An alternative — and much less expensive — path to a US moon landing

[ecarfan]

Eric Berger: The oracle who predicted SLS’s launch in 2023 has thoughts about Artemis III

Back in 2017, when NASA was forecasting a 2019 launch date for Artemis 1, an “unbiased industry insider” predicted that the actual date would be “around 2023”. The critics piled on.

Obviously, this person turned out to be correct.

Then in 2020 the same person predicted that SpaceX would win the HLS contract and it might even be sole source. Again, most space industry observes found that unlikely.

And again, that person turned out to be correct.

Now the same individual is predicting that Artemis III, the first crewed moon mission, will not fly in 2025 as NASA plans but instead will be no sooner than 2028.

A lot can happen in the next several years; politically, economically, technologically. You need to read the article to understand the reasoning that supports this speculation:

The safest and lowest-cost means of completing an Artemis mission to the Moon, therefore, may involve four astronauts launching to a fairly high altitude in low-Earth orbit on Crew Dragon and rendezvousing with a fully fueled Starship. The astronauts would then fly to the Moon, land, and come back to rendezvous with Crew Dragon in Earth orbit. They would then splash down on Earth inside Dragon.

This architecture is less risky because it doesn't involve launching on SLS, nor does it require two rendezvous and dockings in lunar orbit, far from Earth. The crew would only spend a couple of more days aboard Starship than they would during the existing Artemis III plan, so Starship life support should be up to the task. If you care about costs, this plan also excludes the $4.1 billion launch cost of Orion and the SLS rocket and substitutes Crew Dragon, which would be on the order of one-twentieth of the cost.

Such an approach would not only save many billions of dollars by not using SLS, it would also eliminate the need for the very costly Gateway. It’s actually quite brilliant. F9 and Crew Dragon are obviously fully “flight proven”. Starship development for HLS including in-orbit refueling is already ongoing and funded by NASA.

Read the article. Then discuss.

 

[scaesare]

Hehe... that article was one of my "read this when you get a chance later" browser tabs... which I just did.

2028 as a starting point seems pretty reasonable to me too...

I found this point interesting:

It can reasonably be argued that Starship is also not safe to launch on and land back on Earth. It, too, is a large and complex vehicle that will come back through Earth's atmosphere, dissipate heat, and perform delicate maneuvers before landing under the power of its own engines. Even though Starship will launch at least dozens of times per year, the vehicle is unlikely to meet NASA's safety requirements for humans for a long, long time. So Starship-only missions to the Moon are not a near-term solution.

Didn't Crew Dragon become human rated after only a couple of unmanned tests in addition to a number of sucessful Falcon 9 launches?

Why would it take so much longer to qualify Starship? He mentions Starship likely doing "dozens" of flights a year, yet he says it could take "a long, long time" to be cleared.. not sure what that equates too but it feels like he talking years. That could mean Starship has gone up 25+ times at that point... I understand that retro-repulsive landing is new, but would it take years to qualify for human rating at that launch cadence?

Alos interesting: the Space Prophet is predicting we might see Falcon 9 + Crew Dragon end up being the lynchpin for Artemis III....

 

[Cosmacelf]

I love Berger’s writing. The “Space Prophet”! Love it.

But yes, the plan he outlined makes the most sense, which is why it is unlikely to happen, unless SLS has an oopsie moment.

As far as Starship all the way, well, that would require astronauts to chill in orbit while it gets refuelled a bunch of times. Not to mention that Dragon has in flight abort, while Starship does not.

And coming back, NASA is very likely to think their parachutes are safer than Starship’s propulsive landing.

 

[ecarfan]

Didn't Crew Dragon become human rated after only a couple of unmanned tests in addition to a number of sucessful Falcon 9 launches?

Yes, however:

— Crew Dragon was based on Cargo Dragon which had flown successfully many times

— The Crew Dragon re-entry technology (heat shield) was based on Cargo Dragon and ablative heat shield re-entries are longstanding technology that NASA is comfortable with (remember that NASA nixed Elon’s preferred retropropulsive landing proposal).

— As @Cosmacelf noted, Crew Dragon has proven inflight abort capability.

— And as you noted, Crew Dragon was being placed in orbit on the F9 which NASA had a lot of confidence in.

In contrast:

— Starship is an all new vehicle design with a new engine design.

— Starship’s orbital re-entry technique (the “belly flop”) is a radical departure from every crewed vehicle in the past.

— Starship’s inflight abort capability remains to be seen. I believe Elon has said it will have that capability?

So launching crew to LEO and and then returning them to Earth using a completely new vehicle with a radically new re-entry technique is more risky than using SLS (partially new, but basically all old technology) and Orion (standard re-entry technique). So how many perfect flights will NASA require from SpaceX to human rate Starship? I don’t know the exact number but surely lot more than what it took to human rate Crew Dragon.

By only using Starship to take crew to and from the lunar surface, NASA avoids the “belly flop” risk and reduces the new vehicle design risk; if Starship doesn’t make it to the Gateway rendezvous the crew simply boards Orion and goes home.

 

[scaesare]

Yes, however:


...

— Starship’s orbital re-entry technique (the “belly flop”) is a radical departure from every crewed vehicle in the past.

...

Yeah, I suspect this is the major holdup... thinking about it SpaceX scrapped powered module landings in favor of parachute touchdowns in response to NASA concerns as well...

 

[Watts_Up]

... SpaceX scrapped powered module landings in favor of parachute touchdowns in response to NASA concerns as well...

Beside glidding (Space shutle, Virgin Galactic, Boeing X-37...)
and the hundred of Falcon (and few Blue Origin) vertical booster landing,

since the 1985 McDonnell Douglas DC-X,
is there any 'non parachute' Earth landing performed?

 

[Grendal]

Yes, however:

— Crew Dragon was based on Cargo Dragon which had flown successfully many times

— The Crew Dragon re-entry technology (heat shield) was based on Cargo Dragon and ablative heat shield re-entries are longstanding technology that NASA is comfortable with (remember that NASA nixed Elon’s preferred retropropulsive landing proposal).

— As @Cosmacelf noted, Crew Dragon has proven inflight abort capability.

— And as you noted, Crew Dragon was being placed in orbit on the F9 which NASA had a lot of confidence in.

In contrast:

— Starship is an all new vehicle design with a new engine design.

— Starship’s orbital re-entry technique (the “belly flop”) is a radical departure from every crewed vehicle in the past.

— Starship’s inflight abort capability remains to be seen. I believe Elon has said it will have that capability?

So launching crew to LEO and and then returning them to Earth using a completely new vehicle with a radically new re-entry technique is more risky than using SLS (partially new, but basically all old technology) and Orion (standard re-entry technique). So how many perfect flights will NASA require from SpaceX to human rate Starship? I don’t know the exact number but surely lot more than what it took to human rate Crew Dragon.

By only using Starship to take crew to and from the lunar surface, NASA avoids the “belly flop” risk and reduces the new vehicle design risk; if Starship doesn’t make it to the Gateway rendezvous the crew simply boards Orion and goes home.

The potentially good thing is that IF SH and Starship are as reusable as Elon wants, then once they get back a few then SpaceX can examine them. From there, you'll get the "Block 5" upgrades where SpaceX replaces weak parts with robust parts designed to hold up with regular use. We've even seen improvements in the Block 5 F9's. The early boosters seem to good for around 10 launches, while the later boosters are up to 14 and still going. Starship needs to be a rock solid robust design for what SpaceX and Elon wants to do with it. I can easily see Starship getting human rated after 40 to 50 solid launches and landings with no failures. Let's be clear that there will be failures before they reach 40 to 50 launches.

 

[ggr]

Hehe... that article was one of my "read this when you get a chance later" browser tabs... which I just did.

2028 as a starting point seems pretty reasonable to me too...

I found this point interesting:



Didn't Crew Dragon become human rated after only a couple of unmanned tests in addition to a number of sucessful Falcon 9 launches?

Why would it take so much longer to qualify Starship? He mentions Starship likely doing "dozens" of flights a year, yet he says it could take "a long, long time" to be cleared.. not sure what that equates too but it feels like he talking years. That could mean Starship has gone up 25+ times at that point... I understand that retro-repulsive landing is new, but would it take years to qualify for human rating at that launch cadence?

Alos interesting: the Space Prophet is predicting we might see Falcon 9 + Crew Dragon end up being the lynchpin for Artemis III....

NASA is talking about human-rating Orion/SLS after only two successful missions (each) AIUI.

 

[scaesare]

NASA is talking about human-rating Orion/SLS after only two successful missions (each) AIUI.

And that's traditional parachute landings, right?

IF so and it really takes ~50 sucessful landings as @Grendal surmises, as oppossed to 2 for SLS, I find it interesting that NASA is so much more concerned about landing risk than launch risk....

 

[petit_bateau]

Buryan was also a non parachute landing. Not sure if it did a full orbit though.

 

[scaesare]

Buryan was also a non parachute landing. Not sure if it did a full orbit though.

Ahh yeah right... much like the Shuttle. Not sure how I forgot about that... I have an e-copy of this, interesting to leaf through:

As an aside... too bad it never had humans aboard.

Aside #2: Also too bad the remaining ones were inside the cosmodrome building that collapsed

Aside #3: This abandoned full scale wooden mockup sitting out in field is cool:

 

[Grendal]

Buran:

 

[scaesare]

The potentially good thing is that IF SH and Starship are as reusable as Elon wants, then once they get back a few then SpaceX can examine them. From there, you'll get the "Block 5" upgrades where SpaceX replaces weak parts with robust parts designed to hold up with regular use. We've even seen improvements in the Block 5 F9's. The early boosters seem to good for around 10 launches, while the later boosters are up to 14 and still going. Starship needs to be a rock solid robust design for what SpaceX and Elon wants to do with it. I can easily see Starship getting human rated after 40 to 50 solid launches and landings with no failures. Let's be clear that there will be failures before they reach 40 to 50 launches.

And that's traditional parachute landings, right?

IF so and it really takes ~50 sucessful landings as @Grendal surmises, as oppossed to 2 for SLS, I find it interesting that NASA is so much more concerned about landing risk than launch risk....

Resurrecting this year-old thread where we were discussing the Ars article which implied that Starship might need upwards of 40-50 successful flights to get human rating... which I found odd...

This article outlines a discusses what's needed a bit more to make the 2025 goal:

SpaceX has a long road ahead of it before the Starship is ready to take people back to the moon. It has to demonstrate the ability to launch both the Super Heavy and the Starship into low Earth orbit and then recover them intact. Then SpaceX has to turn the two rockets around quickly and launch them again. The company has to master on-orbit refueling if it expects to use the Starship as a lunar lander. It has to fly at least one uncrewed mission to the moon before NASA will trust the vehicle enough to put people on it.

I can see a a couple dozen unsuccessful attempts to get all that proved out. But that says 1 successful flight to moon before human rating. And if that's the bar for the moon, why would the bar for a less stringent mission like earth orbit be any higher?

 

[JB47394]

I can see a a couple dozen unsuccessful attempts to get all that proved out. But that says 1 successful flight to moon before human rating. And if that's the bar for the moon, why would the bar for a less stringent mission like earth orbit be any higher?

Artemis has Starship landing on the Moon. That's a vacuum and 0.2 g.

A LEO Starship mission is landing on the Earth. That's atmospheric heating, atmospheric maneuvering, a flip, then landing. Either on the ground or by sliding between the chopsticks. All in 1.0 g.

Edit: Come to think of it, I wonder if all this flipping around is going to demonstrate a need for collapsing tanks. That is, tanks that, one way or another, stay conformal to the remaining propellant. The comical images that come to mind are a moving bulkhead like a plunger, or a flexible tank that sits inside the metal one and collapses as the propellants are used. Or a combination of the two.

 

[scaesare]

Artemis has Starship landing on the Moon. That's a vacuum and 0.2 g.

A LEO Starship mission is landing on the Earth. That's atmospheric heating, atmospheric maneuvering, a flip, then landing. Either on the ground or by sliding between the chopsticks. All in 1.0 g.

Good point.

But 40-50 successful flights/landings to get human rating?

While landing on the moon may not have the atmospheric and gravitational risks that returning to earth does, it would seem there were a whole host other equally-risky items: Luner Module separation, LM turn-around maneuver, powered landing, ascent engine firing, Command Module docking, LM undocking, CSM re-firing engines...

That's a pretty complicated procedure. Just seems odd to require Starship to fly/land successfully from LEO 50 times, yet if they do the modern equivalent of what Apollo did for the moon once, NASA is good with it...

 

[JB47394]

But 40-50 successful flights/landings to get human rating?

I think Elon was just looking forward to Starships flying multiple times a day, so cranking out 40-50 flights is about a week of elapsed time. I'm exaggerating, but I think that was his general attitude; develop so much confidence in it that there's no chance of a person's life being at risk. After all, if you're going to put dozens of people into one, you really really don't want to crash it.

While landing on the moon may not have the atmospheric and gravitational risks that returning to earth does, it would seem there were a whole host other equally-risky items: Luner Module separation, LM turn-around maneuver, powered landing, ascent engine firing, Command Module docking, LM undocking, CSM re-firing engines...

The Starship reentry risks are not things that you can recover from. If you fail, you're done (Columbia). In space, or even in 0.2 g, you have time (Apollo 13).

 

[scaesare]

I think Elon was just looking forward to Starships flying multiple times a day, so cranking out 40-50 flights is about a week of elapsed time. I'm exaggerating, but I think that was his general attitude; develop so much confidence in it that there's no chance of a person's life being at risk. After all, if you're going to put dozens of people into one, you really really don't want to crash it.

As far as I know, Elon has never said that... it was surmised in this thread that 40-50 wouod be needed based on this Ars article, which said:

Even though Starship will launch at least dozens of times per year, the vehicle is unlikely to meet NASA's safety requirements for humans for a long, long time. So Starship-only missions to the Moon are not a near-term solution.

The Starship reentry risks are not things that you can recover from. If you fail, you're done (Columbia). In space, or even in 0.2 g, you have time (Apollo 13).

Sure, but the failure of several of those things I mentioned is also game over if you can't fix it in space, and Apollo 13 was a bit of a 1-in-a-million example.

In any case, I agree, there's risky items that must be proven out, most certainly. Just feels like a bar of somewhere approaching 50 is out of line for most everything else I've seen NASA require.

But, I'm repeating myself at this point...

 

[JB47394]

As far as I know, Elon has never said that...

Fair enough. I made the assumption - not that it sounds like something Elon would say. He'd say something more like "A bunch of people will probably die in the beginning." Which he did, in the context of pursuing Mars colonization.

 

[ecarfan]

From my post that started this thread:

Now the same individual is predicting that Artemis III, the first crewed moon mission, will not fly in 2025 as NASA plans but instead will be no sooner than 2028.

And now this: NASA to push back moon mission timelines amid spacecraft delays

The U.S. space agency is expected to announce the plans on Tuesday after spending months tracking progress with contractors and considering changes to the Artemis program, a multi-billion dollar effort that includes returning the first astronauts to the moon since the last Apollo mission in 1972. NASA's second Artemis mission is expected to be pushed beyond its planned late-2024 target after issues were uncovered with the Lockheed Martin-built (LMT.N) Orion crew capsule's batteries during vibration tests, two of the people said. The batteries will need to be replaced.

And this:

Artemis 3 - planned to be the first mission landing humans on the moon in late 2025 using the Starship landing system from NASA contractor SpaceX - will likewise be pushed back. Billionaire Elon Musk's SpaceX is taking longer than expected to reach certain development milestones, all four people said.

I still think this mission plan is genius:

The safest and lowest-cost means of completing an Artemis mission to the Moon, therefore, may involve four astronauts launching to a fairly high altitude in low-Earth orbit on Crew Dragon and rendezvousing with a fully fueled Starship. The astronauts would then fly to the Moon, land, and come back to rendezvous with Crew Dragon in Earth orbit. They would then splash down on Earth inside Dragon.

 

[JB47394]

I still think this mission plan is genius:

I'd love to see how that would work. When I do the math on delta-v, a return trip using Starship only works with tanking at the Moon. Where's all the delta-v coming from? It takes 3.94 km/s to get from LEO to LLO (low lunar orbit), 1.72 km/s to land, 1.72 to get back to LLO, then 3.94 to get back to LEO. That's 11.32 km/s. Starship has 6.9 km/s with 100 tons of cargo. If they drop the cargo sufficiently, then Starship can barely get back to LLO. V2 Starship would help, of course.

I assume that's why NASA has HLS staying at the Moon and only Starliner comes back.