Operations begin to de-ice Euclid’s vision | Hacker News Reader

Operations begin to de-ice Euclid’s vision

esa.int

82 pointsN19PEDL240 comments

How do other telescopes avoid/deal with this problem? I don't recall the James Webb telescope having a similar issue, for example.

I would expect James Webb has the same issue, and similar ways of dealing with it. It's just that it never hits the press, since it's a normal part of operations.

I'm not sure why Euclid's icing issue is getting some attention now, from the ESA article this was expected. Maybe it's just that it's slightly worse than anticipated, and they're changing their way of dealing with it?

James Webb underwent a very specific cool down procedure. It included a phase where everything but the instruments were cooled down first, so that water from elsewhere in the spacecraft didn't stick to the instruments.

Moving a bit into speculation territory, I can make a more specific guess. The article for Euclid specifically mentions insulation as the source of the water. It appears that the water in the other components were properly accounted for in Euclid's cool down procedure. For Euclid, insulation directly surrounds the telescope portion. James Webb is build different. There's insulation on the solar panel side, but the giant heat shield is between that and the mirrors and sensors. I don't think there is any insulation on the mirror and sensor side, because besides the sun there's no significant sources of heat where James Webb is.

>It was always expected that water could gradually build up and contaminate Euclid’s vision, as it is very difficult to build and launch a spacecraft from Earth without some of the water in our planet’s atmosphere creeping into it.

If they expect moisture, perhaps final assembly could be done in the Atacama?

Still need to launch from either Guyana or Florida though, so that doesn't solve anything unless you do the integration with the final stage in the Atacama and ship the entire assembled rocket somewhere.

I wonder if this is actually a good application for orbital manufacturing.

Kidding. Kinda…

You could do a Starship* hop from Chile to Guyana, and there transfer it to the ESA's launch vehicle.

*(Meaning the upper stage by itself, once it has lunar lander legs).

Are you saying ESA should have waited for this to be technically possible? Or that it is technically possible now and ESA should have looked in to this option?

It's a speculative, blue-sky idea—as is the context discussion that's proposing to assemble space satellites in a remote desert for optimization reasons.

It's not technically implausible. Starship (upper-stage only) in principle has a very long hop range—it's almost an SSTO by itself [0]. In principle it could also support landing on non-spaceport facilities like deserts—after all it's explicitly designed for landing on the moon, in accordance with the NASA Artemis contract. Would SpaceX be thinking about trying long-range hops on Earth, as a testing mechanism? I'm not sure it'd be completely crazy to consider that.

Musk has extensively talked about the concept of Starship (full stack) as city-to-city passenger transport [1].

[0] https://en.wikipedia.org/wiki/Single-stage-to-orbit#Starship

[1] https://www.theverge.com/2017/9/29/16383048/elon-musk-spacex...

I read the quote and wondered, well, that’s sort of a weird thing to mess up; if it was always expected then why didn’t they take care of it?

It in case anyone else is wondering—it sounds like it is a very thin layer of water

> A few layers of water ice – the width of a strand of DNA

Sounds like they made it about as dry as possible.

Satellites are assembled in clean rooms, which have about 50% relative humidity to avoid electrostatic discharge. The multilayer thermal insulation or MLI recaptures 1% of its own weight after bake-out, within 24 hours in a normal atmosphere. In the vacuum of space at cryogenic temperatures, it can take years and decades for the MLI to dry up. It's a problem for all satellites, but Euclid is particularly sensitive. The Falcon9 fairing was purged with gaseous nitrogen for 4 hours prior to Euclid's launch, to at least minimize the water that could freeze out directly from the atmosphere traveling into space with the satellite.

Sometimes I'm upset that we don't have flying cars or warp drives yet, but then I think of the sheer engineering challenges in something as mundane seeming as this and can be proud again of humanity. I don't know how they ever manage to get these kinds of projects right. The JWST was a pretty crazy feat as well.

> can be proud again of [del]humanity[/del][ins]humidity[/ins]

I fixed that for ya.

Genuinely my first reading

Yeah, those damned science fiction shows make it all look too easy...

You can fix most problems by inverting the field polarity ;)

Enhance!

1.4 billion euros over budget, 12 years in, and they still can't make it work. I am not sure if I am super proud of this.

Science experiments should have some accountability as well.

I cut cost overruns for things like this some slack because I know, irrefutably, that anyone who thinks they can estimate a budget for a one-of-one, never-been-done-before, object like Euclid is a liar.

I know all about the techniques. I've been in the meetings. I've seen the spreadsheets. Your parametric cost model is adorable.

But at the end of the day, you're just lying with math.

And if you have to lie a little bit to get a space telescope built... no sweat.

Is the stated predicted cost really their honest prediction? I thought they always lowball on purpose to make approval easier. Otherwise, I don’t understand why they not just double every prediction by default, that would probably increase accuracy.

1.4 billion appears to be the total cost: https://www.esa.int/ESA_Multimedia/Images/2023/05/Euclid_an_...

Budget back in 2012 was 600 million: https://cerncourier.com/a/euclid-to-link-the-largest-and-sma...

I don't know how badly inflation hits these projects, but that's around 820 million in today's money, which means roughly 580 million over budget. And I've no doubt that 600 million was only floated around in the first place to satisfy a bureaucratic manager who imagines that putting a number next to everything makes things efficient.

Some of this stuff is a gas, and as we know, gases at low temperature or high pressure tend to condense into a liquid. While gases can be compressed into a small space, liquids tend to quite violently exceed the confines of their container.

That's what? Fourteen F-35s? Shared by all of Europe. Or in other words 30 cents per European for each of 6 years. Science is much cheaper than you think. And much cheaper than not doing it.

Is there not a way to control moisture during assembly? If there is, and they did. What amount of moisture are we talking about to cause this issue?

Well, the deposition is estimated as just a few nanometres thick so were talking about something like a grain sand crushed to a fine powder, and then a few grains of that...

Still, I would've thought they'd do a very long vacuum bake-out stage before sealing up the satellite inside the launch rocket.

It's not going to be in a vacuum inside the payload shroud. There's air, and that air contains some amount of moisture. And Florida isn't known for its dry air (it was launched from Cape on a Falcon 9 – but French Guiana would hardly have been better in that respect).

> It was always expected that water could gradually build up and contaminate Euclid’s vision, as it is very difficult to build and launch a spacecraft from Earth without some of the water in our planet’s atmosphere creeping into it.

Satellites are transported in very specialized containers that will often be purged with dry nitrogen. Then after encapsulation in the faring, conditioned air of whatever sort is pumped into the faring from rollout to launch. [Example air conditioning duct attached to a Thor-Agena](https://space.stackexchange.com/questions/39440/why-use-an-a...)

Can ice sublimate at these temperatures?

They did:

> This icing issue is not uncommon, and, as a result, part of the telescope’s commissioning phase included an “outgassing campaign.” This process involved Euclid turning to face the Sun for a total of 96 hours. This process was implemented specifically to reduce the risk of ice build-up after the agency’s Gaia mission faced a similar issue.

Apparently, it wasn't as effective as they hoped.

They should have put one of those silica “do not eat” desiccant packets in there

They say Approximately 10 kg of MLI covers Euclid's two science instruments, but this material can absorb 1% of its own weight in water so at most 100 grams of water.

Assuming they were careful, maybe a tenth of that remains, so around 10 grams of water, a small fraction of which sublimates and eventually meets a mirror and deposits there. They say it's "several layers of molecules".

I suppose they could have used another type of insulation, but since they also have very precise temperature requirements, maybe it was a trade off they had to make.

It's a problem they planned for.

Why don´t they have some strong IR source that can selectively heat the water molecules and zap them into space?

The primary source is this ESA story and it's much better:

https://www.esa.int/Science_Exploration/Space_Science/Euclid...

The current OP is just a summary of this one.

Thanks! We changed it from https://europeanspaceflight.com/esa-euclid-telescope-has-an-....

500 Internal Server Error, so https://archive.is/20240319115956/https://europeanspacefligh...

> 500 Internal Server Error

Firefox gives a SSL_ERROR_NO_CYPHER_OVERLAP (can't vet cert data).

Cert is good. https://www.ssllabs.com/ssltest/analyze.html?d=europeanspace...

My Firefox doesn't complain both on Linux (123) and on Android (latest Nightly)

I tried a few hours later and the page loaded okay.

After some reading I think it has to do with 1) how rigidly Firefox acts in the face of cert chain irregularities & 2) that the site was under heavy load (as indicated by other HN users).

If I'm right, I think Firefox should present more useful error info and have method to bypass - even if method is temp + >1-click.

Why dont they have some IR source that can selectively heat the water molecules and zip them into space?

Reply on news.ycombinator.com