6 ms·

NASA shares first images of meteor damage on James Webb Space Telescope

38 points by ColinWright 4y ago

_8j50 4y ago
What happens when this thing breaks? Will they abandon a $10B device in space? Or maybe send a guy to go out to L1 and "turn it off and back on again"? Lol
- jvanderbot 4y ago
  when hubble broke they sent someone to fix it. next time probably a robot
- gridder 4y ago
  *L2
- belter 4y ago
  It was a design decision. The JWST was not built to be serviceable in Orbit.
- ben_w 4y ago
  They'd do what they can from the ground, but it's basically not physically repairable if it's damaged, so it would be software fixes only. It's too far away for humans; nothing on it is designed as an attachment point; and body heat on the parts on the dark side of the sun shield, let alone engine exhaust, would risk further damage. A robot might be able to fix something, but that risks being more expensive than building another telescope — I don't know how much of the cost was designing it vs. building it, but a second would probably be cheaper than the first.
  - N19PEDL2 4y ago
    > nothing on it is designed as an attachment point I would be curious to know why they didn't implemented anything like this. I know that no robot, space vehicle or any other technology currently exists that could repair such a spacecraft in L2, but I would like to know whether leaving this door open by adding one or more attachment points would have increased the cost of the telescope too much.
    - nradov 4y ago
      Mass and volume are at a premium on any satellite. Adding attachment points and access panels would have reduced the capacity for scientific instruments. Since NASA has no repair capacity now or even seriously planned for the future, it was pointless to design for serviceability. Building and launching a whole replacement JWST would certainly be cheaper, considering that they have some experience now.
sharikous 4y ago
What's lacking in the article is an expiration date of the sort. When will the performance of the telescope actually degrade noticeably assuming the micrometeor collisions follow the expected distribution? As I understood there was a single impact that was bigger than expected. How does the functional time window change if the expected probability of larger impacts is tuned up a bit?
- belter 4y ago
  More details here.... "Characterization of JWST science performance from commissioning": https://news.ycombinator.com/item?id=32136969 https://news.ycombinator.com/item?id=32136969 See comment: under "Micrometeoroids"
  - koheripbal 4y ago
    The issue here is that not only was the large impact a statistical outlier, but so were the other dozen smaller impacts. So unless the team can optimize the orientation of the craft (which they are attempting) to reduce risk (and reduce observation targets), the telescope will likely not meet its projected lifetime. We're all hoping for the best, but the models of micrometeor density at L2 were likely optimistically inaccurate.
koheripbal 4y ago
This large impact was one outlier. So statistically, you can write it off as (hopefully) a fluke. Unfortunately, there were also over a dozen smaller impacts that, according to NASAs model, would have been very unlikely. Combining these data points clearly demonstrates that the micrometeor density model was wrong - which very sadly means that the projected lifetime of JWST is going to be reduced, maybe very significantly. NASA is currently exploring permanently altering the orientation of JWST to minimize impacts, but this also comes at the cost of reducing viewable space - thus limiting which science studies get telescope time. While NASA's PR department does a great job of telling everyone that everything is fine, and social media does its job of being an echo chamber for such glossy reporting, I hope that HN can take a more realistic view at this, very clearly, bad news. ...maybe even think outside the box about solutions or lessons to learn.
- MichaelZuo 4y ago
  What made them so confident in their assessment of a low micrometer density in the first place? Even as a freshman engineering student I remember that the exposed mirror design seemed oddly overconfident.
  - tekla 4y ago
    What does this even mean? Are you saying as a freshman student you had insight into the statistical modeling of micrometeor impact probability and impact? The design of the optics was more or less finalized around the early 2000s. The exposed optics were kind of the entire point of the entire design. A very low quality post.
    - MichaelZuo 4y ago
      I usually don't reply to obvious trolling but I rather not see any eager freshmen mechanical engineering students be discouraged by a random troll. The basic principles of mechanical engineering have remained mostly unchanged for the last half century. NASA, including the mirror designers, have not discovered any new principles in the process of making JWST. Since even the most advanced technologies derive from these well established principles, at least in mechanical engineering, a sufficiently motivated and inquisitive student can learn enough within a year to formulate reasonable hunches.
vanattab 4y ago
How much would it have cost to send and return a chunk of some kind of airogel to measure micrometers before the mission. You probably don't even need to return it just have cameras send back the data. It seems like the science would have been simple and it becasiclly would have just been an extra launch right?
- yetihehe 4y ago
  Not even cameras, three microphones and a plate would be enough. But spending several million dollars for "measurement of micrometeoroids density at L2" is too much, they just relied that they have valid models.
  - topynate 4y ago
    This is a little cynical, but it's not something they'd want to have known. Suppose the worst case is true and the Webb has a useful lifetime of three or four years. What were they going to do if they'd learned that in 2015? Redesign the entire mission? It was already too expensive. One might say that they should have made the measurement back in the '90s, but the Webb was anticipated to be significantly cheaper back then and the cost of a pathfinder mission would have been proportionately greater.
    - vanattab 4y ago
      Does anyone know if we have ever done direct meteoroid detection at any Lagrange point? With how critical the Lx points are for huge numbers of missions and orbital mechanics we would have thought we would have studied it extensively already.
vanattab 4y ago
Why don't they build two or three copies of telescope when they build them. Most of the money must be in the RD, desgin, and technology development right. I know the mirrors and things are expensive and difficult to make and you still have to test each unit but still.
theandrewbailey 4y ago
Is there a reason that JWST isn't in a tube like most other telescopes? (Would it interfere with the super-low temperature observations?) A tube would shield it from a lot of meteoroids.
- topynate 4y ago
  Saves weight, makes it easier to unfold, fewer potential sources of stray light.

38 points by ColinWright 4y ago

What happens when this thing breaks? Will they abandon a $10B device in space? Or maybe send a guy to go out to L1 and "turn it off and back on again"? Lol

jvanderbot 4y ago
when hubble broke they sent someone to fix it. next time probably a robot
gridder 4y ago
*L2
belter 4y ago
It was a design decision. The JWST was not built to be serviceable in Orbit.
ben_w 4y ago
They'd do what they can from the ground, but it's basically not physically repairable if it's damaged, so it would be software fixes only. It's too far away for humans; nothing on it is designed as an attachment point; and body heat on the parts on the dark side of the sun shield, let alone engine exhaust, would risk further damage. A robot might be able to fix something, but that risks being more expensive than building another telescope — I don't know how much of the cost was designing it vs. building it, but a second would probably be cheaper than the first.
- N19PEDL2 4y ago
  > nothing on it is designed as an attachment point I would be curious to know why they didn't implemented anything like this. I know that no robot, space vehicle or any other technology currently exists that could repair such a spacecraft in L2, but I would like to know whether leaving this door open by adding one or more attachment points would have increased the cost of the telescope too much.
  - nradov 4y ago
    Mass and volume are at a premium on any satellite. Adding attachment points and access panels would have reduced the capacity for scientific instruments. Since NASA has no repair capacity now or even seriously planned for the future, it was pointless to design for serviceability. Building and launching a whole replacement JWST would certainly be cheaper, considering that they have some experience now.

sharikous 4y ago

What's lacking in the article is an expiration date of the sort. When will the performance of the telescope actually degrade noticeably assuming the micrometeor collisions follow the expected distribution? As I understood there was a single impact that was bigger than expected. How does the functional time window change if the expected probability of larger impacts is tuned up a bit?

belter 4y ago
More details here.... "Characterization of JWST science performance from commissioning": https://news.ycombinator.com/item?id=32136969 https://news.ycombinator.com/item?id=32136969 See comment: under "Micrometeoroids"
- koheripbal 4y ago
  The issue here is that not only was the large impact a statistical outlier, but so were the other dozen smaller impacts. So unless the team can optimize the orientation of the craft (which they are attempting) to reduce risk (and reduce observation targets), the telescope will likely not meet its projected lifetime. We're all hoping for the best, but the models of micrometeor density at L2 were likely optimistically inaccurate.

koheripbal 4y ago

This large impact was one outlier. So statistically, you can write it off as (hopefully) a fluke. Unfortunately, there were also over a dozen smaller impacts that, according to NASAs model, would have been very unlikely. Combining these data points clearly demonstrates that the micrometeor density model was wrong - which very sadly means that the projected lifetime of JWST is going to be reduced, maybe very significantly. NASA is currently exploring permanently altering the orientation of JWST to minimize impacts, but this also comes at the cost of reducing viewable space - thus limiting which science studies get telescope time. While NASA's PR department does a great job of telling everyone that everything is fine, and social media does its job of being an echo chamber for such glossy reporting, I hope that HN can take a more realistic view at this, very clearly, bad news. ...maybe even think outside the box about solutions or lessons to learn.

MichaelZuo 4y ago
What made them so confident in their assessment of a low micrometer density in the first place? Even as a freshman engineering student I remember that the exposed mirror design seemed oddly overconfident.
- tekla 4y ago
  What does this even mean? Are you saying as a freshman student you had insight into the statistical modeling of micrometeor impact probability and impact? The design of the optics was more or less finalized around the early 2000s. The exposed optics were kind of the entire point of the entire design. A very low quality post.
  - MichaelZuo 4y ago
    I usually don't reply to obvious trolling but I rather not see any eager freshmen mechanical engineering students be discouraged by a random troll. The basic principles of mechanical engineering have remained mostly unchanged for the last half century. NASA, including the mirror designers, have not discovered any new principles in the process of making JWST. Since even the most advanced technologies derive from these well established principles, at least in mechanical engineering, a sufficiently motivated and inquisitive student can learn enough within a year to formulate reasonable hunches.

vanattab 4y ago

How much would it have cost to send and return a chunk of some kind of airogel to measure micrometers before the mission. You probably don't even need to return it just have cameras send back the data. It seems like the science would have been simple and it becasiclly would have just been an extra launch right?

yetihehe 4y ago
Not even cameras, three microphones and a plate would be enough. But spending several million dollars for "measurement of micrometeoroids density at L2" is too much, they just relied that they have valid models.
- topynate 4y ago
  This is a little cynical, but it's not something they'd want to have known. Suppose the worst case is true and the Webb has a useful lifetime of three or four years. What were they going to do if they'd learned that in 2015? Redesign the entire mission? It was already too expensive. One might say that they should have made the measurement back in the '90s, but the Webb was anticipated to be significantly cheaper back then and the cost of a pathfinder mission would have been proportionately greater.
  - vanattab 4y ago
    Does anyone know if we have ever done direct meteoroid detection at any Lagrange point? With how critical the Lx points are for huge numbers of missions and orbital mechanics we would have thought we would have studied it extensively already.

vanattab 4y ago

Why don't they build two or three copies of telescope when they build them. Most of the money must be in the RD, desgin, and technology development right. I know the mirrors and things are expensive and difficult to make and you still have to test each unit but still.

theandrewbailey 4y ago

Is there a reason that JWST isn't in a tube like most other telescopes? (Would it interfere with the super-low temperature observations?) A tube would shield it from a lot of meteoroids.

topynate 4y ago
Saves weight, makes it easier to unfold, fewer potential sources of stray light.