Hacker News Reader

Comment by samwillis | Hacker News Reader

With SpaceX Starship the space industry is going to be very different in 20 years time. Launching new telescopes, even larger ones, will be much easer.

Attempts to refuel will be completely unneeded.

Edit:

This seems to be a controversial opinion (based on down votes). To clarify, my expectation is that if launch costs have come down as much as people suggest, launching a new and improved (potentially significantly larger) space telescope would be a better use of funds than a complicated and risky refuelling mission. There will be many lessons learnt from JWST that by then there will be an appetite for something new.

My hope also is that if the cost to launch has come down so would the development costs of any new satellites, leading to reduced costs for science missions.

View on HN

I suspect you're being excessively dinged for the absolute statement that refueling will be completely unneeded.

It's fair to say that over two decades, additional capabilities will exist.

Given the advances in launch systems, both in situ assembly (large structures w/o complex packaging and deployments), and planned servicing and replaceability could mean that both arguments are true:

- We'll have bigger, more capable, and/or simply more observatories.

- Those themselves will be capable of being serviced and upgraded over their own lifetimes.

Keep in mind that one element of science that's advanced by such projects is engineering systems for space environments themselves. Vacuum, radiation, micrometeorites, station-keeping, gravitational flux and variance, materials handling (cold-welding, lubrcation, moving parts, etc.), thermal management, and design-for-servicing itself, all factor in to considerations.

It seems highly probable to me that any such servicing itself will all but certainly be robotic. At L2, speed-of-light issues are a factor as there is a one-way signal time of five seconds, long enough that direct control of physical manipulators is no longer feasible, or very slow. (Imagine that you reach for a wrench, and ten seconds later, see that your hand has actually extended and contacted it.) And putting humans in such a location, only reachable over many days or weeks, well out of Earth's magnetosphere radiation projection, would be fantastically expensive.

> - We'll have bigger, more capable, and/or simply more observatories.

The LUVOIR is the next generation - https://asd.gsfc.nasa.gov/luvoir/ and https://en.wikipedia.org/wiki/Large_Ultraviolet_Optical_Infr...

The design for it is really impressive.

> LUVOIR Architecture A (LUVOIR-A) features a 15-m diameter primary telescope aperture and four serviceable instruments

JWST has a 6.5 meter mirror.

In 20 years we might be able to spamcraft a bunch of ~3m space telescopes and be able to measure their positions with sufficient precision to enable long baseline interferometry in the IR and optical spectrum. We might never build this large of a single instrument again.

The difference is the light gathering size.

The larger the aperture, the smaller the resolved object that we can look at. For example, looking at the central black hole of M87 which is on the order of "diameter of the solar system" ( https://xkcd.com/2135/ ).

However, the only reason we are able to resolve it is that it is so bright.

But if you want to look at dim objects, then the area of the light gathering is the important value. That is why Arecibo was so important ( https://ui.adsabs.harvard.edu/abs/2005AAS...207.2907D/abstra... )

For the HSA - it used the largest of the radio dishes available - https://science.nrao.edu/facilities/vlba/HSA

The Green Bank Telescope 100m and the Effelsberg 100m telescopes are much more sensitive than the VLA... but the VLA can resolve 0.2 and 0.04 overall.

So yes, we could send up a bunch of 3m space telescopes out to the L2 point - but how many individual miniature instruments are we going to need to make for each, how much propellant will be needed for each, how will you keep the cold side cold (size matters)?

While I'm not a rocket scientist, I'm not sure that those are easy problems to solve... and I feel like they're much harder problems to solve than ones that the JWST and the current line of telescope proposals have.

>the absolute statement that refueling will be completely unneeded.

>It's fair to say that over two decades, additional capabilities will exist.

Capability doesn't create need, in either a technical or economic/resources sense.

Too many people are interpreting "unneeded" as: "a satisfactory amount of fuel for a long-enough lifespan, but more would be better"...

When it really means: "fuel is no longer the most likely controlling/ limiting factor as to what ultimately triggers the end of webb's usable/useful lifespan".

Even if fuel ends up being the actual limiting factor, no ability to extend the (likely) lifespan of ALL the other critical components makes refueling a high-risk, high-cost, low-return venture.

So, I agree that there are likely multiple paths to end-of-mission for JWST.

And that fuel itself, over 20 years, could prove not to be the limiting factor.

There remains the point that you've stated, and are again asserting, an absolute condition, that refueling is completely unneeded. Given that a now + 20 years JWST that is still functional except for fuel might exist, that categorical absolute remains unjustified, regardless of how probable it might be. And again, it's specifically how you're communicating this that seems to be raising objections.

None of us have a perfect view of the future. JWST itself offers only an improved view of the past, after all.

If refueling was already a highly questionable proposition given 5 or 10 years of fuel, having ~20 years of fuel certainly doesn't increase the reward in risk-vs-reward analysis.[0]

In what hypothetical scenario does fuel deplete at a greater than expected rate, in a way that wouldn't also call into question the reliability and expected lifespan of the remaining limiting factors?

Accidental commanded release? A leak? Station-keeping being more demanding than expected? Those would all move risk-v-reward in the wrong direction.

And it doesn't substantially decrease the cost or risks, other than what might be gained by an extended development period, which would need to be weighed against simultaneous advances better applied to and spent on a successor.

So maybe not an absolutism, but an already highly-unlikely proposition who's chances of implementation just decreased dramatically due to the new information at hand.

[0] By ending up with an older/ closer to failure but refueled satellite, and a decreased relative value of remaining potential science/observations, presuming highest value science is done first.

These are good points.

They are not the elements of your initial nor follow-up comments with which I've had issue. And you continue to not address those.

As for what might lead to low-fuel-but-serviceable observatory: reaction-wheel degreadation shy of failure (there are multiple redundant wheels for each axis of rotation), possibly some kind of solar-storm or solar-wind interaction which increases station-keeping requirements.

It's difficult to foresee what previously anticipated scenarios might emerge, for all the obvious reasons. But if I had a $10 billion low-on-fuel observatory on my hands, it was still turning in good science, and my capabilities to provide it with fuel had improved markedly, and there was nothing else obviously imminantly limiting lifetime ... I might opt for the top-up. Even at a cost of a few sheckles.

Keep in mind that the opportunity cost isn't the amortised value of the observatory, but its projected future life and the cost of replacing that functionality. I don't know what the associated values here are. I could envision a reality in which those turned favourable for a refueling mission.

Which, again, contradicts your original, and several-times-sustained, "completely unneeded" assertion.

The horse seems quiet well flogged, and in the interests of a peaceful death, I'm begging off further beating of it.

I think we'd be pretty much in agreement if only 5 years of fuel were left.

> Capability doesn't create need, in either a technical or economic/resources sense.

Of course it does! Today we have all kinds of needs that people 20, 100, 1000 years did not have. That's because the capability to do something created the need for it.

For an obvious example, people need iphones. Nobody had one 20 years ago.

For another, deodorant wasn't needed 100 years ago. The need for it was totally created by the ability to manufacture it.

Semantics, I suppose.

Capability allows the market to exist and creates potential for demand.

For certain levels of actual demand and saturation, social/cultural normalcy and expectations might lead to "need" as you describe it.

But in this case, I meant specifically that Webb, given 20 yrs of remaining fuel, won't suddenly "need" refueling if the experimental technology suddenly becomes avaliable in 19 years, as the remaining system also approaches end of life.

Fuel is no longer the most likely point of failure for determining lifespan. That might have been different if Webb had to use more fuel on its trajectory burn, had only a few years of fuel, and the technology was extant/eminent and cheap.

The risk-v-reward calculus would completely different, but unlikely to change at this point.

If my truck throws a rod on a roadtrip in Alaska and I limp upon a gas station as my 20yo dying truck burns the last of its gasoline... I don't "need" gasoline, I need a flight home and a new truck.

I’m not sure if the launch cost was a driving factor in JWST’s price tag. Granted a lot of the engineering was constrained around how to deal with the massive forces involved in launch but I suspect those forces will still be in play with starship.

Yeah you might be able to get bigger and heavier but it still needs to contend with launch forces.

Now assembling it in space might change things… but I’m sure that has its own challenges. And even then you’d probably want a space elevator instead of rockets to get your people and equipment int local orbit.

I think the idea is that if the launch cost is so much lower you can take more risk with the design as you can more easily justify launching again or take a more iterative design approach.

The Arian 5 costs around $185M to launch, they therefore need to make sure it works first time.

Starship is estimated to be $2-10M.

The JWST cost nearly $10 Billion and 24 years to build on earth. This is not a launch cost problem.

How much of that time and money was R&D and prototyping, and how much of that was the bespoke production of this single unit?

There is some previous discussion on what it would cost to manufacture a copy of JWST in this discussion https://news.ycombinator.com/item?id=29725799

Speculation is somewhere in the 70%-90% of original cost.

If you were making the same thing at the same time, it'd be about 30%. That is, previous space missions that have built 2 flight articles have spent about 30% more.

Doing it after the fact--- some design work would be saved, but how much of the tooling and supply chain still exists? Parts are going to need re-engineering, etc.

JWST, like all things in space, has all sorts of complications caused by the launcher.

Size constraints are particularly obvious. For instance, it's 6.5M diameter mirror needed to fold up for launch to fit inside the less than 5.4M diameter fairing. That would fit comfortably without folding in Starships 9M diameter. The sunshield wouldn't quite fit unfolded (14m x 29m), but you can almost guarantee that it would have had a simpler deployment mechanism.

Mass constraints are maybe more important, but less obvious. JWST weighs 6161 kg, Starship is aiming for a payload of more like 100,0000 kg, and by taking advantage of orbital refuelling the capability to put that anywhere instead of it being severely reduced past LEO. That means you can use heavier mirrors instead of optimizing them for weight, you can build your structures out of easy to work with materials light metal, instead of hard to work with ones like carbon fiber. You can oversize parts, and include redundant ones if you think they might be needed.

Even if you totally ignored the cost of the launch, the capabilities of a super heavy launcher would make a huge difference. The cost is the icing on the top. It means that as long as you can build a second version for substantially cheaper than the first, you don't have to design the first so that it never fails. For instance, you don't have to design your heatshield on earth and never test the design in 0G until it's on your 10 billion dollar satellite that has to work...

The question is how much of that cost was involved in getting it to fit the launch constraints? The materials chosen, the unfolding method, etc. On top of all of that, it has to be made so that it absolutely cannot fail.

Imagine instead launching components that could be docked together in orbit, tested, and then boosted to it's destination orbit.

> I think the idea is that if the launch cost is so much lower you can take more risk with the design as you can more easily justify launching again or take a more iterative design approach.

You can also remove large categories of risk by freely spending mass and volume.

> Starship is estimated to be $2-10M.

Don't believe the hype. If Starship works out, they can beat other paths to orbit by a fair margin, but it's not going to be a >10x reduction. Much likely much less than that, too.

The launch costs weren't a driving factor but the launch restrictions made the engineering costs a driving factor.

There should never be such an expensive program again because hopefully we'd realize its easier to also build a massive 1-off rocket for it instead of years developing awesome folding things. Has all the money SpaceX have ever spent so far even come to 10 billion yet?

The JWST can fit in Spaceship’s cargo area unfolded. Although I’m partial to the idea of putting a giant telescope in a crater on the dark side of the moon

I too am a big fan of SpaceX but I think it is fair to say things like "If SpaceX Starship" rather than assume it is going to work and work affordably.

Agreed. Also a fan of SpaceX, but I think NASA's approach is 100% reasonable: quietly prepare plans to take advantage of starship (see: LUVOIR) but wait until Starship is proven to do the high level push to switch.

Additionally, you might prefer to leverage the low launch cost to use a different approach than JWST uses. Basically, Big Spitzer.

Instead of a huge sunshade, use a telescope tube and a huge amount of liquid helium (even the relatively small amount in Spitzer lasted many years, and this is one instance where mass directly translates into time). Instead of big unfolding mirror, use a monolithic one.

This would actually be technically superior to Webb (colder mirror and instruments than you can get with the typical Webb cryocooler, less starlight impacting the instruments as your telescope tube with baffles shields it better, less diffraction off the mirror segment interfaces) and ought to be potentially much simpler and cheaper. And you could afford to build more than one, allowing you to reduce risk aversion slightly and increase telescope time.

That's negating the cost of the telescope itself, though. JWST may be more expensive than something you could launch with e.g. Starship, due to the need to fit into an Ariane 5 payload, but it's still an insanely high-precision instrument that would surely have cost $billions even without the unfolding procedure. If you could refuel it for, say, $250 million, because Starship, why not do that instead of launching a new telescope for $5 billion?

I agree with your point but I think that $5 billion is an overestimate for a telescope as capable as JWST which is much less mass/volume constrained. ELT, a ground based telescope vastly more capable than JWST, is expected to cost €1.3 billion.

I don't think ground-based telescopes are really comparable to space-based. Just one of the issues with JWST is the exotic construction required for heat management, which would be an issue for any space telescope even if it was teleported into orbit. This is comparably not an issue with ground-based stuff where you have infinite power, coolant, and heat sinks; e.g. for the ELT: "ESO is aiming to implement proven technologies and commercial off-the-shelf components to build the cryogenic infrastructure for the ELT instruments. A combination of open loop Liquid Nitrogen cooling and low-vibration mechanical cryo-coolers will be installed to provide the required temperature levels and cooling capacities."[0]

I grew up at Lick Observatory in the Bay Area, and really love ground-based stuff, but I'm somewhat familiar with their operations and I guarantee it's got nothing on space 'scopes.

[0] https://www.eso.org/sci/libraries/SPIE2018/10706-47.pdf

You yourself are making a good argument for why massive reduction in launch costs (like 3 or 4 orders of magnitude compared to Ariane) changes things dramatically. Heat management becomes easier when you can afford liquid cryogens like Spitzer or a larger and cheaper (more ground-like) cryocooler.

You literally could have 1000 tons of liquid nitrogen delivered to your telescope if you needed to, if it saved $1 billion.

I think people still aren’t understanding the ambitions of Starship. We’re not talking about a 30% improvement in cost. But a 300000% improvement in part because you could even afford human technicians…

I'm well aware of the capabilities SpaceX is promising! This is why I said, "even if the telescope was teleported into orbit." If a new ground-based telescope is still $1 bil then something which has to operate in hard vacuum, high radiation, and with even greater temperature differences, with far less provision for repair (even with Starship), will be hard-pressed to come in at an even smaller budget. Then again, $1 bil ain't what it used to be.

On the other hand, you don’t need a massive, expensive structure to keep the mirror from deforming under gravity, everything is already vacuum insulated for you, humidity isn’t a problem, you get 24/7 energy from the Sun using just solar panels (no diesel generators), you don’t have concerns for building on sacred land, you have like 3-4 times the observing time per day (which directly impacts cost per observing hour), independent of weather, access to both hemispheres, and no (or much less) need for expensive adaptive optics and laser systems to partially correct for the atmosphere.

All true! And mostly better for the ultimate purpose of a telescope, which is why we prefer to put them in space. However none of those advantages offset the greater cost, e.g. your "expensive structure to keep the mirror from deforming under gravity" is simply a steel scaffolding. Admittedly to higher precision than an office building, but not extraordinarily so. I think perhaps you are not respecting the challenges of space and the cost to operate reliably in it, even if launch costs were zero and mass unlimited, so I'd encourage you do do a little more research.

> vastly more capable than JWST

... not at the same wavelengths. The earth's atmosphere is largely opaque to IR, for instance.

And even if not opaque, the thermal signature of the atmosphere itself would degrade signal from distant objects.

There are some things you can only do beyond the atmosphere.

Dust also greatly degrades earth based telescopes abilities.

Skyglow, atmospheric distortion, atmospheric absorption of specific frequencies, wind, the Earth's inclination, orbit, and rotation (limits time-on-target for any specific target of interest), local conditions (static electricity being a major issue for the ESO in Chile, given elevation and lack of humidity, all electronics must be static-tolerant or resistant), etc., etc.

Not that space isn't also hard, but each environment has both affordances and limitations.

really though, if Starship is that capable, and JWST is still usable, I think you do both. If only to test really crazy things that you haven't been able to look at before because the time was too valuable.

Lower launch costs could enable cheaper R&D for the satellite itself by enabling more testing of components in orbit rather than trying to nail the calculations ahead of time.

Refuelling satellites is currently a hot topic in the aerospace world. No matter how cheap launch gets (and there are some very real limits to how cheap it can get for orbits like what the JWT is in), the hardware itself is quite expensive and it's often worth maintaining older scientific missions for continuity.

Propellant is often a key limitation to life of a satellite, especially for earth-observations sats in low earth orbit (the opposite of the JWT), and it's well worth refuelling a satellite in principle.

The JWT is a _lot_ of expensive hardware. It's not crazy at all to think about refuelling, especially if it's expending significant propellant to say in a stable orbit +/- maneuver for imaging operations.

All that having been said, refuelling is still a long way from reality, and despite a lot of companies marketing refuelling ports and similar things, it will be a long while time until it's a realistic solution. Over the lifetime of the JWT, it probably will become feasible, though.

Refuelling seems to be more like to be done by bolting another engine + propellant, than actually refuelling the in-service engine, based on current concepts.

Some commercial companies (e.g. Orbit Fab) are selling refuelling ports for satellites, FWIW. NASA put a lot of research into it as well, IIRC. For the JWT it's probably quite different, but for commercial earth observation sats, there are lots of folks considering including some of the refuelling port designs in current hardware even though the actually "refuel in space" service isn't available yet.

Cost is important, certainly, but the available space inside of the payload fairing is going to be a huge boon for future telescopes. From what I recall, a tremendous amount of effort went into folding up the JWST just so it would fit in the small space provided by the biggest rocket they could get in 1997.

yeah, the capabilities of JWST was limited by its size which was limited by payload capacity. Also its cost was driven up by lots of things to deal with the size problems, like all of its folding stuff. We could put up something cheaper and better today if we just had the payload capacity.

I suspect the only "controversy" in your comment is that you said something positive about SpaceX.

I believe the GP is correct: wear and tear on existing components means that fuel won't be the only issue in 20 years. Hubble has been serviced 5 times [1] for example.

Part of the complexity (and thus cost) for JWST was all the issues introduced by trying to fit a 6.5m mirror on existing rockets. All those moving parts and points of failure have design costs.

What you may find in 20 years is that we'll want to launch an even larger mirror and/or use interferometry between multiple telescopes to improve effective resolution.

[1]: https://www.space.com/15892-hubble-space-telescope.html#:~:t....)

It's weird, Musk was the golden child on HN about 2 years ago when he was electrifying cars, now a pariah because of his wealth.

its very hard to build something. its really easy to tear something/someone down

There's two things that go into that IMHO:

1. People like to hate on things and people just because they're popular when there's literally no other reason; and

2. A large number of people will let their personal opinion drive their behaviour. Elon is a dick, no question. Because of that some people will just downvote or speak ill of SpaceX or Tesla. Worse, this dislike can be used to rationalize pretty antisocial behaviour (eg doxxing, swatting).

What SpaceX has done is nothing short of astounding whether you like him or not. I'm not sure of the exact number but I believe SpaceX has raised <$20B and brought launch costs from >$50k/kg (Space Shuttle) to $1-2k/kg to LEO. It seems highly likely that Starship will drive that below $1k/kg, possibly by a lot.

By comparison, a single SLS/Artemis launch is likely to cost more than that.

And if you think it's just a question of throwing money at the problem, just look at the abject failure that is Blue Origin (fun fact: Blue Origin was founded 3-4 years before SpaceX).

I don't see him being treated as a pariah on here. He is being judged by the many things he does which as times goes on will vary more and more. Some will attract praise and some will attract scorn.

I would bet big that most people on here have a fairly balanced opinion of him that roughly would look like:

- Almost single handedly accelerated the electrification of cars by a couple of decades

- Together with many amazing engineers, scientists and more, led to huge reductions in launch costs and increased frequency of launches

- He acts carelessly online with respect to the law around securities

- He acts carelessly online when he wants to insult someone

- He doesn't really care much about light pollution that affects astronomers

- Overhypes many of his ideas and doesn't recognise any criticism of his failures

- Has an implausible idea that we can colonise Mars this century

I am both impressed by the man and have a few misgivings. Like any human being he is multifaceted.

Hyperloop BS was a big turning point for me. That flipped my opinion of him from a smart guy with a lot of resources to do what he wanted to a rich idiot that got lucky hiring great people.

Listen to some in depth interviews with him on technical subjects. He’s a jerk, but certainly not an idiot.

He certainly comes across as a jerk at times but I am wary of thinking we have a good handle on what he is really like at all.

I think the one thing we can be sure of is that he is no idiot. He's very very smart at many things.

But like anyone he does idiotic things at times but I suppose they are amplified by his strange life.

It makes more sense if you consider it's just a cover for developing habitat construction tech.

> It's weird, Musk was the golden child on HN about 2 years ago when he was electrifying cars, now a pariah because of his wealth.

I wouldn't say he's a "pariah because of his wealth." If anything, he's probably a pariah for consistently, wildly over-promising (overshooting "optimist" to land in "liar" territory) and being a jerk (e.g. defaming that the cave diver, among other things).

Then add to that obnoxious fanboys who've made him a weird tech Jesus, and who defend him and bring up his brain farts at every opportunity. I, at least, wouldn't mind never hearing about him again. Lets hear about the cool people who actually did the cool stuff (and just happen to work for companies he's invested in) instead.

Elon has stated that his main goal is to save humanity with technology - so he's made himself tech Jesus.

I think a certain Thai rescue plan and attacking an individual may have something to do with that. Musk is a great entrepreneur, but not a nice person.

The kids were lucky they didn't need his help in the end (less flooding than worst case scenario) and the old white expat wasn't even a rescuer.

Please at least let it reach orbital flight before we start dreaming.

cost to refuel JWST: $500M

cost of a new JWST without a foldable mirror and a static sunshield: $2B

why not do both?

How dare you share your opinion and positive outlook about SpaceX.

I wouldn't worry about the junks, they're fake internet points that mean nothing.

Reply on news.ycombinator.com