The Big Picture: Hubble Repair

By Phil Plait | September 2, 2008 7:35 pm

On October 8, the Shuttle Atlantis will launch into space. This mission is special: it’s the only one left on the schedule not going to the Space Station. Instead, it will rendezvous with the Hubble Space Telescope and perform the final servicing mission on the old girl.

The Big Picture, hosted at the Boston Globe, has a fantastic series of images showing NASA’s preparation for this historic mission. My favorite shot is this one:

practicing the repair of STIS

It shows astronaut Mike Massimino working on my old camera, STIS. On orbit, he’ll have to remove 111 screws to open up the camera, change out the shorted-out circuit board, and then replace all those screws.

I wouldn’t take his job on a bet. On the ground. Not wearing as space suit.

Still, I can’t wait to see this mission lift off. STIS shorted out years ago, but I spent a long time before launch in 1996 getting it ready, and a longer time after it was lofted into space testing and using it. I’d very much like to see it being used again. Not by me; my time is probably past. But STIS is a fine piece of machinery, and a new generation of astronomers should be partaking of its precision and engineering.


Comments (54)

Links to this Post

  1. Science Doesn’t Sleep (9.3.08) | BEYONDbones | September 3, 2008
  1. I believe that job calls for a special kind of screwdriver. You were at the *Con, got one you can lend him?

  2. Overlord

    Man, the RSS reader only sending a couple of lines through now REALLY sucks. Is there no way to tell your new bosses to stop this?

  3. Kaptain K

    IIRC, he doesn’t have to replace the screws. They’ve come up with a snap-on cover that eliminates the need for the screws!

  4. @Overlord:

    I second that.

    I want to be able to read the whole post (along with fancy pics) without having to middle-click the feed in GReader.

  5. third that.. whole post in the Greader. I would go as far as to say i wont click on 1 ad on this page.. or look at anything else on discover.. simply because they force me to come here.

    I’m sure other rss readers will join in.

  6. I’m sure you can’t control this, El Jefe, but I’ve got the same complaint as Overlord and Freiddie. Let us know what can be done.


    P.S. – Dig the blog. You’re my anti-vax woo hero.

  7. Xeno


    The post appears truncated in the RSS reader. Perhaps this is a sly attempt to gain ad revenue? I wouldn’t know, since FireFox has AdBlock Plus and NoScript installed.

    Anyway, I third the annoyance-call.

  8. BP

    Just for the record, these images are not exclusive or anything, and these images and tons more like them can regularly be found by checking the STS-125 media sites. I check them all the time.

  9. Daniel

    I cant wait to see what we get from hubble with this new unit. Hopefully they upgraded on the design.

  10. Rory

    I fourth the complaints on the truncated rss feed!!

  11. Dave Hall

    Man this is going to cost a bundle. . . Twenty five bucks for the part, and several million for the labor and service call.

    Just hope it ain’t a weekend call.

  12. Jeffersonian

    “Not by me; my time is probably past”
    Please explain.

    [btw, this blog format still doesn’t have a left margin (as parsed by Firefox), thus clipping the first letter of each sentence slightly]

  13. madge

    I read a proposal by Sir Patrick Moore that once Hubble comes to the end of it’s life in about ten years time, instead of de-orbitting it we should boost it to a higher more stable orbit, where we can keep track of it and then be able to return to it in the future to recommission and upgrade it as technology allows. That way we won’t lose it. I understand the replacement Webb telescope is a one hit wonder. Once it is inplace it will be too far out for any servicing mission to reach (yet)

  14. CanadianLeigh

    @ Madge,
    I say NASA should take out a real good Sears service agreement on the Webb telescope. How the service tech gets there if needed is their problem.

  15. Allanimal

    Please, return to a full RSS feed.

    111 screws…. I just know that they’ll finish the repair with a couple screws and a few other thingamabobs left over. At least that’s how it works whenever I try to assemble or repair anything. I guess that’s why they aren’t sending me….

  16. Another “full article in RSS feed” suggestion here.

  17. Elissa Fazio

    Have to add another comment about how annoying the truncated RSS feed is – please go back to full version…

  18. Wendy

    I can’t even imagine being an astronaut… I have no idea how they do it. It’s amazing! (And freaky!)

  19. I tenth the truncated RSS feed complaint.

  20. From Phil:
    “The RSS feed has been set to only send out the first few lines of the post. A lot of hosted blogs do this, and the Hive Overmind wanted to test it out.”

    Looks like it’s now safe to tell them that your readers think that, well, it kinda sucks.

  21. Stephen

    I have to say I much prefer screws, that one can see, to snap-on covers where one just has to guess where one has to exert force, how much, and in which direction. Having said that: 111 screws to open the camera? What was the designer thinking?

  22. Yeah, “it kinda sucks” was my general opinion as well. And, as I use it on my blog, and have only just realised how truly annoying it is, I’m going to change back to full feed, as the Hive Overminds should do also :)

  23. madge

    @ CanadianLeigh
    If those service techs are anything like ours The James Webb will wait in all day for them and they will only turn up during the two minutes the telescope has popped into the garden to hang out the laundry :(

  24. Please change the RSS feed back to publishing the full article. I read you blog from an RSS reader, and I will not be clicking through the links to read them. I really disdain when blogs do this. It cuts me off as an audience member.

    Even if I click through some, I will read far less of your blog this way. I love Bad Astronomy and really don’t want that to happen.

    Josh B.

  25. I still think it’s odd that the James Webb is being touted as Hubble’s “replacement”. It’s going to be an infrared scope, which makes it more akin to Spitzer… Not that that’s a bad thing. It’s just not quite what they’re claiming.

    111 screws though! If he dropped them, that’s enough for a small meteor shower!

  26. Sci_Tchr

    The screws are TINY! I can’t imagine manipulating them with space gloves.

    Long live the Hubble!

  27. Todd W.

    Hmm. I have a solution: cordless drill with a reloadable clip.

  28. Ferin

    I’m sure it’s alovely pice of equipment, but why does it need 111 screws?

  29. Curt Winthorp

    I eleventh the sentiment on having the full post go into the RSS feed. If I can read the whole post there, I might get interested enough to click on it and go see comments, related posts, and the all-important Discover ads. If I can only read the first couple of lines… hey, life is short and I’m not going to bother clicking out of my feed reader to read more to see whether I might want to read more or not.

    That’s why I (and many others) have a feed reader, so I don’t have to click all over the Tubes to read what I want. And if the feed is useless… off the list it goes, and rarely do we remember to visit. Phil, please kindly tell your editors to let you be nice to your readers, and Discover will come out ahead.

  30. Brian

    I just want to log my complaint with the truncated blog in GReader as well. Please return to the way it was. Thanks.

  31. Sorry to rain on everyone’s “please restore the full read in the RSS reader” parade, but remember: Discover Magazine needs to pay Phil Plait (among other people).
    If you don’t click through to the blog, Discover doesn’t get traffic.
    And no traffic means no one will want to advertise on Phil’s blog.
    And that means Phil won’t get paid.

    This is, of course, a simplified revenue model, but there’s some truth to it.
    The other thing it does is prevent those scummy spam blogs from stealing content and auto-posting it to steal traffic.

  32. Gary Ansorge

    When I worked for Univac(35 years ago), the company had a dept that was run by field engineers and was dedicated to introducing design engineers to the concept of replacement and repair. For someone who doesn’t have to crawl around on the floor, move cabinets with a forklift, etc, the whole idea of making things easily accessible is an alien idea.

    111 tiny screws??? If those guys had to repair their design under water, wearing big, bulky gloves, I bet they would have come to the conclusion six BIG screws were adequate,,,

    GAry 7

  33. DLC

    I seventeenth the complaint about the truncated RSS feeds. I stopped reading Ars Technicia because they did that, and I would be sad to drop this blog too.

    There is too much good information here to waste behind an intentionaly crippled system.

  34. Ginger Yellow

    “Having said that: 111 screws to open the camera? What was the designer thinking?”

    Presumably the FSM forgot that astronauts don’t have as many noodly appendages.

  35. Invader Xan, I used to think that, but someone came up with a good explanation for why it IS more like Hubble than like Spitzer. Of course I don’t remember the entire argument now. Part of it may be that the science to be explored by JWST is more similar to the science that Hubble has contributed to, but at higher redshifts. Another may be that although it is infrared like Spitzer, which is really just demarcated where our eyes stop being sensitive, it focuses more on the near-IR rather than the mid-IR.

    But I’m just guessing here!

  36. For those wondering why so many screws, a little background:

    This particular cover was never designed to be removed in space. The normal cycle has been to build a new instrument when we want to replace something. The instruments themselves are designed to be replaced: Turn one handle a few times (with a fancy space-capable screwgun), demate four circular connectors (which have nice wing-tabs so they’re easy to rotate), and slide out the old instrument. Reverse the process to put in the new one. That’s how they’re going to remove COSTAR and replace it with COS (Cosmic Origins Spectrograph).

    With STIS, the failure happened late enough in the game that it was too late to build a whole new instrument. Then the question is, “Can we repair it?” It was never meant to be repaired like that, so it took a lot of work to figure out what needed to be done.

    So, why so many screws? The basic answer is this: Rockets are really loud. Obvious, huh? Well, that is, they shake the hell out of things during launch. Which means you need to build stuff strong. And light, of course, because lighter is easier to get to orbit. (Servicing mission 4 will be using about all of the lift capability of Atlantis, which is the most powerful Shuttle.) And for a large cover like that, to make it strong you need lots of fasteners. Which means lots of screws.

    ACS (Advanced Camera for Surveys) is also being repaired. That only needs to have 32 screws removed, but then we take out 4 circuit boards and replace them, and strap on a new power supply. There’s also a metal grid that has to be cut, on that one. ACS only failed a year and a half ago, so it turns out to have been very helpful that the STIS repair had already been in progress. A lot of technology developed for STIS has been adapted for ACS repair.

    Now if the hurricanes will cooperate, this is going to be a pretty amazing mission!

  37. drow

    heck, i’d like to see the RSS feed include the full article, all comments, hyperlinks to relevant pages at NASA and elsewhere, and a built-in javascript-driven D&D 4e character sheet. let me know when Discover sees the light and does all that, will ya?

  38. 01101001

    People (BA included) don’t seem to have read up on the screws and are making assumptions. There are detailed descriptions of EVA plans, but this press conference discusses the STIS screws in simple language,

    “The challenge was the 111 screws that are holding it on, and the screws are not captivated. So they have to go in there and take all these screws out. Well, you can imagine what went through a lot of people’s minds when we first started thinking about this, you know, the 111 screws floating all around inside Hubble. That was unacceptable.

    “So we came up with a very clever device called the fastener capture plate which is basically made out [inaudible]-type material, and this plate goes over top of the MEB cover that is aligned and fastened on there. Then this fastener capture plate has a series of little holes in it that line up little screws. The holes are small enough to allow the tool bit to go in, so you can turn the screw, but they are small enough to keep the screw from falling out. So, once you get all 111 screws taken care of, the cover stays attached to the fastener capture plate and moves the whole thing out. So all the debris and all the screws is captured in there.

    “And then we have come up with an astronaut-friendly replacement cover. Once we are done servicing, they take the new cover, put it on, and there are two latches. They just throw the latches, and Bingo, it is on there, and then there is a third latch that they throw that has some fingers that grab the electronics board and mates them to the cover because this cover also acts like [inaudible]. Okay. So that was challenge number one.”

  39. Johnny Vector:

    So, why so many screws? The basic answer is this:

    Thanks for the not-always-obvious explanation. (Do you speak from experience?) It’s like the other oft-heard complaint about why the really expensive cameras on the spacecraft are so “low quality”, and they can “buy a better camera for $50 at Wal-Mart”.

  40. madge

    If I can find it there is a great video floating around (pun intended) of an astronaut aboard the ISS taking a wristwatch apart and putting it all back together again. It is awesome! And I agree with Johnny Vector that the cover was never designed to be removed in space. They have had to design a new screwdriver and a catch-all cover to collect all the screws and any metal shards produced which could also cause havoc. NASA’s problem solving skills rock.

  41. madge

    I found it! The actual repair footage comes on around 3:30.


  42. Todd W.

    @Johnny Vector and madge

    Umm…why not use lock washers or lock nuts? I understand the need for redundancy, but wow!

  43. Ken B: Thanks, and yes, I do speak from experience. Of course it’s possible to design a cover light and strong enough to survive without a hundred screws, but then it becomes much more expensive. You do that for every possible place you might have to repair, and very quickly you’ve priced yourself out of the competition for new instruments. So the instrument was broken, but not designed for easy repair. But the choice was repair or leave it busted. And well, the Hubble development team is not the type to shy from a challenge, so of course the decision was to fix it.

    ACS, failing much closer to the planned repair mission, was a harder decision yet. There was a long period of “Well go ahead and see how far you get, and if it looks like you’ll get there in time we’ll make it official.” But hey, no problem, just open up a can of can-do attitude (and some really smart engineers), and again, good to go.

    I hadn’t heard the complaint about the cameras being lower quality than the $50 Wal-Mart ones. Hmm, this may call for some outreach work. I guess the Wal-Mart cameras are better quality, if by “quality” you mean “number of advertised pixels”. If on the other hand, you define quality based on video noise, sensitivity, actual angular resolution, quantum efficiency, crosstalk, etc., that’s a different story. After the trouble I’ve had getting decent shots at rock concerts (with a camera I chose based on its low light capability), I cry inside a little every time I take a test image with the ACS spare detector. In a sealed box, taped everywhere, with two layers of black cloth over the top, it still picks up light leaking in. Now that would be an awesome camera to take to the next Carbon Leaf show!

    Course the whole keeping it at -80 degrees C. is a bit of a difficulty. As is the 10cm x 10cm CCD, and the 100 seconds it takes to read it out. But man, talk about image quality! Oh yeah, plus, it all works in space, with radiation and the vibration of launch, and no air to help cool the parts, and…

    And madge, the screwdriver and fastener capture plate (2 really, one for STIS and one for ACS) are just the beginning. There are more new tools designed for this mission than for any before.

  44. 01101001, why do you assume I haven’t read up on the screws? I know they will be contained in a bag-like device; that’s actually made clear in the photo I linked to.

  45. Todd W:

    Umm…why not use lock washers or lock nuts? I understand the need for redundancy, but wow!

    Ah, good question! Hey, the answer is obvious to me, but I didn’t really explain it, did I? Sorry about that.

    It’s not a question of each fastener failing; even #4-40 screws (yes, English units, what can I say?) are pretty strong as far as holding the cover on. And we do lock them down so they don’t back out during the vibrations of launch. Usually with a locking insert or a dab of glue, rather than with lock washers; lock washers tend to create little metal shards, which can float around and short things out later.

    The reason we need lots of fasteners is that we need to tie down the cover at lots of points along the edge. Box covers are (generally) relatively thin aluminum, so they’re fairly flexible. If we just put in screws at the corners, they would tend to flex in the middle of each edge, both during launch and due to temperature changes in orbit. And we generally need the covers to be light-tight, to prevent electromagnetic interference. Also in many cases the heat has to flow through the box covers, which also means they need to be tightly held down. When you do the analysis, in the end you want to put a fastener every couple inches or less. They can be small fasteners, but you need a lot of them. Which makes repair difficult. Especially when you’re in a pressurized space suit!

    I hope that’s a little clearer.

  46. Um, Phil? The FCP is not a bag! And Morey Amsterdam is not a sammich!

    A box. It’s definitely more of a box. The FCP I mean, not Morey Amsterdam.

  47. Howdy. I’m DISCOVER’s Web editor — i.e., vassal of The Hive Overmind. I hear all of you on the RSS difficulties, and we’ll work out a better approach. Thanks for the abundant feedback.

    As for the post, all of this talk of screws makes me inevitably think of Deep Space Homer: “You fool! Now we may never know if ants can be trained to sort tiny screws in space.”

    DISCOVER mag (and Web site) recently covered the save-the-Hubble mission, focusing on Massimino. The last picture in the story’s photo gallery shows a “screw capture tool,” similar to what BA and 01101001 are talking about.

  48. Hi Amos,

    The photo in your gallery is actually the STIS “Fastener Capture Plate”. There is a separate screw capture tool (I think that’s what it’s called, but the name may be slightly different), used to take out four larger (#10-32) screws in the ACS repair. It looks like a screwdriver with a little claw that reaches out to grab the screw as you remove it. Not nearly as photogenic as the FCP, for sure. Good article.

    We briefly considered using ants to sort the screws, but Dan Castellaneta wasn’t available, and Hubble only hires the best!

  49. @Johnny Vector
    I was afraid you might say that — they did sound awfully similar. Are you in an especially good position to know — e.g., work for NASA? You can email me, if you’d like:

  50. @Amos

    Well, you might say that. And you might be right. Let’s just say that once we get the ACS Repair hardware down to the cape next week, I’m going to take a short vacation before the launch.

    And no problem on the specific names of the tools. It’s hard enough keeping the names straight even working with them on a regular basis.

  51. 01101001

    BA: “01101001, why do you assume I haven’t read up on the screws? I know they will be contained in a bag-like device; that’s actually made clear in the photo I linked to.”

    Oh, sorry. When you used “replace” you meant substitute something different for the 111 screws, i.e. the easy new cover with the latches. I thought you meant “put the 111 screws back in place” and I suspect others may have read it that way, too.

  52. Gary Ansorge

    Johnny Vector: Excellent post. I had forgot that the ‘scope wasn’t anticipated to ever have maintenance.
    That explains a lot.


    GAry 7

  53. bill anderson

    what happens if they strip the screw head off during removal? will they then use a drill bit to drill the head off or what


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar