NASA’s bet on Starship for the Moon | Hacker News Reader

NASA’s bet on Starship for the Moon

arstechnica.com

164 pointsalexrustic120 comments

NASA's "bold bet" is also an insurance policy on the fact that SpaceX might have just built this Moon landing Starship version on its own and get back to the moon first. Especially if it got a large private backer, like Yusaku Maezawa is backing the dearMoon project. NASA's relevancy would definitely take a hit if that happened, so funding it themselves is a good idea. I don't see the other two bids getting private funding to go forward without NASA support.

Imagine in an alternate universe where they decided to back an SLS type program over SpaceX, perhaps due to corruption of some sort.

In that case, there's probably better than 10-1 odds that SpaceX would have beat them there for a fraction of the price. They would have looked stupid and been the laughing stock of the country. Future budgets would have been seriously cut.

They made the smart decision here to back SpaceX. Love or hate Elon Musk, his companies get results in an incredible fashion. So far everyone betting against them has got horribly burned. They are excellent examples of American ingenuity and enterprise.

They already look stupid for being part of SLS and having prominent admins say things like landing a rocket isn't that hard and NASA could have done it in the 80's if they'd wanted to, or the Space Shuttle which was a massive downgrade from the Saturn series.

Well, this guy finally considers himself too old to run for re-election again:

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

SLS had a big fan in him, and he used to punch above his weight in space matters.

Going to the moon is not a good value proposition as far as science per dollar. SpaceX and other players only care about supporting science to the extent it makes them money. In my opinion NASA never had the budget for a moon mission anyway. Scrapping space telescopes and other far more useful instruments to pull a publicity stunt would be irresponsible. NASA is still the only party willing to support the less glamorous but more important work. This division of labor is good. Companies can do things that make good marketing material and NASA can continue managing the science that motivates the entire excercise in the first place.

The goal of NASA, or spaceflight at all, has never been to maximize science per dollar. That is a new invented benchmark.

The goal was, and IMO still is, to put ourselves in space.

Science/$ is something of a "new invented benchmark", but I think it's useful to a point.

I fully agree that "put ourselves in space" was the prime driver for NASA up through Apollo, and has certainly played a role since, but NASA as a driver of scientific discovery has become, I think, the main NASA thing.

"Science" isn't even a fungible unit, so you can't even say "science/$". Even a specific type of science, say "Earth Science", isn't even a fungible unit. So it doesn't make much sense to talk about "Science/$".

This can be true when trying to compare science A against science B. However, there is no controversy when comparing a program designed to deliver an instrument package with one designed to build a brand.

Also, there are frameworks for evaluating science output per dollar. This is what funding agencies do when creating and managing grants. There is alot of subjectivity involved which is why DOE program managers, for example, are pulled from practicing experts in the relevant field and tend not to serve very long in those positions before circulating back.

My only reference to something like this is Kerbal Space Program, which uses science as a pseudo-currency to unlock new items etc. I figured it was just a video game thing and did not reflect real life at all.

This is strange to think about. Are these valuations public? Can science be traded between government agencies or other countries? Does science that gets made become devalued over time? Is there like venture capital where VCs deploy science instead of dollars? Has the value stabilized yet such that we can predict how much science we'd get with an arbitrary amount of money or vice versa?

Genuinely interested in learning more since it feels like if we can properly tie "science" to $, we can predict the future by looking at budgets, no?

NASA evolved from NACA, they have always been an aeronautical research group with a focus on practical application. They do way more than just (manned) (space)flight.

It's not about science. Just like the ISS it's about engineering. It's about developing space infrastructure to enable both future science and future space migration/exploration.

> SpaceX and other players only care about supporting science to the extent it makes them money.

In some ways true, but it's hard for me to accept they're a cold, calculating money-grubber despite the common refrain that corporations only exist to make money. SpaceX will gain an absolute massive amount of public goodwill by bringing humans back to another celestial body after 50+ years. That goodwill pays off over the long term in money terms.

A corporation exists to do whatever its owners wish to do with it. It is only true that "corporations only exist to make money" when that's the sole or primary wish of their owners. Musk owns over 50% of SpaceX equity (and over 70% of voting rights), and it is pretty clear his primary motivation is not simply "make as much money as possible". On the contrary, I think his primary motivation with SpaceX is to try to make the science fiction stories he read as a kid come (partially) true, and to try to leave his mark on human history. Making money is just a means to that end, not the end-in-itself.

(The remainder of SpaceX is owned by investment funds, Google, etc, whose motivation is much closer to just "make money".)

> That goodwill pays off over the long term in money terms.

no it doesn't. Goodwill merely keeps the company in the limelight, and if they sell some goods/services, this goodwill is "free" marketing. Good will doesn't pay debts nor obligations.

> it's hard for me to accept they're a cold, calculating money-grubber

better accept it, and the sooner, the better!

> Scrapping space telescopes and other far more useful instruments

A moon telescope is actually a very promising idea.

https://www.nasa.gov/directorates/spacetech/niac/2020_Phase_...

I think if SpaceX only cared about things to the extent it makes them money, they would have picked a simpler business and made a lot more money by now.

I agree with you about science-per-dollar, but I don't think that was ever the criteria for large projects like this (or ISS). Sometimes the payoff is in learning to do the next hard thing, building the next generation of tools, etc.

I think if there's money to be made, there's a business that will pop up to fill that niche.

The missing part here is that there needs to be enough money to be made to offset the risk and startup costs to justify upfront investment.

The last decades have been dominated by tech giants building mostly software, and this has set the bar for risk/expectations when ideas are competing for dollars.

It is not at all obvious to me that a hugely expensive and insanely high risk venture such as putting humans on the moon would magically materialize as a product of a hypothetical free market in today's world.

Moonshots require risks and rewards shared by larger society. It would be impossible, for example, for NASA to capture the complete value of technology developed to go to the moon. However, the body of knowledge filtering out into society at large justifies NASAs continued funding (that and jobs in key states /s).

It would be hard to justify that SpaceX was a rational business venture without hindsight. And many emoyees who work for SpaceX are known to have made economic tradeoffs. There is definitely something beyond "free market" to the story.

Indeed SpaceX is basically controlled by Musk ("Owner Elon Musk Trust (54% equity; 78% voting control") and he's said he's kept it private basically so he can work on making humanity multiplanetary without having to answer to shareholders as to whether that increases shareholder value or not.

You're just ignoring how fucking insanely out of this world rich are SpaceX shareholders going to be in 30 years. Space mining and manufacturing at scale will turn the world on its head, and Musk is the king of scaling technology. The first mover advantage of having thousands of rocketships with a convenient refuel stop and human settlement mid-trip to the asteroid belt is gigantic. Do you think he's really trying to bore tunnels under Las Vegas or wherever? Nah, that boring machine is going to be stuck to the nose of a Starship. SpaceX will be the CHOAM and Spacing Guild of our age.

Yeah, Musk could've made money faster. He also could've simply retired after PayPal. Evidently he's not a person who'd do that.

That depends on how much it costs to go to the Moon.

> All of this has led to criticisms that SLS is a jobs program. Indeed, it provides jobs in all 50 states and supports hundreds of small businesses

They don't call it the "Senate Launch System" for no reason...

Dunno why it’s called the Senate Launch System when Senate Lunch System is right there.

Last I read, it was in only 48 states. Strictly speaking they needed only 26. It will be very hard to kill SLS even with SpaceX able to provide launch capacity at <0.3% of the price.

Betting here they will actually launch one SLS, or even two, but not crewed. The contractors will need to be shifted to one or more other wasteful make-work projects before SLS can be allowed to die. Betting on DoD work.

> It will be very hard to kill SLS even with SpaceX able to provide launch capacity at <0.3% of the price.

I think dearMoon is the thing that's going to kill SLS/Orion. Once private astronauts have safely done an Apollo 8 / Artemis II type mission using Starship all the way there and back, it is going to be hard to say "no" to NASA using it for their astronauts too.

SpaceX will probably even do a private lunar surface mission. I think they'll let NASA go first because they don't want to steal NASA's thunder–they might change their mind about that if NASA is hopelessly delayed, but I think that's unlikely since SLS/Orion are close to finished and the biggest risk to Artemis III timelines is now Starship not SLS/Orion. Once SpaceX does a private mission for significantly less than what NASA is paying, it is going to be very hard for NASA/Congress to justify continuing to pay for SLS/Orion.

> SLS/Orion are close to finished

I'm not well versed in this field so forgive my ignorance but I remember learning about SLS in middle school and being told that it should be operational in 4-5 years. I'm about to graduate college. Do we actually know if SLS is close to finished or are we just taking their word for it?

SLS/Orion are years behind schedule and billions over budget – but they are fully constructed and the testing before the first mission (Artemis I) has basically been completed. The last major pre-flight test milestone was the "green run" of the SLS Core Stage at Stennis Space Center in Mississippi, which was completed successfully last month–after an earlier unsuccessful attempt in January. Right now, the first SLS Core Stage is on a barge on way to Kennedy Space Center, where it will be stacked in the Vehicle Assembly Building with Orion, the ESM (European Service Module), the ICPS upper stage, and the solid rocket boosters (all of which are already at Kennedy). The first flight of SLS+Orion, Artemis I, is scheduled for November – that is an uncrewed test flight in which Orion will orbit the moon and then return to Earth. It is widely expected to slip a bit, but still, if they miss November, it will happen early next year. The next mission planned after that is Artemis II, which will be a crewed Apollo 8-style mission, with astronauts orbiting the moon but not landing on it. Artemis II is scheduled for August 2023. The gap between Artemis I and Artemis II is decent-sized and hopefully gives them time to address any issues that come up, but at the same time don't be surprised if Artemis II slips by a few months too. Finally, the first lunar landing mission, Artemis III, is tentatively targeted to October 2024–but many think 2025 or 2026 are more realistic. Even if SLS/Orion are ready for that, nobody knows whether HLS Starship will be. SpaceX moves a lot faster than the SLS/Orion contractors (Boeing, Aerojet Rocketdyne, Northrop Grumman, Lockheed Martin and Airbus), but Starship is at a much earlier stage of development, and Starship is novel and technically challenging, whereas SLS/Orion is a technologically much more conservative design.

You seem to be saying most of the money for the first SLS has already been spent, so they might as well shoot it.

If there were anything big needing to be put into orbit right away, this one might be able to get it there. But of course as a first test, nothing valuable will be on board. Maybe they could loft a tank of LOX in a solar-powered refrigeration tank, for use by whoever needs it later.

It would be silly to build and launch a second one.

This isn't a "bold bet", it's a backstop. It's what (little) the Senate will pay for right now. Their one choice. It helps ensure that NASA has at least one way to get large amounts of cargo onto the moon. No actual manned flight is included, and personally, I don't expect Starship to be man-rated soon, or - perhaps - ever. It's not needed for the most dangerous task, which is getting astronauts up to orbit, that's already possible via Dragon. Avoiding putting kerosene soot into the stratosphere is the only reason to prefer sending astronauts into orbit via Starship.

> It helps ensure that NASA has at least one way to get large amounts of cargo onto the moon. No actual manned flight is included, and personally, I don't expect Starship to be man-rated soon

That's not true. SpaceX's bid includes at least two lunar landings – at least one uncrewed, and one crewed. (They may have to have more than one uncrewed landing if things unexpectedly go wrong with the first one.)

For the crewed landing, Starship will launch from earth, be refuelled in LEO, then travel to lunar orbit and wait there. Up to 3 months later, Orion will launch on SLS with four astronauts and also travel to lunar orbit. Once in lunar orbit, Orion and Starship will rendezvous and dock, and two astronauts will transfer to Starship (the other two stay on Orion). Starship will land those two astronauts on the moon, and then re-dock with Orion, and the four astronauts will use Orion to return to earth. The Starship will be disposed of, probably by landing it again on the Moon's surface, where it will await possible future reuse by a lunar base. (Ideally it would be reused multiple times before disposal, but I don't think that's going to happen at first – how do you inspect/repair it in-between uses if it can't return to Earth?)

Starship is not being crew-rated for launch and re-entry as part of this contract, although that is the long-term goal. It is only being crew-rated for in-space use and the lunar landing. Orion will be the vehicle used for crewed launch and re-entry for lunar missions initially. Once Starship is crew-rated for launch and re-entry, that may change.

Citations would be welcome. The program includes manned landing, but this contract covers just one part of that program. I looked again, and can't find the contract you say exists.

> It's what (little) the Senate will pay for right now. Their one choice.

This is no correct. The typical NASA move in this situation would have been to extend the analysis phase and pull the competitors along.

Also, the selection statement quite clearly showed that Starship was by far the best solution in every single segment evaluated, technical, price and management.

Had Starship not won the evaluation. NASA would have been required to first engaged the winner in price negotiation and try to square the budget. If this would not have been possible they might have looked to the cheaper competitor or do something else.

However, all of that doesn't matter because Starship simply won outright and they could fit it in the budget.

> No actual manned flight is included

As the other comment stated, this is just flat wrong.

Reminds me of a recent mind blowing article about how bad the SLS really is.(1.) It's good to know Nasa is backing a company that has current expertise, not reusing generation old designs mandated by Congress.

Things like the shuttle era engine design requiring a full tear-down after each launch. And the solid rocket boosters having a 1 in 100 failure rate.

(1.) https://caseyhandmer.wordpress.com/2021/02/24/sls-is-cancell...

Thanks for this. The perfect level of cathartic salt.

Does this mean that the Lunar Gateway project will be scrapped? https://www.nasa.gov/gateway

The current official plan is for Starship to take astronauts from the Gateway to the surface of the moon and back. That plan is ridiculously complex and expensive when you could skip the Gateway and ride Starship all the way to the moon instead. At some point Starship will be ready and SLS and the Gateway will still not be ready and then NASA can decide to bypass the bullshit.

That was my take on this too. NASA is buying themselves a backup plan for further SLS delays.

I can't believe that I'm understanding this correctly? Just to be sure: NASA's plan is to fly the insanely expensive hasn't even launched into space yet SLS, onto some non-existing docking station, then SpaceX fly to the moon. Repeat back (does Starship have to dock again why can't it just go straight back into earth)?

While Starship can go to from earth to moon with a refueling stop alone? Even if sticking to the space dock why not the WAYYYY cheaper Dragon which just worked today (using a recycled rocket)?

I am way out of my comfort zone discussing this, but I really like the idea of building oribiting infrastructure. Not sure why they are not using an earth-moon Lagrange point. I agree that it would probably be cheaper and more practical to go directly to the moon with a simpler refuling scheme. I do however also believe that some infrastructure is needed for further space travel and a larger more permanent off-earth presence.

Starting new missions from outside a strong gravity field seems like such a huge advantage. I guess most people want a destination and not just a more long term established presence.

I imagine that we'd need hundreds or thousands of people in space to make such a base/fuel depot worth it - IIRC it's less fuel to use a transfer orbit than to go to L4 or L1 and stop on the way.

Such a thing might make sense once there are, say, 50 or 100 starships.

For the foreseeable future, all missions will "start" in Earth's gravity well due to the fact that the o2 and ch4 are all down here. It will be some years until we have significant fuel stores in space.

> Starting new missions from outside a strong gravity field seems like such a huge advantage.

You don’t get to do this, period. All your crew and all your materials are ultimately coming from Earth (or, in some plans, are mined in-situ). You use considerably more resources getting to your waypoint than you save by leaving from there.

Being farther out of the gravity well isn’t entirely positive either; you’re losing out on significant savings from the Oberth effect. Earth-Moon Lagrange points are well past the minimum-cost orbit.

> All your crew and all your materials are ultimately coming from Earth

Wait till we start making LOX (and later, I think, metals) on the Moon from regolith. The process is old and tested on Earth, the missing part is actually experimenting with this on the Moon. And as soon as we have more or less robust Earth-Moon transportation, this will become important, as it will significantly simplify near Moon operations.

Do you know NASA runs, right now, competitions for e.g. lunar ice extraction and purifying?

> All your crew and all your materials are ultimately coming from Earth

This is the key state differentiator.

If you bring everything from Earth, it doesn't make sense. (So why are we doing it now?)

But once you start procuring resources from outside gravity well, things look different.

Exactly what do you plan to acquire/create outside of Earth, and from where to where do you plan to transfer it realistically? Except for fully autonomous human colonies which will mine stuff nearby and use, almost no transfer of resources in space makes sense, it is way too costly and there is no demand (even imaginary). Since there is no need for that, there is also no need for transfer stations (warehouses) anywhere, be it a planet orbit or Lagrange points.

Propelant depots, those make all the difference.

They make sense for an individual mission. Bring up fuel and leave it in the parking orbit. But the optimal parking orbits are going to be different for different destinations, vehicles, and times.

It makes sense if you want to use the rockets you already have instead of building the absolutely titanic rockets you would need to carry everything in one trip. There are metallurgical limits to just how big you can make the rockets as well.

But this isn't necessarily advocating for a full time in-orbit refuelling station either. The plan might be to instead launch 2 or 3 rockets, 1 with the stuff and the others with fuel. They meet up in orbit and the fuel rockets refill the stuff rocket so it can go on to Mars or wherever, then they return to Earth for the next mission.

Only some orbits would intersect with L1 point, others will require immense fuel waste to get there and then change orbit to original target. Also there are no gas stations in space, for multiple reasons - there is no steady demand to fuel in space, it is at best sporadic and spiky, and in the mean time you'll need to keep up cryo fuel from boiling off; then it is hard to operate anything remotely, so Gateway gas station would be harder to operate than ISS in LEO; there is no need to place gas station farther than LEO because in space we don't need to bother with actual distances, but with gravity. Earth to Mars is 62 million kilometers, so speaking in analogies it would make sense to refuel at half of the distance, similar to the traveling by car from New York to San Francisco. But in reality it is not. Getting 300 kilometers from Earth to LEO takes 9 km/s of delta V, getting the rest 62000000 kilometers from LEO to Mars takes 5-6 km/s of delta V. LEO is much more than half way to Mars already and there is literally zero point is refuelling anywhere else (excluding fuel generation in space possibility).

> does Starship have to dock again why can't it just go straight back into earth

Because Moon-bound Starship will be quite different from Earth-bound - not enough to not call it Starship, but still - the Moon-bound will be optimized for less gravity (different legs?), no atmosphere (no wings), braking using engines in the middle of the body, not in the tail section, Moon-specific systems like ladder or elevator etc.

AFAIK lunar Starship still has Raptors in the tail that do most of the burns, the "belt" engines are only used for the final landing phase to reduce the ammount of regolith kicked up by the rocket plumes contact with the surface.

Interesting. So would still do a 'dock' and could come back on a different Starship, Dragon, or SLS it sounds like

Separating the "get to/from moon" vehicle from the "land on/start from moon) vehicle has indeed been the plan for many years now.

hasn't even launched into space yet SLS

I'm no fan of SLS, but check your terminology here: You're talking about Starship in the present tense, but it hasn't launched _either_, so you can't say it "can" go to the moon etc either.

Starship is awesome, but it's an insanely ambitious project. No one knows how long it'll take to get it right.

Raptor engines are, yes, cutting edge. Serious stuff.

Starship itself?.. Being built with such technologies, so cheap (Elon mentioned that Boca Chica works are quite small percents of what SpaceX is doing in terms of money), so fast - and yet we already have a successful-ish landing from ~10 km height.

Of course, we don't know for sure. But it surely doesn't seem too bad, especially now - and for some it was clear when SpaceX was just created.

Well, Starship has definitely reached bigger height than SLS so far (unless you count some of the reused Shuttle engines). ;-)

Also considering the planned SLS "landing" mode they are also on par so far.

That's totally 100% fair critique.

But SLS is in no way able to go to moon or earth with what they are currently building?

What makes this so weird to me is that SpaceX does have Dragon already which just brought more people into orbit this week - e.g. proven and way cheaper of what SLS is supposed to do in this mission - unless I'm missing or not understanding something.

So you're right the starship portion to moon (and hopefully mars) is similar to SLS in that it's not realized ye

But looking at the part SLS is supposed to play in this production, that role has already been acted successfully by Dragon. I'd say spacex gets the space oscar for it, though i doubt they get the trophy this weekend :)

Orion does have several spec advantages over Crew Dragon such as larger volume and longer endurance. https://www.quora.com/What-are-the-key-differences-between-t... Given SpaceX's dramatically faster and cheaper development, it might be possible to deliver an upgraded Crew Dragon Plus (or whatever you want to call it) before Orion is ready.

> it might be possible to deliver an upgraded Crew Dragon Plus (or whatever you want to call it) before Orion is ready.

Orion basically is ready. There is an uncrewed full stack test flight around the Moon scheduled for November this year, Artemis I. Decent chance it will slip by a few months, but still highly likely to happen by early next year. And then in 2023 they have planned a crewed test flight (orbiting the Moon but not landing, like Apollo 8).

I think SpaceX wants to focus as much of their development resources as possible on Starship. They'd rather focus their energies on that rather than on coming up with new versions of Crew Dragon. Starship, if successful, is going to kill SLS/Orion in the medium-to-long term. I don't think there is any great benefit to SpaceX in trying to kill it in the short-term. That'd be walking into a potentially expensive and risky political battle with vested interests in Congress, and why get them offside when doing so isn't on your critical path to your ultimate goal?

I thought that was crazy at first but the Orion capsule uses proven reentry tech. SpaceX hasn't landed Starship successfully yet and hasn't proven their reentry strategy, and coming back from the moon will be harder because they'll be coming in faster. I can't blame NASA for not committing their astronauts to it yet.

But yeah, I think SpaceX will prove it all works and then NASA has the option to change the plan. In the meantime, officially keeping SLS keeps Congress happy.

Starship can come back from the moon to LEO then the crew transfers into a Crew Dragon and lands.

Yeah, once you have a large ship out of earths gravity well, would be a waste to land it back on earth. Heck, even if it breaks, park it in a stable orbit so the next generation can salvage it for parts as needed.

PSA: Most orbits around Moon are not long term stable due to underground mass concentrations.

Neat! If you park a satellite at moon geosychronous orbit how unstable is it? Would it be lost in the order of months or decades?

Does that also effect the Lagrange points for the moon? Do human currently have anything parked at earth moon lagrange points?

There appears to be no such thing.

"the radius for a selenostationary orbit is much farther than the Hill radius, meaning that no stable orbit can be achieved as it would be too much perturbed by the Earth and/or the Sun"

"all of the Lagrangian points of the Earth-Moon system are stationary relative to the Moon surface"

...but then it is explained that stationary doesn't mean stable.

Links: https://astronomy.stackexchange.com/questions/20499/is-it-po...

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

It would need lunar or high elliptic refueling to do that. Which is possible, but more complicated.

The Lunar Starship has enough dV to land on the moon and return to orbit, about 3.44 km/s but boosting from the moon to Earth and braking into Low Earth Orbit actually takes more dV than that, about 3.94 km/s. Unless they build in extra 0.5 km/s capacity or about an extra 20%, it flat out won't be able to do it.

Here's where I got the idea and some different calculations: https://caseyhandmer.wordpress.com/2021/03/26/lunar-starship...

The diagrams say "Crew transfer in LEO or direct entry of Starship" as if they're interchangeable options, but these are not equivalent. Returning from the Moon and braking into LEO takes just as much dV as transferring from LEO up to Lunar orbit. Per the diagram that's 3.94 km/s as I said.

Conversely landing on the moon and returning to Lunar orbit takes 1.72 km/s twice, which is again consistent with my numbers if we assume the payload mass is constant. That may not be the case of course but any cargo taken from the Lunar surface to Lunar orbit will also need to be brought back to Earth somehow.

currently the lunar version of Starship is not designed to land back on earth - it lacks both the TPS and aero surfaces required for entry & landing.

The Lunar Starship requires a functioning Starship capable of reentering in order to refuel the Lunar Starship before it heads to the moon (and it will need to eventually travel back to the Earth for refueling after some number of uses).

I don't think the lunar Starship is ever intended for returning to the Earth after it's launched - it has no flaps and no heat shield.

It will either be discarded after use or refueled and reused fully in space.

Lunar Starship won't be (intentionally, excepting emergencies or retirement) returning to Earth's Surface, but the plan might involve it returning closer to Earth for refueling.

Sounds a lot like Gateway is Kubernetes the space version..

The gateway is where international partners are currently hooked in. If NASA is to cancel the Gateway without absolutely destroying it's relationship with international partners, it would need a replacement mission (say Moon surface modules) to make up for it.

> The current official plan is for Starship to take astronauts from the Gateway to the surface of the moon and back.

Even before announcing the choice of Starship for HLS, NASA was already talking about bypassing Gateway for the first crewed lunar landing (Artemis III) and only visiting it on later missions – https://spacenews.com/nasa-takes-gateway-off-the-critical-pa...

They haven't completely decided yet though. They don't know when the bits will be ready. If come time for the first crewed landing, SLS+Orion and Starship are ready but Gateway isn't, they'll just do the landing without Gateway. On the other hand, if Gateway is already ready by then, they might consider using it. One advantage Gateway has, is it gives the two astronauts who stay-behind on Orion a bit more room. It would also give them something to do – testing out Gateway systems.

I don't think the Lunar gateway is so useful for people going to/from the moon, but for future trips to Mars and back. Having a place to stop and refuel would greatly reduce the size of rocket you need to launch from Earth.

The choice of having it in Earth orbit vs. Lunar orbit comes with some tradeoffs. It may make more sense to leave it in Earth orbit, but ultimately having it at all substantially changes the calculus on a Mars trip.

Unless the fuel is being produced on the Moon, it would be more efficient for Mars-bound spaceships to fuel up in LEO. All Earth-departing mass has to pass through LEO, so it's simpler to rendezvous there rather than enter, and then leave, Lunar orbit.

If launches are cheap, then refueling in LEO makes a lot of sense. Historically, this has not been the case at all, but it's all part of SpaceX's model for Starship.

Lunar gravity is actually weak enough to make a space elevator from conventional materials possible. Even Kevlar would suffice.

If fuel for interplanetary missions is produced on the Moon, a space elevator would make a huge sense for refueling of the ships.

Doesn't the moon rotate too slowly to make a space elevator work? IIRC you can't take up a ground-stationary orbit around the Moon because the Earth would constantly knock your satellite around with its big gravity well.

I think this is where the Lagrange points L1 and L2 would help.

You can have a ground stationary satellite if you tether it to the moon. Though obviously the tether has to deal with that extra load.

The best sites for permanent outposts are the poles though, it would be quite a drive from the equator.

A lunar space elevator will never happen. First of all the gravity is unstable.

And even if you do so, the main advantage of an elevator can also be had with an accelerator, there is no need to build an elevator. Because there is no atmosphere you can simply fling stuff into orbit for pretty much exactly the same price but far, far, far, far less investment.

The one place elevator actually makes sense is Mars potentially. But that is a very long way off.

Even lunar space elevator will cost an insane amount of money (and likely still not possible in this century), and it will be irrational to build it just to get some cheap fuel into orbit.

And even then gas station in Lunar orbit is useless. If there is a need to refuel, you will simply get fuel into orbit directly from whatever container in the elevator to the ship nearby. There is zero point in inserting a proxy station in the middle of this process.

However we juggle the possibilities of space travel, in 99% of all schemes gas station in space is useless.

Basically, yes.

Gateway is smaller, doesn't have two airlocks, etc. If you want to build a station in LLO, just build it from a few starship hulls.

Also, the stupid polar orbit around the moon isn't necessary if you have moon starlink (no comm loss on dark side). This saves energy otherwise needed for inclination changes.

> Also, the stupid polar orbit around the moon isn't necessary if you have moon starlink (no comm loss on dark side). This saves energy otherwise needed for inclination changes.

What? Gateway is planned to be in a Halo orbit at L1. Getting to the lunar poles is actually cheaper from L1 than directly getting there Apollo style, and the poles are of interest for exploration due to areas of permanent sunlight (for power) and permanent shade (for potential ice).

Moon starlink? For comms you just need one or two satellites at L2. The Chinese already have one there. And it's not the dark side, it's the far side.

Goram people being wrong on the internet making me correct them grumble grumble.

Getting to the lunar poles from Earth through Gateway is a diversion that takes more energy than going direct.

> What? Gateway is planned to be in a Halo orbit at L1.

Can you provide a source for this? I’m not seeing anything that says gateway is ever expected to even get particularly close to L1.

> Gateway is planned to be in a Halo orbit at L1

https://en.wikipedia.org/wiki/Lunar_Gateway#Orbit_and_operat...

No, it doesn't really change anything about the gateway plan. It was always unnecessary technically, and necessary politically (to support a lunar landing mission).

It's mentioned in your link, but I'd like to call it out more clearly: NASA has also recently selected SpaceX to be the sole launch provider for the initial Lunar Gateway, and the sole launch provider for resupplying it as well.

The lunar tollbooth? Hope so.

https://spacenews.com/op-ed-lunar-gateway-or-moon-direct/

To me this seems extremely risky. There are so many things that still need to be solved on Starship such as in LEO refueling.

I hope NASA knows what they are doing.

It _is_ very risky, but the source selection statement lays out in great detail how both other bids were much worse.

SpaceX needs to solve this to go to Mars, NASA is simply paying them to learn (similar to how full fare launches allowed them to learn to reuse the first stage and keep it for themselves).

I don't want to sound too cavalier, but launching and landing are objectively hard - need a combination of power and finesse. Isn't refuelling just connecting a pipe and an ullage motor?

https://en.wikipedia.org/wiki/Propellant_depot#Engineering_d... has a pretty good list of concerns.

Starship's idea of (1) you send the fuel up in another Starship and (2) you refuel by bumping uglies and then accelerating in a direction that makes the fuel simply fall into the empty vehicle, together solves a lot of problems that imply a static depot.

Dynetics also required refueling and the National Team required assembly in NHRO. SpaceX was seen as less risky than the other two because their refueling was in low Earth orbit and was done prior to human launch.

There would be many other core unsolved technical challenges if they had gone with a traditional system like the SLS. For example:

- Standing it upright

- Fueling it

- Launching

- Building it

End of the day, Starship is a new and not-fully-proven design, but (parts of it) _have_ launched, and NASA knows that SpaceX is committed to the vision either way. Much better than a contractor design which exists only on paper and only to win a bid.

Worth noting orbital refueling may not actually be that difficult, it was strongly rumored that Senate Launch System backers put the kibosh on refueling because it would jeopardize the SLS jobs program if it worked:

"We had released a series of papers showing how a depot/refueling architecture would enable a human exploration program using existing (at the time) commercial rockets," Sowers tweeted on Wednesday. "Boeing became furious and tried to get me fired. Kudos to my CEO for protecting me. But we were banned from even saying the 'd' word out loud. Sad part is that ULA did a lot of pathfinding work in that area and could have owned the refueling/depot market, enriching Boeing (and Lockheed) in the process. But it was shut down because it threatened SLS."

https://arstechnica.com/science/2019/08/rocket-scientist-say...

Orbital refueling is done for decades; e.g. Progresses supplied Soviet space stations with fuel to maintain the orbit.

What's trickier is orbital refueling of cryogenic liquids - LOX, liquid methane. It still looks easier than refueling with LH2, and ULA did a lot of work with that (search for "propellant depot"). Active cooling might be required; humanity had experience storing cryogenics in space for long time (e.g. Buran, which used LOX for RCS, was able to store it for a month on orbit), but we still might want a better solution (the one which actively turns vapors back into the liquid phase using onboard energy and radiators).

Yet there is little which seems too complex in the refueling idea.

IIRC ULA had a plan to run an internal combustion engine off the hydrogen/oxygen boil off, providing power for stage systems and or cooling/recondenser in their ACCESS upper stage concept.

Of course ideally you don't want to loose any propelants to boil off but it still seems like a nice idea.

I hope they will go back to pushing the limits more again. Since the Space Shuttle and now the SLS they got stuck in a rut and iterated very slowly. It’s unbelievable how long the SLS, which basically repeats what has been done 50 years ago, is taking and how much it costs. Better put that money into several programs that have a chance of failing but will also deliver new technology.

Yup. Break the tyranny of mission assurance.

I had it the other way around. I was worried about SpaceX overselling their actual capabilities until NASA gave them the contract.

NASA rated SpaceX's approach equal or better in technical areas and management areas than the competitors, and it was the cheapest.

Much more need to be solved for the other two options, so the choice seems logical. Whether the best option is a good enough option, is a different question.

They should also solve for that whole explosion thing it's been doing.

(This is mostly a joke I'm sure they will)

It's kind of mind boggling to me as a non-American how the US can spend such vast amounts of money on defense and so relatively little on space. Surely even from a purely egoistic perspective the prestige and potential profit of being the only serious player in "space industrialism" must be worth a few less fighter jets?

The second flight, which will carry astronauts to the Moon, could launch as early as 2024

Extremely unlikely. The first Apollo landing was the result of years of research and the expensive development of whole industries and skills. One smallish company has no hope of duplicating that environment in such a short time.

Apollo was more then just a lander, it was a gigantic new rocket, a capsule, a service module a lander and an accent stage.

SpaceX is by no means a small company, its arguably the most successful space company ever. They are also the most vertically integrated space company. They launch far more then anybody, largest sat operator, largest provider of cargo and crew for human space flight.

Also your comparison completely misses a few things, Apollo had to invent a huge amount of things and build infrastructure that is now standard and well known well understood. Just think about the difference in complexity of computers. NASA still has to knowledge and much more knowledge about the moon and human space flight since then.

SpaceX can reuse a huge amount of things from existing vehicles and capsules. The Raptors have been under serious development since 2014. Apollo was a huge number of different very unique parts, Starship uses Raptor engines for all parts from earth to the moon.

2024 is still a very difficult target but its not because you need to develop a whole new industry. Its because the vehicle is absurdly big and has crazy operations.

It’s not so bold; they’re going with the lowest bidder.

Politically bold—near suicidal, bureaucratically.

Probably the SLS suppliers will need to be shifted to a DoD boondoggle or heads will roll. We have heard rumblings recently about starting up a $trillions missile program, which might be that boondoggle.

I hope as a test Musk launches a cyber truck to the moon and drives around. The ultimate flex.

I suspect they will. Cybertruck's chassis and battery capacity will dominate every other lunar rover architecture, and a ton of NRE has already been spent making it a production vehicle. Lunarizing the platform will be cheap, and Starship has ample payload mass margins.

> Lunarizing the platform will be cheap

What makes you say that? I would assume almost nothing would carry over.

Lighter materials, revised batteries, adjusted suspension, different wheels, space-buggy interior.

Similar terrain relative navigation, charge control system, motors, etc.

Does it have to though? It would still be a PR miracle even if it was just a non-functional showpiece. I would even say that sending a battery pack up there just for the meme of being able to remotely drive it around is not worth the potential risk of compromising the mission (IIRC the Roadster had most of its electronics stripped out before launch as well.)

Reply on news.ycombinator.com