An extraordinary idea, mind you, but I don't think it's terribly farfetched.

    • Re:Wouldn't it be neat... (Score:5, Informative)

      by TWX ( 665546 ) on Thursday June 24, 2021 @02:15PM (#61517240)

      Probably more cost-effective to launch X37B with tools/parts and launch crew for a spacewalk from a capsule-based vehicle.

      The problem with the Rockwell Space Shuttle Orbiter was it wasn't as reflyable as the project intended. The amount of refurbishment required prior to a subsequent flight was extreme, and the launch stages themselves were either one-shot or were so expensive to refurbish that there wasn't a lot of cost-savings theree either. Not to mention the demonstrated risks of SRBs on a man-rated craft.

      • Re: (Score:2)

        by cusco ( 717999 )

        Colombia was supposed to be a prototype, replaced after a decade or less with improved technology acquired from flying it. It was never intended to be the only model of space shuttle, much less intended to be flown until it failed. Nixon and Ford took NASA's design for a space shuttle and gave it to Congress and the Pentagon to play with, what came back was something totally different, with a fraction of the budget and directions that certain parts had to be constructed by certain companies in certain con

    • The ISS has been up there so long I am surprised a small satellite to satellite sized shuttle hasn't been implemented yet.

      • Inclination changes are more expensive than they would seem at first glance. The IIS is on a 51.6 degree incline so the Russians can get at it. Hubble is at 28.5 degrees because that's easier to get to from Florida/Cape Canaveral.

        If you were to put the IIS at a higher orbit, same as Hubble, the inclination change alone would be about 3,000 meters of delta V or change in velocity.

        For comparison to get INTO orbit you need to get going about 8,000 meters per second. The Shuttle, which got Hubble up there, o

        • Re: (Score:1)

          by bardrt ( 1831426 )

          If you were to put the IIS at a higher orbit...

          I'd argue against putting IIS into orbit at all, given how often iisreset decides to throw a "Restart attempt failed. The service cannot accept control messages at this time."

      • Re: (Score:2)

        by cusco ( 717999 )

        NASA has proposed satellite servicing and refueling craft every few years since about 1973, Congress refuses to fund development.

    • ...if we had a reusable launch vehicle--a space-plane perhaps--that could shuttle parts and astronaut

      Yeah, except when we tried that it cost more to launch the manned space plane than to just make another copy of the Hubble and launch it.

      to and from the telescope safely

      "Safely". LOL

      Notice that the NRO hasn't monkeyed around trying to fix their spy satellites, which are essentially the same design as the Hubble. They simply launch more as needed.

      • Re: (Score:2)

        by cusco ( 717999 )

        You can do that when your funding is essentially unlimited. Unfortunately by 2023 the Space Farce alone will have a larger budget than all of NASA.

  • Can we just take a moment... (Score:3)

    by TWX ( 665546 ) on Thursday June 24, 2021 @02:01PM (#61517190)

    ...to appreciate that the guy speaking for the team working on the computer systems for the Hubble Space Telescope is named Hertz?

  • Why wouldn't they launch a new telescope? I mean this one has done it's job and I'm sure is way past planned life.

    • "Why wouldn't they launch a new telescope?"

      $10B USD?

      • Re: (Score:2)

        by dgatwood ( 11270 )

        That seems a little high to me. It cost $4.7B at launch time, including R&D. Subtract $1.2B of R&D, and I'd expect it to be closer to $5B delivered — maybe even less if they still have spare parts.

        Now if you wanted to build a better one, that's a different question. :-)

        • Our favorite 3 letter agency already "gifted" NASA with 2 other Hubbles that they didn't end up needing so presumably less...

          • Re: (Score:1)

            by wfj2fd ( 4643467 )
            They're your favorite? I mean, they have some cool logos and designs, but I thought everyone loved the agency and company more. And it was just the mirrors, not the full systems.

          • Ok, I give: Why is the National Reconnaissance Office (NRO) your favorite 3 letter agency?

        • The $4.7B was in 1990 though, that is close to $10B today. And as with most things built 31 years ago you probably have to start quite a lot of R&D from scratch (suppliers no longer existing, some plans completely gone and so on.

          • Re: (Score:2)

            by dgatwood ( 11270 )

            We're talking about government contracts. You can safely assume they have long-term contracts to provide the parts. Also, they only need one. :-)

    • Re: (Score:2)

      by tragedy ( 27079 )

      Cost-benefit analysis of a repair, one would hope. They may not go that route, of course.

    • JWST was originally supposed to have launched in 2007 to replace Hubble (well, their missions are slightly different, but it would outperform Hubble in nearly all respects).. That date was important because the Shuttles (Hubble's only means of repair and replenishment) would hit their 25-year use-by date a few years later, and would be retired (certain critical structural parts were only certified to last 25 years). They padded a few years to account for schedule slips.

      Well, it's still slipping. You can

    • Why wouldn't they launch a new telescope? I mean this one has done it's job and I'm sure is way past planned life.

      As a matter of fact, yes , they are soon going to launch a new telescope https://en.wikipedia.org/wiki/... [wikipedia.org]

    • Re: (Score:2)

      by cusco ( 717999 )

      All NASA budgets since founding: $650 billion
      2021 Pentagon budget: $753 billion

      This is why we can't have nice things.

  • Most of Hubble's components have redundant back-ups...

    Are you telling me they only have a single backup? That's certainly not up to Starfleet specifications.

    GILORA: Starfleet code requires a second backup?

    O'BRIEN: In case the first backup fails.

    GILORA: What are the chances that both a primary system and its backup would fail at the same time?

    O'BRIEN: It's very unlikely, but in a crunch I wouldn't like to be caught without a second backup.

    • The computer has 4 memory modules, only 1 of which is in use. NASA > Startfleet.

    • Re: (Score:2)

      by tragedy ( 27079 )

      Star Trek has never been all that good at the logical implications of the technologies they introduce. With replicators and transporters, any Star Trek ship or space station should be able to simply replicate any broken component, teleport it away from where it is, then teleport in a replacement. All in a few seconds. Sure, they could make up excuses for certain components like being made of materials that can't be replicated, or the inverse tachyon field around the component interfering with the teleporter

  • Did they re-use one of those?

  • Did they renew the extender warranty? I'll bet they missed that call . . .

  • Have you tried turning it off and on again?

  • viruses.

  • If all systems have redundant backup, the switch ALL systems to their backup systems to get the device live. Then switch one at a time to the non backup system, when it goes down, you've found it.

    • Turns out the issue was in a power supply, and when you flipped back to the shorted one you blew out all the backup systems. Oops.

  • Have they tried turning it off and on again?

  • NASA has problems with THE Telescope?!

    Or does Slashdot have problems with THE Editor?

Slashdot Top Deals

Math is like love -- a simple idea but it can get complicated. -- R. Drabek

Close