A giant European telescope rises as U.S. rivals await rescue | Hacker News Reader

A giant European telescope rises as U.S. rivals await rescue

science.org

110 pointsVagantem87 comments

Tom Scott did a video on the ELT about a month ago:

* https://www.youtube.com/watch?v=QqRREz0iBes

Most of the video is an explanation of (a) why ground-based telescopes are still useful, and (b) why ganging up multiple telescopes (to create a large 'virtual' area) only goes so far, and then he gets into the ELT in part three of the video.

It's big. The scale is remarkable.

I think its total mirror area is more than that of all existing (largest) professional reflecting optical telescopes combined

39.3m aperture, mirror area scales quadratically with that https://en.wikipedia.org/wiki/Extremely_Large_Telescope

https://en.wikipedia.org/wiki/List_of_largest_optical_reflec...

https://en.wikipedia.org/wiki/List_of_large_optical_telescop...

I love their naming sense.

Very Large Telescope

Extremely Large Telescope

Overwhelmingly Large Telescope

Relevant: https://maz.one/insanely-large-machines.html

next should be called 'Such Telescope' and 'So much Telescope wow'

Definitely A Telescope

Chonky boy Telescope

Telescope AF

Oh lawd he comin

It follows how frequency band called

I think the neatest part is fluid floating, and having to time the light physically because there is STILL too much data to convolve it in real time.

> why ground-based telescopes are still useful

Any layman reading on why optical interferometry hasn't been solved for astronomy, and what it'll take to get there?

Light wavelengths are in the order of nanometers. Radio wavelengths (were interferometry is "solved") are in the order of kilometers to centimeters. While it is possible to build structures that are unmoving in the order of centimeters, it is hard to impossible to do that in the order of nanometers. That doesn't make direct optical interferometry impossible, just quite a bit harder, see e.g. LIGO.

And for doing timing-based interferometry over a distance, the needed timing precision limits your resolution. 1km/c is in the order of 10^-6s (µs), "easy", 1cm/c is in the order of 10^-11s (ps) "possible", 1nm/c is at 10^-18s (as) "quite impossible".

And for the layman, a factor of 10^-6, or 6 orders of magnitude in time are like "a second to two weeks" and "two weeks to far back into the stone age".

Phenomenal explanation. Thank you!

> 1km/c is in the order of 10^-6s (µs), "easy", 1cm/c is in the order of 10^-11s (ps) "possible", 1nm/c is at 10^-18s (as) "quite impossible".

Wow. That's really sad that the orders of magnitude in resolution required escapes beyond our limits of detection.

What do you mean by solved ? optical interferometry exists since the 70s, and is used a lot. The VLT, Keck observatory, CHARA, LBT, NPOI. It's not has big as classical observation because there are a lot of disadvantages, but it exists

Presuming economical access to space is going to scale up a lot in a few years, and taking the many billions of dollars, years of building, and human obstacles to building on the ground -- if it were me with the billions, I'd be thinking about creating a workshop in orbit and then building genuinely big scopes up there, where you're so much less constrained by "the whole thing's gotta fit in a booster, survive not just Earth gravity and weather but a rocket launch, and then deploy and totally work up where nobody can fix it."

These sums of money are so pathetically small. We routinely drop similar sums on seemingly random stuff, none of which ever returns even a small amount of awesome.

I'm sure they will accept more money for the same thing, if you'd like. I live in San Francisco, where locals constantly complained that there wasn't enough funding for essential things. As a result, we funded the things and it turns out that if you want to buy a thing that costs X, the person selling it to you will happily accept 10X if you want to spend that much on it.

I don't think the only possible takeaway should be “whatever the project is, you’ll always get the same result no matter how much or how little you spend”.

That seems fallacious. I think there were other forces at work in the SF allegory, that probably don't apply to all projects.

I don't think that was the proposition.

What do you think it was?

Spending more doesn't guarantee a better result.

You misinterpreted that as spending more never gives a better result.

Ah okay. I didn't see any language where they gave an indication that it was possible to spend more and get a better result.

There wasn't any. But just as I can say "having owned one, I can say buying a BMW is no guarantee of quality", I'm not implying that all BMWs are low quality, and don't need to explicitly say that stop people inferring it. It's not a logical inference.

Which "stuff" are you referring to?

Football stadiums, defense.

Defense is pretty awesome. It’s needed..

Some might say football is pretty awesome and needed. After all, football is life

Social network. See : twitter

One Fusion reactor costs what musk did invest in twitter. -.- Would be way more awesome than people talking shit all day.

IIRC, the cost of mapping human genome was a small fraction of Twitter cost.

War

These giant facilities are in a sense humanity's artificial "exo-sensors", an augmentation of our ability to perceive and understand the Universe. They are costly in comparison to resources accessible to an indvidual but still manageable when done collectively by large countries or country coalitions. Its not clear when and how competition can be a plus in this space. If the core technology is similar/known it may make sense to just pool the widest possible international resources and build the largest possible thing. But there is also something to be said about competing teams taking calculated risks with exploring new directions.

> humanity's artificial "exo-sensors", an augmentation of our ability to perceive and understand the Universe

Indeed. McLuhanesque Extensions of Man.

Tom Scott recently posted a great, 30-minute video about this telescope. Highly recommended.

https://www.youtube.com/watch?v=QqRREz0iBes

Recently there was a video link posted here about the process of aligning the JWST's 18 mirror segments that shed light on the complexity. The task looked challenging at 18, but it goes to mind boggling at 798 segments of the ELT primary mirror. The level of impressiveness is astronomical.

Being on the ground makes the task both easier and harder. It's easier because unlike JWST they don't have just one chance to get it right. If something goes wrong, they can just fix it.

But it's also a more challenging problem to solve because the Earth's gravity deforms the mirrors, and each segment gets deformed differently, and each segment, in turn, gets deformed differently when the mirror is pointed in different directions. So you have to have an extremely accurate model of the deformation as a function of orientation.

> likely to be more than $100 million annually

Oof.

Should the US have just partnered with Europe? Seems like a bigger telescope would have been better than two individual projects (with one struggling significantly).

There might be enough use demand to saturate multiple gigantic telescopes. The US should fund at least one as well.

Technically a bigger telescope has more throughput, right? It can expose the same spot for less time, then move onto the next observation.

Generally there is a lower bound because it takes a few minutes to slew to a new spot on the sky and stabilize. Very large telescopes are slower, so this will take longer, possibly 5--10 minutes. So you typically want to spend at least 20 minutes per target so that most of your time is spent observing rather than slewing. That's why there's still a lot of demand for 1-m class telescopes.

There are also issues with saturation. Large telescopes simply cannot observe brighter targets because they saturate the detectors too quickly. (This is not so much of a problem for deep field objects since none of them have high surface brightness anyway, but can be a problem for brighter stars.)

Only if it were completely automated, or run by the full–time staff instead of by visiting astronomers. Plus, even if a new telescope is 10× better than the old one, nobody will agree to observe for 1/10th the time; it would negate the advantage of the new scope.

It does make me wonder where that sort of money goes. That's roughly $275k per day.

At $1k per person per day you can have 75 staff doing whatever they do (maintainence etc) and still have 200 left over. Computers? One would have thought they'd have on-site processing as part of the build cost. Electricity to run said computers? Food for the cafeteria?

Presumably we can ignore accounting things like depreciation.

Not that it matters I'm sure it's expensive to run, it's just that $275 k per day seems like a lot...

The Tom Scott video linked else where in this thread shows that the existing 4 telescopes have on-prem facilities to resurface the mirrors with a layer of 99% pure aluminum 1000 atoms thick. I'm guessing that quality material fabricated to be installed that precisely is not cheap. The mirrors are moved around on custom built hover crafts. When the 4 scopes are used at the same time, the "plumbing" to get the lights to the instruments at the same time is so precise, it is measured in nanometers.

There is nothing about these facilities that is normal off the shelf components. Everything will have a cost associated that reflects that. What ever back of the napkin math you are attempting doesn't even come close to being accurate.

The person you are replying to is asking about operating costs, not manufacturing costs.

Well operating a highly custom and very expensive is going to have astronomical operating costs as well.

Resurfacing is an operating cost it needs to be done periodically.

I'm going to need some more info on that, because that doesn't sound right to me.

There's no way they are regrinding and resurfacing the mirror periodically - cleaning, sure. But redoing the surface? I'd like to see more info please.

Just go watch the referenced link. It is a very informative video. It's only the top response of this whole thread, but just to make it convenient[0]. I can only lead you to water. I can't make you drink it. Continuing to refute it after that just let's us know exactly what you're about

[0]https://news.ycombinator.com/item?id=38092084

As requested, I drank, and I saw the part about them refinishing the top surface layer. As I suspected it was not them redoing the glass, rather removing the aluminum and redepositing it.

The perfectly shaped glass is the expensive part - coating it is not the expensive part. So despite drinking, I did not get my answer on how this telescope costs $100million a year to operate.

What part of my comment suggests I was talking about manufacturing costs?

This part: "quality material fabricated to be installed"

But they do this operation every few years. It is an ongoing process of installing the aluminum surface. Applying/Re-applying maybe words better fit for your understanding?

It's nice to change your words after the fact, but taking your words, as originally written, your response to a question about operations was a reply about manufacturing.

Additionally it does not costs millions to operate this resurfacing machine. Building it? Sure, I could see that. But running it? No way.

275 k if you average it out over the year, but things like a mirror replacement will require a lot of work, companies, materials and people.

Think also of the data processing, and how much money your employer funnels to AWS.

The ESO observatory in Paranal has its own electricity power plant, both natural gas and - recently - solar.

Computers: multiple TB of data per day will be generated and will need to be archived, backed up, and made easily accessible. Everything on the telescope and the instruments mounted on it is computer-controlled, with countless computer and electronics racks to power and maintain.

Staff: not only salaries and food, they have to run a whole hotel-like facility for people who live on the site for weeks at a time, plus travel costs both from the rest of Chile and from the main HQ in Germany. It's in the middle of the most forbidding desert on Earth, so water and everything else has to be trucked in. Facilities are spread out, so there are dozens of trucks and cheap Fiat cars to go around, with a gas station to fill them up. Cooks, cleaners, engineers, technicians come and go nearly every day.

It all adds up quickly.

I don't remember where I read it, but I'm pretty sure proposals to directly image exoplanets rely on pointing multiple telescopes at them simultaneously. So this might be ok.

The biggest win would be to convince the NRO to flip some of their telescopes 180 degrees

I'm sure the spy sats could take some pretty pictures, but I doubt you can get much science done with their image sensors.

I don’t think these two telescopes are interchangeable. Pretty sure the Webb telescope was designed for seeing specific light spectrums.

ELT is in southern hemisphere, TMT will be in the nothern. Just one cannot see the whole sky. So it's valuable to have both.

> Next month, the first batch of polished mirror segments should set off from France by ship in a temperature-controlled container.

Why do they need a temperature-controlled container for shipping? I would have expected that temperature control would only be needed once they're deployed and in service.

You want to avoid condensation (and other depositions) on the mirror surface. That is most easily avoided if the mirror is slightly warmer than the environment. Of course the container is sealed as well, but there's always some moisture in the air anyway, and it's worthwhile protecting the mirrors from that.

Mirrors deform under temperature changes, and the precision target is measured in tenths of nanometers.

Adjust both for those low orbit satellite plus gravity. But I think it is great to have some around. If nothing else, let us not discovery end day asteroid days before it arrived.

https://en.wikipedia.org/wiki/Extremely_Large_Telescope

It has around 250 times the light gathering area of the Hubble Space Telescope and, according to the ELT's specifications, would provide images 16 times sharper than those from Hubble.

Edit: Wow, this construction image from August:

https://upload.wikimedia.org/wikipedia/commons/thumb/7/75/Al...

Extremely Large Telescope is the budget version.

The original plan was Overwhelmingly Large Telescope https://www.eso.org/sci/facilities/eelt/owl/OWL_design.html

God I still hope that becomes a thing.

Somewhat unlikely: there is a race on between arrays of cheaper telescopes linked by software and ever larger single telescopes and optical bags of tricks to combine the images from multiple smaller telescopes. So an even larger instrument might well be obsolete before it could be constructed and that's before we get into the cost of such a project.

Yeah, and on top of that, making your telescope bigger does not have a linear cost increase. I know it's unlikely, but I can dream!

There are some interesting parametric cost models for telescopes floating around and various papers describing the models (and the data that went into the models).

From that page: “Keeping OWL 3264 segments aligned to a few millionths of a millimeter is a daunting but not impossible task”

And that’s for a 100m diameter main mirror.

"The ESO specialists expect resolutions from OWL that are up to 40 times higher than those of the Hubble Space Telescope. If the 100-meter mirror cannot be financed, a 60-meter variant is being planned. The name OWL would remain the same. Because then the project is jokingly called »Once was larger«."

https://www.itespresso.de/2006/04/17/groesstes-observatorium...

--> Google or other translate

Question from a total noob: How modular are these segmented mirrors? Ie, if the budget runs to 'only' 39m now, is it possible to upgrade it to 42m later by adding segments to what's already there?

I have no special expertise on the subject but by reading and watching about it I think that’s more complex than you might think because there’s a lot a extra systems to make the mirror work with that set up. So it’s not a matter of just bolting in a new ring of hexagonal mirrors and I suspect the majority of the cost comes not from the actual mirrors themselves.

The segments themselves are 100% modular and you could tile an US state with them. The main problem is engineering: the whole thing needs to behave like a single object, turning 360 degrees and moving between horizon and zenith (horizontal and vertical), with a minimum of mechanical flexure. And also resisting wind load. In essence, it's a lightweight building that has to move freely, and this one won't get bigger.

They are modular but not very interchangable (there are 133 different designs of the segments in the primary mirror).

Making a much larger mirror would also face another obstacle: the focal length is designed to be practical for a ~40m mirror.

Despite public perception to the contrary, the Hubble Space Telescope is not an especially large telescope. Telescopes on the ground of a similar size are a dime a dozen. What made HST special is that it was in space and could get much higher resolution than ground based telescopes at the time. With modern advances in adaptive optics that advantage has generally been negated, so a telescope needs to have some other reason to be put in space.

Wasn't it being cold in space so there's more infra red coming in also a factor?

More than temperature, the main problem is the atmosphere. Hubble can see a lot of infrared, and also ultraviolet, because it is in space. Ground-based IR telescopes are somewhat limited, and UV is non-existent.

For ground-based telescopes, I am continually impressed with what amateur astronomers construct in their back gardens, e.g., https://youtu.be/tFZVmVRItbg

Reply on news.ycombinator.com