SpaceX to NASA quote : simplified mission architecture : Technical discussion

[TomH]

I first posted this in the Frankenrocket thread and later in germane threads.

I'm going to suggest that if those in charge believe that we simply have to make a 7th landing on Luna before China makes their first, and it rises to the level of a National Emergency, here is a method that might suffice.

In a dual launch architecture, SLS/Orion (since they exist and that money is already spent) keeps its already assigned role per the Artemis program.

SpaceX still provides the lander BUT, the LV is a fully expendable stripped down V4 SH and SS with no refueling. The lander is fully disposable and two staged like the Apollo LM, but has more Altair like dimensions. The Descent Stage is essentially a copy of Hoppy. The Ascent Stage is a modified Dragon, sans TPS, possibly reoriented, the trunk components possibly as now or possibly reconfigured, and a whole lot of Super Dracos somehow configured as ascent stage engines.

It is a distraction. It deviates from the POR. It expends more money. It delays getting an actually productive Lunar program underway, it delays getting the Mars program underway. But for those with such fragile egos that everything else has to stop so that the 7th Lunar landing goes to the US rather than China, this could do it. IMHO, this is about as simple and fast as it could be done. Would I do it? No, I would stick with the POR. I don't think it matters if China lands a tiny flags and footprints mission shortly before the US lands an exceptionally advanced ship. We landed there almost 56 years ago. History will not recall anything about this so-called second-space race other than a minor footnote. If anything, history will only note how fragile and easily affected certain egos were in the mid 2020s.

[spider_best9]

My vision for a slightly simplified mission would be like this:

- The Lander: a shortened HLS with a modified Crew Dragon on top to serve as both an escape vehicle and Lunar Ascent Vehicle. Modified to have a pass-through tunnel in the bottom, though the trunk, into the Lunar habitat. The trunk would hold extra fuel for the Super Dracos longer burn times during ascent.
The HLS part would still perform the descent and remain on the moon.

- Refueling: done from a Depot, filled with expendable tankers, maybe as little as 3 of them.

- Return to Earth: still done with Orion as before.

So that would be as little as 5 launches on SpaceX's part.

[thespacecow]

The cheapest single-HLS case I can come up with is the "D2 on the nose" variant, which only requires about 1600t of prop and a single, uncrewed refueling.  However, there's a catch, as you'll see:

1) HLS launches, uses its residual prop to boost to VLEO+400m/s (about 200 x 1740).
2) Crew launches on F9/D2, into the 200 x 1740 orbit, where it docks with the HLS.
3) HLS does TLI, with the D2 still docked to the nose.
4) HLS goes to LLO, where the D2 undocks.
5) HLS does lunar surface mission.
6) HLS re-docks with D2 after surface mission.
7) HLS+D2 does TEI.
8 ) Before entry interface, crew climbs into the D2 and does direct EDL.
9) HLS either disposes of itself or aerocaptures into LEO.

As we've discussed, ad nauseam, on the thread I linked above, this requires non-trivial but still fairly modest mods to the D2, because it has to work in a deep space environment (different radiation and thermal requirements), the heat shield has to be qualified for ~11km/s entry speed, and the D2 and docking system have to withstand about 1G eyeballs-out acceleration from the Raptors during TLI, LOI, and TEI.  But the prop to LEO load is small, there are no crewed refuelings, and the HLS doesn't have to return propulsively to LEO to do RPOD with the D2, where it may be subject to "the RAAN problem".²

I bet something like this is what they proposed.

Only this mission architecture fits all the criteria:
1. Simplified mission architecture and conops
2. Faster return to the Moon
3. Improving crew safety
4. As Elon indicated, they're going to do the entire mission themselves

Using 2 HLS is obviously not simpler, it's more complicated in terms of conops since you'll need additional refueling. Using a single Starship for the entire mission has questionable safety in the near term, and will put more milestones on Starship which is already late and on the critical path.

D2 + HLS also has a unique feature which is on the way to the Moon and back, the crew always have two spacecraft at their disposal, so if one has some problems - like an Apollo 13 scenario - they can rely on the other one to survive. I think this will go well with NASA who values this kind of redundancy.

[Kazioo]

- The Lander: a shortened HLS

The proposal likely isn't about any simplifications to the actual Starship HLS.

Why? Because SpaceX used the forbidden phrase: "improving crew safety". A theoretical less capable vehicle made quicker cannot be described as safer than the real deal that is supposed to be the future of space exploration. That would be a big self-own SpaceX would never say.

So my guess this is just a political counterattack to something that was a political attack in the first place: they will probably propose that all NASA needs to simplify the mission and improve crew safety is to delete SLS and maybe even Orion.

[spider_best9]

Why? Because SpaceX used the forbidden phrase: "improving crew safety". A theoretical less capable vehicle made quicker cannot be described as safer than the real deal that is supposed to be the future of space exploration. That would be a big self-own SpaceX would never say.

I would say that a Lander that only needs to land and an ascent/escape vehicle(Modified Dragon) based on a proven to work reliably craft it would improve crew safety and speed up the timeline.

[DanClemmensen]

Why? Because SpaceX used the forbidden phrase: "improving crew safety". A theoretical less capable vehicle made quicker cannot be described as safer than the real deal that is supposed to be the future of space exploration. That would be a big self-own SpaceX would never say.

I would say that a Lander that only needs to land and an ascent/escape vehicle(Modified Dragon) based on a proven to work reliably craft it would improve crew safety and speed up the timeline.

No new design or Kerbal mashup can be completed and land crew on the moon before 2029. The design and implementation process just take too long.

I predict that the SpaceX alternative mission architecture has no new hardware elements and just uses stuff that is already being developed for Artemis III, plus Crew Dragon for Earth-LEO and LEO-Earth.

[Brigantine]

Very similar to the BLEO Dragon concept mentioned upthread, but with Orion:

- HLS #1 fueled in LEO, does a BLT (less than 3.26 km/s to lunar capture)
- HLS #2 (OTV) fueled in LEO
- crewed Orion to LEO, docks with HLS #2
- HLS #2 tugs Orion to LLO, using low thrust or even lunar landing thrusters depending on what forces the docking adapter can handle (3.94 km/s plus Oberth penalty, with extra mass of Orion and reduced ISP)

Crew stays with (but not always in) fully fueled Orion as far as LLO, with abort to Earth via Orion

- HLS #1 from BLT/HLO to LLO for rendezvous with HLS #2 (fuel transfer depending on performance) and then Orion (crew)
(0.68 km/s plus some extra to make it timely)
- HLS #2 disposal (lunar impact, or keep it as comms relay until crew returns to Earth)

Challenge: the exact details of the LLO to be accessible from BLT/HLO, TLI, and the (polar?) landing site. possibly a somewhat elliptical orbit with apolune over one of the poles (thermal management & plane change considerations). May require non-trivial stationkeeping.

(HLS #1 tugs Orion from 'LLO' to an actual LLO)
- HLS #1 does the whole landing thing (1.73 km/s)
- HLS #1 returns to LLO/Orion (1.73 km/s)
- Crew returns in Orion via TEI (0.82 km/s plus penalties for plane-change)

Orion (with service module) can use between 1346 m/s and 1587 m/s of delta-v.

∆V requirements:
<7.9 km/s for HLS #1 (<3.94 then refueling option then 3.46. Payload for lunar surface ops)
>3.94 km/s for HLS #2 OTV (with 27 tons Orion+ESM 'payload')
>0.82 km/s for Orion

Advantages:
- No exotic variants, no V4s. Just HLS, tankers and Orion-ESM (depots optional)
- Tankers all go to LEO (maybe one HLS-HLS fuel transfer in LLO)
- Although there are many tanker flights, it is two identical HLS refueling campaigns separated in time
- Starship does not require human rating below LEO
- No waiting for week-long orbits in the event of contingencies
- Uses the EDL heatshield with BLEO flight heritage (albeit not without issue) including abort options
- No refueling with crew (although maybe one HLS-HLS fuel transfer with the crew waiting nearby in Orion)
- Crew spends most of their time on a Starship (except for the last ~3 days)

If Starship performance falls a bit below expectations, this plan still has room to shoehorn in refueling in elliptical orbits
[EDIT: and maybe even using ICPS to boost HLS #2 into a more elliptical orbit before disposing ICPS and transferring crew - though I doubt the docking hardware can handle it]

[baddux]

Very similar to the BLEO Dragon concept mentioned upthread, but with Orion:

I don't understand why two HLS is needed. Just one fully fueled HLS that docks Orion in LEO, performs TLI and LOI. Then descent and ascent, dock with Orion that does TEI and HLS is disposed.

This would be Constellation style, Ares V + Altair replaced with Starship.

[TheRadicalModerate]

Very similar to the BLEO Dragon concept mentioned upthread, but with Orion:

I don't understand why two HLS is needed. Just one fully fueled HLS that docks Orion in LEO, performs TLI and LOI. Then descent and ascent, dock with Orion that does TEI and HLS is disposed.

This would be Constellation style, Ares V + Altair replaced with Starship.

What are you going to use to launch the Orion?  The idea is to save the cost of SLS/Orion.

There are really only two viable single-HLS options, unless the HLS Starship can be launched with crew and do its own EDL:

1) Launch - LEO - refuel(uncrewed) - RPOD(D2) - undock(D2) - TLI - LLO - LS - LLO - refuel(crewed) - LEOpropulsive - RPOD(D2) - idle

Pros:
- Little to no modification of D2

Cons:
- Requires crewed post-ascent refueling in LLO (no abort option if it fails)
- RAAN problem (D2 and HLS not in same orbital plane if there's an abort)
- Somewhat more expensive in terms of prop needed in LEO

2) Launch(no refuel) - ~200x1600 - refuel(uncrewed) - RPOD(D2) - TLI(with D2) - LLO - undock(D2) - LS - LLO - RPOD(D2) - TEI(with D2) - undock(D2) - directEDL(D2) - HLSdisposalOrSlowAerobrake

Pros:
- No crewed refuelings.
- No RAAN problem (D2 is automatically in the same orbit in LEO)
- Cheaper in terms of prop.

Cons:
- High radiation environment when D2 and HLS dock in ~200x1600
- D2 mods required for 11km/s reentry
- D2 avionics need to be radiation-hardened for deep space
- D2/HLS dock need to support about 1G acceleration, eyeballs-out
- Possible D2 thermal changes for cislunar space and LLO.
- Still a critical D2-HLS RPOD in LLO, with no good abort option.

The main advantage of the two-HLS profiles is that there are lots of abort options, and the launch-lunarOrbit-EDL leg is independent of the type of HLS it's meeting in lunar orbit.  It can also support NRHO RPODs, if Gateway were to somehow survive when SLS/Orion did not (unlikely), or if the waiting HLS required NRHO access for some reason.

[baddux]

Very similar to the BLEO Dragon concept mentioned upthread, but with Orion:

I don't understand why two HLS is needed. Just one fully fueled HLS that docks Orion in LEO, performs TLI and LOI. Then descent and ascent, dock with Orion that does TEI and HLS is disposed.

This would be Constellation style, Ares V + Altair replaced with Starship.

What are you going to use to launch the Orion?  The idea is to save the cost of SLS/Orion.

If NASA+SpaceX want to do "simplified mission architecture" to beat the Chinese then they probably want to use most mature and tested hardware, in this case Orion which is the only thing that can currently land the crew from the Moon. SLS is the quickest way to get Orion to LEO/HEO and as mentioned earlier the money is largely already spent.

Although I don't believe SpaceX suggest this, they probably propose one of those two you described (I would say the option 2. which does not need to get the tanker to LLO).

[TomH]

The idea is to save the cost of SLS/Orion.

My understanding of "the idea" is that we have to make our 7th lunar landing before China makes their first, juts so we can save face. So expedited schedule matters more than money (not to me, to those pushing this).

Besides, this SLS/Orion is already paid for, isn't it? Is it better to fly it, or put it in a museum?

[TheRadicalModerate]

If NASA+SpaceX want to do "simplified mission architecture" to beat the Chinese then they probably want to use most mature and tested hardware, in this case Orion which is the only thing that can currently land the crew from the Moon. SLS is the quickest way to get Orion to LEO/HEO and as mentioned earlier the money is largely already spent.

Fair enough.  Although I suggest you read Casey Handmer's Orion post and still see if you feel the same way.  (Lots of ranting in this, but I also learned a few additional scary things.)

[DanClemmensen]

Very similar to the BLEO Dragon concept mentioned upthread, but with Orion:

I don't understand why two HLS is needed. Just one fully fueled HLS that docks Orion in LEO, performs TLI and LOI. Then descent and ascent, dock with Orion that does TEI and HLS is disposed.

This would be Constellation style, Ares V + Altair replaced with Starship.

What are you going to use to launch the Orion?  The idea is to save the cost of SLS/Orion.

If NASA+SpaceX want to do "simplified mission architecture" to beat the Chinese then they probably want to use most mature and tested hardware, in this case Orion which is the only thing that can currently land the crew from the Moon. SLS is the quickest way to get Orion to LEO/HEO and as mentioned earlier the money is largely already spent.

Although I don't believe SpaceX suggest this, they probably propose one of those two you described (I would say the option 2. which does not need to get the tanker to LLO).

Orion is not "proven", or "most tested", although I suppose in terms of age it's "mature" enough to vote. I really, really hope Artemis II is successful, but it is not yet completed. Artemis I's Orion showed troubling heat shield degradation.

Orion will not return a crew form the Moon. It would return the crew from NRHO. The proposed alternatives do not intend to "return a crew from the Moon". They intend to return a crew to LEO, and then use Dragon 2 to return the crew form LEO to Earth. Dragon 2 has flown 32 orbital missions, all including successful returns to Earth, 18 of them crewed.

Money already spent does not matter: sunk costs are sunk. Cost of the next mission matters. The SLS/Orion portion of Artemis III will cost roughly $6 Billion new money.

[baddux]

Orion will not return a crew form the Moon. It would return the crew from NRHO. The proposed alternatives do not intend to "return a crew from the Moon". They intend to return a crew to LEO, and then use Dragon 2 to return the crew form LEO to Earth. Dragon 2 has flown 32 orbital missions, all including successful returns to Earth, 18 of them crewed.

I was referring to this alternative, which does return Dragon from the Moon (LLO):

2) Launch(no refuel) - ~200x1600 - refuel(uncrewed) - RPOD(D2) - TLI(with D2) - LLO - undock(D2) - LS - LLO - RPOD(D2) - TEI(with D2) - undock(D2) - directEDL(D2) - HLSdisposalOrSlowAerobrake

If SpaceX can return Starship HLS to LEO, I'm all for it, but to me it sounds unrealistic to be the most simple architecture to do Artemis 3 as soon as possible.

[spacenut]

From the pictures on here, SpaceX already has about 6 Starships under construction.  Don't know how many Superheavies, but as shown one has already been reused.  The simplification sees to be do do away with SLS/Orion and use Dragon to transfer and retrieve astronauts in earth orbit.  Starship lander would go to the moon, get in LLO and then land, take off, refuel in LLO from a transfered tanker/depot, and go back to earth.  Use atmosphere drag to slow down and get into orbit for Dragon to bring them home.  Lunar Starship could be reused refueled and used again.   No need for the Artemis station or SLS/Orion.   

[DanClemmensen]

Orion will not return a crew form the Moon. It would return the crew from NRHO. The proposed alternatives do not intend to "return a crew from the Moon". They intend to return a crew to LEO, and then use Dragon 2 to return the crew form LEO to Earth. Dragon 2 has flown 32 orbital missions, all including successful returns to Earth, 18 of them crewed.

I was referring to this alternative, which does return Dragon from the Moon (LLO):

2) Launch(no refuel) - ~200x1600 - refuel(uncrewed) - RPOD(D2) - TLI(with D2) - LLO - undock(D2) - LS - LLO - RPOD(D2) - TEI(with D2) - undock(D2) - directEDL(D2) - HLSdisposalOrSlowAerobrake

If SpaceX can return Starship HLS to LEO, I'm all for it, but to me it sounds unrealistic to be the most simple architecture to do Artemis 3 as soon as possible.

Sorry, I lost track of which of the very, very many alternatives we were discussing. Please take a look at the long thread at
    https://forum.nasaspaceflight.com/index.php?topic=59662.0
and particularly at the first post. This is a brute-force approach, optimized for simplicity and zero new design rather than cost.

Dragon 2 from Earth to LEO
One instance of Starship HLS acting as an OTV (Orbital transfer vehicle) takes crew to NRHO
another Starship HLS  (HLS) takes crew to lunar surface and back to NRHO   (just like the Artemis III PoR)
OTV takes crew to LEO
Dragon 2 returns crew to Earth.

The architecture uses propellant depots and tankers to perform refilling for OTV and HLS. All refilling is completed before crew is launched form Earth.

[stormhelm]

Interesting how space x all of a sudden released video of simplified starship lander and saying achieved milestones in design and starting to build hardware.
Yes—SpaceX just dropped a major update on its Starship lunar lander, revealing a simplified design and claiming 49 milestones completed, including hardware fabrication. This is a direct response to NASA’s renewed scrutiny and China’s accelerating lunar ambitions. Below assessment send to me and I totally agree..

Here’s what’s unfolding:

Why the Sudden Update?
NASA Acting Administrator Sean Duffy recently reopened the Artemis III lunar lander contract due to delays in SpaceX’s Starship development. China’s progress toward a 2030 crewed Moon landing added pressure. Elon Musk initially responded with social media attacks, but SpaceX quickly pivoted to a more strategic move: releasing a detailed blog post titled “To the Moon and Beyond” on October 30.

 What’s New in the “Simplified” Starship Lander?
SpaceX is now formally assessing a streamlined mission architecture that aims to:

Accelerate timelines for Artemis III (now targeting 2027–2028)

Improve crew safety

Reduce operational complexity while retaining Starship’s core capabilities

Key features of the simplified lander:

Pressurized cabin volume >600 m³, nearly ⅔ of the ISS

Dual airlocks for surface operations

Cargo variants capable of landing up to 100 metric tons (rovers, habitats, reactors)

In-orbit refueling remains part of the plan, with a test targeted for 2026

 Milestones Claimed
SpaceX says it has completed 49 contract milestones, including:

Life support and thermal control system demos

Landing leg structural tests on simulated lunar regolith

Raptor engine firings under lunar-like conditions

Docking adapter qualification for Orion-Starship linkup

Cabin systems testing for air distribution and sanitation

 Hardware Already in Progress
SpaceX has begun fabricating a flight-capable Starship HLS cabin, and has produced:

Over three dozen Starships

More than 600 Raptor engines

Extensive infrastructure for launch and testing

 Strategic Implications
This update isn’t just technical—it’s political and competitive. NASA’s move to reopen the contract signals urgency, and SpaceX’s response is both a defense of its progress and a bid to stay ahead of rivals like Blue Origin. Musk’s claim that Starship could handle “the whole moon mission” hints at ambitions beyond just being a lander.



Doubtful they will have refueling test in 2026 and let alone land on the moon time frame despite news.
 The 2026 refueling test and 2027–2028 lunar landing timeline are ambitious—borderline optimistic—given the technical and logistical hurdles SpaceX still faces. Here's a breakdown of why many experts say:

 In-Orbit Refueling: The Bottleneck
Starship’s lunar architecture hinges on cryogenic propellant transfer in orbit, which has never been done at scale. Challenges include:

Boil-off mitigation: Keeping super-cold methane and oxygen stable during transfer

Fluid dynamics in microgravity: Managing slosh, pressure, and flow without gravity

Docking precision: Multiple Starships must rendezvous and transfer fuel reliably

NASA’s own GAO report flagged this as a critical risk, and SpaceX has yet to demonstrate even a basic cryo transfer in orbit.

 Hardware vs. Flight-Ready Systems
SpaceX has built dozens of Starships and hundreds of Raptor engines, but:

Most are test articles or prototypes, not lunar-optimized

The HLS cabin is still in early fabrication stages

No orbital refueling depot or tanker variant has flown

Even if they accelerate builds, integration and qualification for crewed lunar ops is a whole different beast.

 Testing Timeline Compression
To meet a 2026 refueling test and 2027 landing, SpaceX would need:

Cryo transfer demo in orbit by late 2026

Full lunar lander flight test (uncrewed) by early 2027

Crew-rated lander, life support, and surface systems validated by mid-2027

That’s a stacked schedule, and any slip in one area cascades across the whole mission.

 Artemis III Dependencies
Starship HLS isn’t standalone—it must:

Dock with Orion in lunar orbit

Support NASA crew systems

Align with SLS launch cadence

If SLS or Orion slips, Starship can’t land crew even if it’s ready. And vice versa.

 Strategic Signaling vs. Reality
SpaceX’s blog post and milestone claims may be more about reassuring NASA and fending off competitors than reflecting near-term readiness. It’s a classic Musk move: set aggressive targets to galvanize internal teams and shape public perception.

Will backfire after his childish insults to acting head of NASA.

Musk is trying to buy time he doesn`t have and probability of failures high. Competitors will have solutions ready before then.
Exactly—this is a classic Musk maneuver: set aggressive timelines to pressure teams, shape public perception, and keep NASA invested. But the technical reality is far more precarious.

 Why the Risk of Failure Is High
Cryogenic refueling in orbit is unproven and extremely complex. SpaceX hasn’t demonstrated even basic fluid transfer in microgravity with cryogenic propellants.

Lunar landing legs and regolith interaction remain untested in real conditions. Simulations help, but they don’t replicate the Moon’s dust dynamics or thermal extremes.

Life support systems for Starship HLS are still in early testing. NASA requires redundancy, fault tolerance, and full integration with Orion systems.

Massive launch cadence needed for refueling: up to 16 Starship launches per lunar mission. That’s a logistical and regulatory nightmare.

 What Could Go Wrong
A single Raptor engine anomaly during ascent or landing could jeopardize the mission.

Boil-off or transfer failure during refueling could strand the lander in orbit.

Thermal or dust damage on the lunar surface could compromise crew safety.

Delays in FAA approvals, pad readiness, or SLS/Orion integration could derail the timeline.

Why Competitors Might Leapfrog
Blue Origin’s Blue Moon lander is modular and built on legacy systems. It’s slower but more conservative.

Dynetics’ Alpaca is smaller, but designed for simplicity and rapid iteration.

International partners (ESA, JAXA, ISRO) are developing cargo landers and surface systems that could plug into alternate architectures.

If SpaceX slips past 2027, NASA may pivot to a dual-provider model or accelerate alternate lander contracts. Musk’s strategy is to stay ahead by sheer momentum—but the margin for error is razor-thin.

In short ain`t going to happen.

[alugobi]

In short ain`t going to happen.

Can.worms.opened.

[DanClemmensen]

Interesting how space x all of a sudden released video of simplified starship lander ...
Yes—SpaceX just dropped a major update on its Starship lunar lander, revealing a simplified design

Sorry, but that's incorrect. The new information is not about a simplified Starship HLS. It describes the HLS they have been working on the whole time, and it is quite clearly more capable than is needed to meet the Artemis III requirements. There is no hint in this announcement that they are "simplifying" HLS.

NASA Acting Administrator Sean Duffy recently reopened the Artemis III lunar lander contract due to delays in SpaceX’s Starship development.

This is also incorrect. Duffy does not have the power to re-open this contract unilaterally, and there is no funding for this. He announced his intention to do something, but the details are vague.

What’s New in the “Simplified” Starship Lander?
SpaceX is now formally assessing a streamlined mission architecture that aims to:

Accelerate timelines for Artemis III (now targeting 2027–2028)

Please notice that they are proposing a new mission architecture, not a new lander. Furthermore they stated that they have discussed possible new architectures with NASA repeatedly over the last few years. The only "new" thing in this announcement is that they are now formally proposing a new mission architecture in response to Duffy's announcement.

[Jim]

In short ain`t going to happen.

 A lot of nonsense and wrong in that post.