3 ms·

Here is a pic showing the bullet holes in the mirror: https://starhopper.org/2020/11/19/a-telescope-with-bullet-holes/ https://starhopper.org/2020/11/19/a-teles

by mrb 2y ago

Here is a pic showing the bullet holes in the mirror: https://starhopper.org/2020/11/19/a-telescope-with-bullet-holes/ https://starhopper.org/2020/11/19/a-telescope-with-bullet-ho...

Rinzler89 2y ago
How did the mirror not just shatter though?
- dredmorbius 2y ago
  The mirror (third largest in the world at the time) is composed of fused silica. Apparently manufactured in segments, then joined. The cooling process itself took two years. That's described in one of the linked YouTube videos in this thread: <https://news.ycombinator.com/item?id=41459171 https://news.ycombinator.com/item?id=41459171> The result is a glass that's remarkably immune to shattering. Possibly similar to a Prince Rupert's Drop, though that's my own speculation. <https://en.wikipedia.org/wiki/Fused_quartz https://en.wikipedia.org/wiki/Fused_quartz>
  - adolph 2y ago
    Given years long cooling process (aka annealing), the mirror would be the opposite of a Prince Rupert's Drop, which is created by quenching the glass very quickly, similar to the process known as tempering. https://en.wikipedia.org/wiki/Annealing_(glass) https://en.wikipedia.org/wiki/Annealing_(glass) https://en.wikipedia.org/wiki/Tempered_glass https://en.wikipedia.org/wiki/Tempered_glass
    - dredmorbius 2y ago
      "Possibly" is doing some heavy lifting in my comment. The similarity is in the end result, not, obviously, the fabrication process. The strength of the Prince Rupert's Drop comes from the differential tensions and stresses in the head-end of the drop. The tail, of course, is remarkably fragile and will lead to the explosive failure of the entire drop. The head can withstand hammer and even bullet impacts, as Destin Sandlin has demonstrated multiple times on his "Smarter Every Day" YouTube channel. I'd suspect a similar tensile arrangement in fused silica, though of course I may be wrong on that.
      - alwa 2y ago
        Isn’t the original cooling process you’re describing “annealing,” by which glasses are cooled slowly so as to allow internal tensions to equalize? A task requiring exponentially more time as the size of the slab of glass increases? I seem to remember glassmakers using a device called a polariscope to actually see those internal stresses [0]. Which makes me wonder if there are optical considerations as well as physical ones—although I guess if the whole thing is getting silvered, probably not. My understanding, which could be flawed, was that perfectly annealed glass resists cracking because there isn’t a clear fault line to cleave across, something akin to the way hot water in a perfectly smooth glass has trouble boiling if deprived of a nucleation site. While a Rupert’s drop achieves its stability by something more akin to the trope of a “Mexican standoff,” [1], where countervailing tensions are so densely frozen in place that a little extra tension from the outside makes little difference (but when one crack forms, all bets are off and the tensions resolve explosively). I’m also reminded of one of my very favorite moments of video: the silvering pass during the manufacture of the Rubin Observatory’s 23.5-ton, 8.4-meter-diameter mirror; which took from 2008 until 2019 to manufacture: https://m.youtube.com/watch?v=ftKsP3do4nM&t=38 https://m.youtube.com/watch?v=ftKsP3do4nM&t=38 [0] https://www.public.asu.edu/~aomdw/GSI/Glass_Strain.html https://www.public.asu.edu/~aomdw/GSI/Glass_Strain.html [1] https://www.backstage.com/magazine/article/how-to-do-a-mexican-standoff-77522/ https://www.backstage.com/magazine/article/how-to-do-a-mexic...
        dredmorbius 2y ago
        I'm well out of my depth, so, at this point, I don't know.
- klyrs 2y ago
  According to TFA, they're craters, not proper holes. The force of the bullet wasn't enough to puncture the glass, just puverize the points of impact
- hooper 2y ago
  Part of this that hasn't been mentioned is that the mirror is thicker than you might expect. The observatory's website says 12.5 inches (though it will vary somewhat across the curved surface).

by mrb 2y ago

Here is a pic showing the bullet holes in the mirror: https://starhopper.org/2020/11/19/a-telescope-with-bullet-holes/ https://starhopper.org/2020/11/19/a-telescope-with-bullet-ho...

Rinzler89 2y ago

How did the mirror not just shatter though?

dredmorbius 2y ago
The mirror (third largest in the world at the time) is composed of fused silica. Apparently manufactured in segments, then joined. The cooling process itself took two years. That's described in one of the linked YouTube videos in this thread: <https://news.ycombinator.com/item?id=41459171 https://news.ycombinator.com/item?id=41459171> The result is a glass that's remarkably immune to shattering. Possibly similar to a Prince Rupert's Drop, though that's my own speculation. <https://en.wikipedia.org/wiki/Fused_quartz https://en.wikipedia.org/wiki/Fused_quartz>
- adolph 2y ago
  Given years long cooling process (aka annealing), the mirror would be the opposite of a Prince Rupert's Drop, which is created by quenching the glass very quickly, similar to the process known as tempering. https://en.wikipedia.org/wiki/Annealing_(glass) https://en.wikipedia.org/wiki/Annealing_(glass) https://en.wikipedia.org/wiki/Tempered_glass https://en.wikipedia.org/wiki/Tempered_glass
  - dredmorbius 2y ago
    "Possibly" is doing some heavy lifting in my comment. The similarity is in the end result, not, obviously, the fabrication process. The strength of the Prince Rupert's Drop comes from the differential tensions and stresses in the head-end of the drop. The tail, of course, is remarkably fragile and will lead to the explosive failure of the entire drop. The head can withstand hammer and even bullet impacts, as Destin Sandlin has demonstrated multiple times on his "Smarter Every Day" YouTube channel. I'd suspect a similar tensile arrangement in fused silica, though of course I may be wrong on that.
    - alwa 2y ago
      Isn’t the original cooling process you’re describing “annealing,” by which glasses are cooled slowly so as to allow internal tensions to equalize? A task requiring exponentially more time as the size of the slab of glass increases? I seem to remember glassmakers using a device called a polariscope to actually see those internal stresses [0]. Which makes me wonder if there are optical considerations as well as physical ones—although I guess if the whole thing is getting silvered, probably not. My understanding, which could be flawed, was that perfectly annealed glass resists cracking because there isn’t a clear fault line to cleave across, something akin to the way hot water in a perfectly smooth glass has trouble boiling if deprived of a nucleation site. While a Rupert’s drop achieves its stability by something more akin to the trope of a “Mexican standoff,” [1], where countervailing tensions are so densely frozen in place that a little extra tension from the outside makes little difference (but when one crack forms, all bets are off and the tensions resolve explosively). I’m also reminded of one of my very favorite moments of video: the silvering pass during the manufacture of the Rubin Observatory’s 23.5-ton, 8.4-meter-diameter mirror; which took from 2008 until 2019 to manufacture: https://m.youtube.com/watch?v=ftKsP3do4nM&t=38 https://m.youtube.com/watch?v=ftKsP3do4nM&t=38 [0] https://www.public.asu.edu/~aomdw/GSI/Glass_Strain.html https://www.public.asu.edu/~aomdw/GSI/Glass_Strain.html [1] https://www.backstage.com/magazine/article/how-to-do-a-mexican-standoff-77522/ https://www.backstage.com/magazine/article/how-to-do-a-mexic...
      - dredmorbius 2y ago
        I'm well out of my depth, so, at this point, I don't know.
klyrs 2y ago
According to TFA, they're craters, not proper holes. The force of the bullet wasn't enough to puncture the glass, just puverize the points of impact
hooper 2y ago
Part of this that hasn't been mentioned is that the mirror is thicker than you might expect. The observatory's website says 12.5 inches (though it will vary somewhat across the curved surface).