The Genius Geometry of the James Webb Space Telescope
Vložit
- čas přidán 5. 06. 2024
- Details about the Jane Street Academy of Math and Programming: www.janestreet.com/join-jane-... Apply by 15 March for full-scholarship attendance. See you there!
NASA made a great "Webb’s Diffraction Spikes" infographic.
webbtelescope.org/contents/me...
The differences between Webb and Hubble
www.jwst.nasa.gov/content/abo...
Webb deepfield
www.nasa.gov/image-feature/go...
Hubble deepfield
en.wikipedia.org/wiki/Hubble_...
Here is the WebbPSF python package.
webbpsf.readthedocs.io/en/lat...
LIVE: Biden previews first images from Webb Space Telescope (reveal is at 05:37)
• LIVE: Biden previews f...
Bill Nelson born September 29, 1942. Joe Biden born November 20, 1942. They were both 79 at the time.
The images of space were all produced by NASA and the Space Telescope Science Institute who have kindly made them freely available for the rest of us to use.
Huge thanks to my Patreon supports. That shiny cardboard isn't cheap. / standupmaths
CORRECTIONS
- None yet, let me know if you spot anything!
Filming and editing by Alex Genn-Bash
Written and performed by Matt Parker
Produced by Nicole Jacobus
Music by Howard Carter
Design by Simon Wright and Adam Robinson
MATT PARKER: Stand-up Mathematician
Website: standupmaths.com/
US book: www.penguinrandomhouse.com/bo...
UK book: mathsgear.co.uk/collections/b... - Zábava
"Hexagons are the hexagreatest"
My day has just been ruined.
Here I’ll say it for you “they are the Best-a-gons”
@@merezko4339thanks I needed that
well that phrase is trademarked, so...
I'm joking, of course, but can you imagine?
Don't worry my brother, hexagons are the bestagons. May the hexagon be with you
The second i heard him say that I had a feeling someone would comment on it
7:10 "It's the Hubble, close enough is good enough."
That is a deep cut.
And a deeper field
A Parker Telescope
Deep but out of spec
It's an old reference, about 30 years old, but it checks out!
It's a Parker paraboloid!
CGP gray may be happy...we just won't know until his next upload
Fr
Hexagon is the bestagon
24years from now. Can't wait.
I look forward to finding out in 2027.
No, it’s the hexa-greatest. Matt hath spoken
Kid: Can we get the bestagon hexagons?
Mom: but we have hexagons at home.
The Hexagons at home :The hexagreatest
Lol this was way better then the joke I was going to make about hexagons being the bestagons
Hexagreatest is the Parker bestagon
both bestagon and hexagreatest are absolute cringe, but hexagreatest slitghly less, so you'd be lucky to have them instead of bestagons
they're both cringe, but hexagreatest>bestagon
Hexagreatest is polygawesome!
The bit you seem to have missed between the first JWST image and the Hubble Deep Field is that the Hubble shot is about 22 DAYS of exposure time. The JWST exposure was only about 12.5 hours, yet it is 100 times deeper than Hubble. That's what makes it amazing - it shows just how big a step up JWST is over Hubble. Just wait until you see the 22 day JWST deep field.
How about a 100 day JWST deep field
@@hexagonist23 100 days wouldn't be as good as you might think. You reach diminishing returns due to the fact that the sensitivity only increases with the square root of the increase in exposure. So 100 days is only about 2.1 times better than 22 days, not 4.5 times better as you might think. And imagine all the other science you're missing during those 100 days! No, what we need is a bigger mirror.
@@stargazer7644 Ok, hear me out. What if we build a static observatory on the far side of the moon, with a 50 meter mirror?
Hubble's first Deep Field image that Matt showed at 3:03 was actually shot over 10 days with over 100 hours of exposure time. You are probably talking about Hubble's eXtreme Deep Field image (not shown in the video) that accumulated about 23 days of exposure over the course of a decade.
@@stargazer7644 what about 100,000 days?
Klaus Teuber was the first person to create a model of the James Webb Space Telescope in 1995 when he invented Catan.
I’ll trade you 2 wood for 2 wood.
Does anyone have wood for sheep?
It's unfortunate that they were never able to gather sufficient resources to build his version. The best they could do was 2 cities and a victory card.
how have I not gotten a brick it's literally on a 6 tile
I hate to burst your bubble but many tabletop games used hex maps long before Catan. Avalon Hill’s Gettysburg from the 1960s and GDW’s Traveller come to mind right off the top of my head.
Everyone's talking about CGP Grey's "Hexagons are the bestagons" but nobody remembers Matt's "There is only one true parabola"
Probably because we want to forget the eldritch horror that is the end of that video
GLORIA IN-X-SQUARIS
PARABOLATI CONFIRMED
you mean my ring tone?
And there is only one true paraboloid!
Not true. Hubble Telescope's mirror was shaped after the Parker parabola
What truly blows my mind is that the JWST's mirrors make the Hubble look miniscule. But ... I've stood next to the Hubble telescope - or at least a clone of it - at the National Air & Space Museum a dozen times. It's literally the size of a school bus! The fact that it's now our "small" space telescope is astonishing.
"Please let this be a normal field trip"
"With the Frizz? No way!"
A school bus of 2.4 meters? That's a tiny school bus. Or did you mean the height of a bus?
@@jiminybb he meant both height and length...
@@jiminybb no, I mean the entire telescope. It's 13 meters long and 4.2 meters in diameter -- basically the same as a large school bus
@@daeken Oh my bad! I read it as "I've stood next to the Hubble telescope's mirror"
"This model I've made of the James Webb Telescope is to scale, uhm actually it's roughly π% of the original but that was by accident ..."
-Matt Πarker moment
Classic!
The hεχαgρεατεsτ
Πiracy is a big issue.
0:31 hexagreatest lmao
this video was so funny
2:30 "skipping over Bill's uninspired space ramble"
4:23 "what they do have in common, other than being in space, is they both have a space"
hexabest
Dont forget 15:37 xd
I forgot how absolutely terrible that first image reveal was and yet how hyped I still was for it regardless hahaha
While it may not have been spectacular or new, I think it was specifically chosen because it did have the same field surveyed by Hubble, so that they could truly see how much of a difference it would make.
The White House jumped in at the last second to steal some pr glory, it was kinda weird.
It definitely seemed a lot more exciting in the moment. It's like the feeling when you watch a movie you loved as a kid and realize it's low-quality.
Thank you. You're the first to actually explain how we're seeing what we see, and I've watch hundreds of videos about the JWST. That really brought things in to focus. Those mirrors or reflectors made to scale really shows how different they are.
Brought it into focus, eh? I see what you did there...
I remember @LaunchPadAstronomy explaining this months ago...
@@leogama3422 Well, he explained it in a way that I understand it, most just said it's caused by using all those mirrors.
@suzyturquoiseblue Well, he explained it in a way that I understand it, most just said it's caused by using all those mirrors.
@@broadleyn If one person smiled it was worth it.
I had heard already that the secondary mirror supports were lined up to hide some of the diffraction pattern, but I didn't understand how that would leave two small horizontal lines. I thought it would have to be at an angle or only one line. For one, I didn't know the pattern was not an equilateral triangle and two, the thing that really surprised me, was that the diffraction lines were orthogonal to the supports. It blew my mind that the vertical support would cause a horizontal line. Looking closely at some JWST pictures, I can now see how the other four lines of the supports lay on top of the 6 larger lines - they are just a tad bit brighter than the two other large lines without the support diffraction on top.
I want to see a follow up video where Matt calculates the correct length of the secondary mirror.
The focal length is quite long. The distance to the secondary mirror is about 10 times the diameter of the primary mirror.
@@gordonrichardson2972 No, that would be more than 60m. I don't know any numbers by heart but I doubt the secondary mirror is more than 10m from the primary mirror
Remember that the distance between primary and secondary mirror is not the same as the focal length
@@gordonrichardson2972 I looked it up, its focal length is 131.4m and the secondary mirror is 7.2m from the primary mirror
This man has done many absurd & tedious maths antics, but all of a sudden won’t look up or calculate a simple support beam length?
Where does he have the real Matt tied up? Lol🕵️
@@Astromath I looked it up, but only found the focal length, and misjudged the rest of the answer.
why didnt they just hide all of the supports in 3 different corners of the hexagons? they could have done steps of 1/3 around from the first cornet they picked, and would have ensured that all 3 difraction directions have the same amount of distortion rather than having 2 lines be added on by the support plus the extra bar
I'm not certain, but I think it was a constraint from the way the mirror folded up.
i was also wondering that. My theory is, that it has to do something with the logistics and the way it was been set up in space. Maybe this architecture was way easier to set up and to transport that it outweighed the inconvenience of the additional diffraction pattern. But again, this is just my thought, I have done only very little reasearch in this.
My question exactly after watching the video. But after looking up pictures how the telescope was transported and set up, I'm sure it was the far better option in terms of structural implementation, and they had to make this compromise.
first of all you have to remember that the telescope is so big it had to be folded to be launched, so you also have to deal with the issue of how to unfold it, automatically and without breaking. which takes precedence on where to put the supports. second, every Kg of payload on a vessel you have to pay something (money AND performance), so if you had to use 4 struts instead of 3 just to hide one little diffraction spike, you have to pay for it.
My understanding is that it's just not a big deal as far as the science is concerned. Like in a perfect world they'd have found a solution but with all the other priorities it just didn't really need to be fixed
15:36 This part broke me. We need to know more about Matt's temporal and cloning powers. I'm eagerly waiting for the PCU (Parker Cinematic Universe).
Haha I thought “How rude for Matt to interrupt Matt”
Who do you think runs the "Matt_Parker_2" channel?
It'd be great if there was a term for "large cluster of stars" that started with M and P, so his MCU equivalent could also be the MPMP.
Okay, this really was genius. It's one of those things I never heard about before but I was glad I heard it from you because you explained it so well and in a way that I don't think a person who just does space stuff would have been able to make it so clear and elegant.
hexigons are indeed the hexiest shape
Hexagreatest left me with the same feeling as a sneeze that never happened.
I know it’s a joke, but somebody needs to make sure he sees how his presentation came across. They can’t keep wasting these educational opportunities on someone’s ego. Great job.
It's a good thing you've already established yourself as someone who's not afraid to make mistakes and leave it in. Otherwise I would have had a harder time forgiving you taping those supports the way you did initially.
I think that's the first time I've heard the phrase "pi percent." This channel never ceases to amaze.
Matt, would you have anything against using your model idea for a sciency-crafty lesson? I imagine it would be a great and fairly easy DIY project for kids while also a science conversation starter.
I guarantee he wouldn't just not mind, but be very happy to know you help spread science education like that!
You're in trouble when your physicist wife hears you call the Hubble mirror "supersymmetric" 😄
Additional fact about the JWST secondary mirror ... it's spherical, meaning that each instrument gets a focused view *despite them being offset from each other*, but it's a slightly shifted view due to looking at the secondary mirror from a different position/angle. Think of it like one of those mirrors you see in some London Underground tube stations where the tunnels make a right-angle turn (or possibly have seen opposite some driveways with poor visibility). I was puzzled for a long time as to why the examples of the instruments' field of view denoted them with actually different views, but this is why.
Also, I was hoping Matt was going to talk about the geometry of the sun shades, as they're very precise so as to be as good as possible at ejecting infra-red photons out the sides rather than letting them heat up the subsequent layers and thus the instrument/mirror side. However, I'm not too surprised, because when I heard about this from Smarter Every Day's video about measuring them very precisely ... I recall it being mentioned that the shape and the method used to develop it was proprietary. In essence the surfaces of any two adjacent layers are designed such that any photons bouncing between them will end up going out the edges.
i’m guessing that works similarly to audio absorption spike panels, except the waves move the reverse direction, and it needs to be a more complex shape
The JWST's secondary mirror is a hyperboloid with a conic constant of -1.6598 +/-0.0005, not spherical. DOI 10.1117/12.889150
@@y2kkmac Damn, I don't even know what that means, but I'm impressed.
@@y2kkmac bro just added a citation in a youtube comment. love that.
@@y2kkmac Ah, it seems I was either misinformed, or misremembered that part.
The *important* thing is that the secondary mirror is shaped such that every one of the instruments gets a full view of the image from the primary mirror, just pointing slightly differently for a slightly different view.
When you're expecting "best-a-gon", but then "hexa-greatest" comes in like a punch to the gut
The new JWST image was not higher resolution than the old Hubble image. It was just in a different part of the spectrum. JWST has a much larger mirror than Hubble, but it detects light of much longer wavelengths, so in the end the resolution it achieves ranges from significantly better than Hubble (at the shortest wavelengths, in the red part of the spectrum) to significantly worse than Hubble (at the longest wavelengths, in the mid-infrared).
I often forget that the universe conspires to make resolution a wavelength dependent value instead of a fixed one. The universe is weird.
11:05 first time a proper explanation of this effect truly clicked for me. Thanks!
Seeing pictures of the Tarantula and Carena Nebulæ behind Matt is heartwarming.
To be fair to Bill and Joe they were in their 50s when JWST was started. :)
🤣🤣
They weren't particularly competent then -- and look at them now!
Why didn't they put them evenly spaced like one in every second corner? So they would still be 3 but it would all be more symmetric.
Remember that the primary mirror was folded up during its journey to L2, in order to fit into its spacecraft. The secondary support structure had to be designed to be compatible with that. My guess is that simplicity and reliability during unfurling won out over optical elegance.
@@schafer6811 Oh ok. Sounds good. Thanks
To add to Steve's point, it's fairly trivial to remove the diffraction spikes anyway. You just take two pictures rotated a few degrees and overlay them. Anywhere the pictures agree is "real," anywhere they disagree is an artifact. So, like Matt said, they leave the spikes in for the pretty photos because it's not worth the effort to get two photos, and the spikes contribute to the "pretty photo" mission, anyway.
@@DrakonIL ok
They wanted them to align with 2 sides of the hexagons.
Matt when someone isn't excited enough about science communication: "Fine, I'll do it myself."
I took that same picture of the Cosmic Cliffs you have behind you and blew it up to fit on three 30x40in canvases for my living area... spectacular.
I will reflect on the contents of this video
Sorry Matt, but as an astronomer, I'd have to say it's wrong to think of diffraction spikes as being caused by edges. It's caused by the symmetry of the overall shape. If the telescope had two circular apertures side by side, there would be 2 diffraction spikes even though there are no straight vertical edges. To be more precise, the point spread function of the telescope is a Fourier transform of the aperture.
Is that for the mirror shape? It doesn't seem to explain the support spikes.
@@SimonBuchanNz Shape of the actual clear aperture, which does not include the area obstructed by the second mirror or its supports.
Right, that makes sense when I "unfold" the light path in my head, the whole thing is the aperture!
This was a fun video. I do love how much of the stellar art from my younger years reflects the view from the hubble; 4 point stars were the norm. But now, new media will likely reflect this new diffraction pattern. Lookin forward to our 6+ point stars.
I'm curious about how this might effect sci-fi cinematography... or would it be something with animation/editing?
@@bsharpmajorscale I think we'll see examples in both, though I do expect more traditional artistic renditions to continue as well, depending on who's working on a given project.
I asked about this about 7 months ago in reddit and this explains it way better.
In my astronomy classes we look at the Astronomy Picture of the Day as a warm up to class each day, and I've got 60 high schoolers who absolutely could tell you whether an image was taken by Hubble, JWST, or another observatory with similar diffraction spikes as Hubble. Once you know the diffraction spike patterns the 8 spikes of a JWST image are unmistakable, and a combination of 4 spikes and a ridiculous level of resolution generally mean a Hubble image. Most amateur astronomers whose images feature diffraction spikes wind up having four as well, but you generally can't see hundreds of galaxies in the background of those images.
It wasn't the resolution that made the Webb image better, it was the fact that the Hubble's image was exposed for about a week, while the Webb image was exposed for a few hours. (I forget the exact durations, but it was something like that.)
Hubble's deep field image was exposed for 10 days vs. 12.5 hours for the Webb image
@@RedSunT Thanks! So even if the Webb image was the same as Hubble's, the fact that it required over 20 times less exposure makes it about 20 times better. More time would give more resolution. Matt is a mathematician and should have understood this.
Loved this video, Matt. You explain things so well. I feel like I'm a hexpert now.
beautifully explained, bravo!
0:30 He's gonna say it! 0:32 HE DID WHAT?
you are such a great science communicator! love this
Mate… you have NEVER produced a rubbish video…. But this is very probably THE BEST video about the JWST… well done Sir! 👏👏👏👏🤜
Super interesting none MATH video. Or a Sit-down math video!
Please make more videos about Geometry in Astrophysics & their equipment.... This is much more exciting to watch... Thank you... 🙏
More exciting than what!? I've personally loved the random videos recently
It's not often I have a physical reaction to a joke, but that Hexagreatest line made my stomach flop.
Good job, I guess...
Good to see you get your Bec on!!
Great explanation of the diffraction artifacts on JWST.
Loving everyone who came here to say Hexagons are the Bestagons. My people.
Love the seamless transition at 9:43 from "Fun Side-Fact" Matt back to (I presume) "Boring Main-Point" Matt
Great video, Matt!
Loved this video :)
At 12:40 you could also say they kept the vertical support to simplify the unfolding/deployment of the antenna (: (just adding engineering aspects to the physics aspects of a mathematician explanation)
Love the sass when talking about the heat shield on JWST @ 10:31
Thank you matt, always something I wanted to know about but never looked. So grateful I stumbled across it on my feed.
Instant like on that delivery of "hexagreatest"
simple intuitive explanation. dig it
I just learnt so much
Thank you for all you do; you are amazing
I'm reading a book called "Designed to the Core" written by astrophysicist Hugh Ross in which he describes the incredible advances in our understanding of the universe at the galaxy, galaxy cluster, and super cluster levels. The book is quite amazing.
Ah, geometry and space. The best things.
Wait until you hear about... ...space groups
Vey cool video and subject. I was literally working on my own version of that mirror model when I started this video, it has 65mm sides but I'm also making a smaller one and planning on one like the Sussex folks made.
Hmm I am pretty sure Matt misquoted “Hexagons are Bestagons” on purpose to get more engagement. Well played Matt.
Wouldn't having the supports for the secondary mirror be 120 degrees apart from one another allow it to hide all the refraction caused by them behind the mirrors refractions? I guess the position they took also had to consider the engineering of how it would pack and expand itself when deployed?
Yes it would have hid the diffraction pattern. I think, that you are right. Probably it has something to do with the way it has been transported and set up. But I'm not sure either
I think width when folded, and the direction the mirrors fold, meant they couldn't do that.
Was thinking the same. I was a little bit shocked whe Matt went 30° the other way with the second support, and then straight up with the third.
Love the joy that finding pi in random things brings to him
Er 12:41 shows two supports on the corners and one on a side. Matt's model 14:50 shows one on a corner and two on the sides. Stick to numbers Matt. Love the channel and I hope I win a prize.
This editiing is really good!
Hey we were just talking about this in my class earlier today!
I always heard it as "Hexagon is the Bestagon," and it's demonstrably true.
That's the joke. He subverted our expectation for comic effect.
Isn’t the Hubble mirror a hyperboloid? It being a Ritchey-Chrétien design
Close enough is good enough! I like that line!
Let's all thank Jane Street for making Matt say "math"
The maths is lovely, the model is adequate. Wisdom for the ages 🙂
You didn’t fool me, I can tell apart JWST and Hubble by the lens flare!
“Uninspired space ramble.” I LOLed.
0:30 that is a "parker quote" if I ever heard one xD
Hexagons are The Bestagons 🥰
Love to see the geometry of telescopes being featured here! If anyone reading this wants to know more about the ground-based telescope which won't produce any diffraction spikes that Matt mentioned at the end, check out this video all about it (it's called the Giant Magellan Telescope) that explains how it all works 🔭🔭:
czcams.com/video/orFNSSMII3g/video.html
I like when funny math man does science man stuff
i appreciate the expansion of the mattiverse with side-fun-fact matt, great addition
is there any information if we can expect a great crossover of all mattiverse characters in one video?^^
Diffraction spikes occur in my Newtonian Reflector due to the thin braces that support the 45 degree diagonal mirror within the main tube of the telescope.
Some say, the JWST can look back in time as far as Bill's childhood
excellent, thought i understood it before but didn't know about the 90 degrees thing so thought it was the left and right suports that were causing the spikes.
F*kin love Hexigons
Great video, as always!
This got me thinking, though. Why didn't they use a 4-legged support to completely hide its diffraction inside the ones from the mirrors?
Probably because three-legged support is more stable
@@Ferraro96 I think it was probably more due to lower unfolding complexity the fewer parts there are, and also that a fourth diffraction source adds total diffraction, even if it's overlaid.
Or three equally spaced ones.
more weight added is also not always desirable
Probably decided the extra weight of a 4th leg wasn't worth it when 3 was all they needed for stability.
"Close enough is good enough", somehow apt for Hubble, andthat's what you get for not checking the equipment that you use for checking (so meta and recursive, but....)
No, that's what you get for not testing the final assembly of a 1.5 billion dollar telescope before launching it into space. They did do tests of the mirror that showed the figure was wrong, but they ignored those results because they felt that test wasn't as accurate as the primary test that was setup wrong.
@@stargazer7644 Well they did rely on a single measuring instrument during its construction, and then you can fall prey to a rogue paint fleck throwing off your calibration. Any consequent ignoring of the "slight" error merely compounded their folly. At least JWST can adjust the figure of the mirror on the fly, so lessons were learned (I know, it uses a totally different technology, and that is good thing).
@@TheOwlman JWST doesn't adjust the positioning of its mirrors to avoid what happened to Hubble. They did an all-up full integration test on the ground to avoid what happened to Hubble. JWST is composed of 19 independent mirrors. THAT's why they're adjustable - so they can be configured to act as one mirror.
@@stargazer7644 I realise that, I was more commenting on the change in technology from a single monolithic mirror, relying on a fixed curvature - something that is being applied to terrestrial optical telescopes. Edit: Though that is also part of the modern adaptive optics too.
Amazing!
You mentioned how there's three types in there, if you look super closely at the images, you can actually see that each of the main spikes are split into three
In my fantasy world, SBF watche a stand-up maths video before going to jane street.
Would you see the diffraction lines if you took a photo of a small light source in a dark room through the centre hexagon? Would be good to see if it worked the same way (although the diffraction lines may not be to scale since the supports weren't).
Thanks boss
The outro. Great contrast
Nice video.
Finally, a good explanation.
Actually, the main mirror of the Hubble space telescope is not parabolic (they would be if it was a classical Cassegrain design) but it is a Ritchey-Cretien design (sometimes called Ritchey-Cretien-Cassegrain) which uses a hyperbolic primary mirror.
I can't find many details about the JWST design but all I found was the information it might be a Korsch type design using an elipsoid primary mirror.
The effort he went through saying "hexagreatest" instead of "hexagons are the bestagons" from CGP Gray
That's cool! I knew about diffraction from the supports, but I hadn't considered that from the edges of the mirrors. Duh!
You are the only person likely to follow up on my theory >> the particle density in deep space is apparently 0.12 particles per cubic centimeter; light year is 9.461 e+17 centimeters long; so >> 11.3532 e+16 particles per light year. Photon is 8.418 e-15 diameter. Running the numbers it looks like a Photon will strike a particle at least ONCE EVERY LIGHT YEAR. ((or, its wave will interact with the other's wave if you prefer)) Therefore, galaxies further away are NOT going faster. Rather, there is just more stuff between us causing more Red Shift. The further away a galaxy the more the red shift not because of speed, rather, more stuff.
Your theory explains everything perfectly except for the small problem that light does not redshift when it interacts with matter. Light interacts with far, far more particles going through the air above your head than it does crossing billions of light years of interstellar space. It isn't redshifted at all by that air.
of course you can tell the two deep fields apart. the JWST deep field is looking at a more nearby galaxy group and so it has a bunch of less red-shifted larger objects in the foreground. and of course the stars are six pointed--and both images contain at least a few stars.