Starship Artemis Contract & Lunar Starship

[JamesH65]

What wall? Wall implies no way forward without lots of effort. I don't see that.

Setback maybe, but no wall. Which makes the rest of the post moot.

[JIS]

C'mon. Where exactly is the SpaceX lead of 2-4 years? New Glen 7x2 is de facto finished product.

@ JIS, In your opinion, when will NG launch for the fourth time? All modern LVs have taken about two years. (Except Starship, which is too weird.)
Why do you think NG will do any better? Note that BO does not consider it to be a finished product. They have announced important changes even for NG-3.

NG is launching commercial payload and clearly demonstrated the mission profile. The next launch is lunar landing demo which takes yet another step. The capability is clearly there. The fact they are planning some small upgrades for NG-3 or NG-4 is certainly relevant and increases uncertainty but still NG is further along than Starship.

Starship hasn't demonstrated full mission profile or capability to do lunar landing yet. They are also in the middle of major upgrade of their launcher. The difference between Starship V2 and V3 is much bigger than between NG-2 and NG-3 missions. No one knows how many launches it will take for Starship to even achieve LEO insertion.

Upgrading to Starship v4 of NG 9x4 will be a major step for both companies - I'm not talking about this. I'm just trying to estimate chances for near term crewed lunar mission and from this point of view BO has bigger chances right now. IMO the reason is they have more Lunar optimised architecture while SpaceX makes little if any effort for such optimisation.

   

[JIS]

"When you get to the end zone, try to act as if you've been there before".

Apply that to the alleged current "moon race" with China.

The race is silly and promotes using out of date tech.  The new tech will be ready when it's ready.

Aren't delays on SX HLS caused because they moon proposal is low priority offshoot of the Mars plans?
 

Broadly speaking, delays in SS in general are due to just how far beyond state of the art it is.

Detractors try to spin it differently, but nothing on anyone's drawing boards today is remotely comparable.

Specifically for HLS, yes, SpaceX is prioritizing the 2026 Mars campaign over HLS demo, but that's just common sense.  Launch windows should take precedence, and HLS could fly a few months later, there's really no months level urgency.

It seems to me that the recent delays in Starship development were often caused by low quality manufacturing and fast track development.

For example delays caused by development of thermal protection or Raptor 3 are fully justifiable. Those are pushing the boundaries. But having the streak of COP vessels failures and leaks (IFT9) point to quality issues.   

So yes, they are making the progress but let's not be naive and expect those kind of failures will go away. No, they are the feature of SpaceX work and they will manifest across the development phase for years to come.

On the other side they seem to be doing well once the product development is truly finished (Falcon 9). I hope this allows them to achieve Booster and ground infrastructure high reliability soon. It would be really bad if they manage to destroy their only working launchpad. Even if they have more launchpads soon it would still be a major setback.

Blowing up Starship is just business as usual. I can imagine it makes NASA nervous about crewed HLS prospects. 

[DanClemmensen]

C'mon. Where exactly is the SpaceX lead of 2-4 years? New Glen 7x2 is de facto finished product.

@ JIS, In your opinion, when will NG launch for the fourth time? All modern LVs have taken about two years. (Except Starship, which is too weird.)
Why do you think NG will do any better? Note that BO does not consider it to be a finished product. They have announced important changes even for NG-3.

NG is launching commercial payload and clearly demonstrated the mission profile. The next launch is lunar landing demo which takes yet another step. The capability is clearly there. The fact they are planning some small upgrades for NG-3 or NG-4 is certainly relevant and increases uncertainty but still NG is further along than Starship.

Starship hasn't demonstrated full mission profile or capability to do lunar landing yet. They are also in the middle of major upgrade of their launcher. The difference between Starship V2 and V3 is much bigger than between NG-2 and NG-3 missions. No one knows how many launches it will take for Starship to even achieve LEO insertion.

Upgrading to Starship v4 of NG 9x4 will be a major step for both companies - I'm not talking about this. I'm just trying to estimate chances for near term crewed lunar mission and from this point of view BO has bigger chances right now. IMO the reason is they have more Lunar optimised architecture while SpaceX makes little if any effort for such optimisation.   

I see. You prefer to not answer my question, which was:
  "@ JIS, In your opinion, when will NG launch for the fourth time?"

[OTV Booster]

"When you get to the end zone, try to act as if you've been there before".

Apply that to the alleged current "moon race" with China.

The race is silly and promotes using out of date tech.  The new tech will be ready when it's ready.

Aren't delays on SX HLS caused because they moon proposal is low priority offshoot of the Mars plans?
 

Broadly speaking, delays in SS in general are due to just how far beyond state of the art it is.

Detractors try to spin it differently, but nothing on anyone's drawing boards today is remotely comparable.

Specifically for HLS, yes, SpaceX is prioritizing the 2026 Mars campaign over HLS demo, but that's just common sense.  Launch windows should take precedence, and HLS could fly a few months later, there's really no months level urgency.

It seems to me that the recent delays in Starship development were often caused by low quality manufacturing and fast track development.

For example delays caused by development of thermal protection or Raptor 3 are fully justifiable. Those are pushing the boundaries. But having the streak of COP vessels failures and leaks (IFT9) point to quality issues.   

So yes, they are making the progress but let's not be naive and expect those kind of failures will go away. No, they are the feature of SpaceX work and they will manifest across the development phase for years to come.

On the other side they seem to be doing well once the product development is truly finished (Falcon 9). I hope this allows them to achieve Booster and ground infrastructure high reliability soon. It would be really bad if they manage to destroy their only working launchpad. Even if they have more launchpads soon it would still be a major setback.

Blowing up Starship is just business as usual. I can imagine it makes NASA nervous about crewed HLS prospects.

Fast track development is part of the SX secret sauce. Low quality work has not been SX SOP in the past but you may have a point here. Or it may be some combo of bad luck and low quality. We'll see how it shakes out.

[meekGee]

"When you get to the end zone, try to act as if you've been there before".

Apply that to the alleged current "moon race" with China.

The race is silly and promotes using out of date tech.  The new tech will be ready when it's ready.

Aren't delays on SX HLS caused because they moon proposal is low priority offshoot of the Mars plans?
 

Broadly speaking, delays in SS in general are due to just how far beyond state of the art it is.

Detractors try to spin it differently, but nothing on anyone's drawing boards today is remotely comparable.

Specifically for HLS, yes, SpaceX is prioritizing the 2026 Mars campaign over HLS demo, but that's just common sense.  Launch windows should take precedence, and HLS could fly a few months later, there's really no months level urgency.

It seems to me that the recent delays in Starship development were often caused by low quality manufacturing and fast track development.

For example delays caused by development of thermal protection or Raptor 3 are fully justifiable. Those are pushing the boundaries. But having the streak of COP vessels failures and leaks (IFT9) point to quality issues.   

So yes, they are making the progress but let's not be naive and expect those kind of failures will go away. No, they are the feature of SpaceX work and they will manifest across the development phase for years to come.

On the other side they seem to be doing well once the product development is truly finished (Falcon 9). I hope this allows them to achieve Booster and ground infrastructure high reliability soon. It would be really bad if they manage to destroy their only working launchpad. Even if they have more launchpads soon it would still be a major setback.

Blowing up Starship is just business as usual. I can imagine it makes NASA nervous about crewed HLS prospects.

Fast track development is part of the SX secret sauce. Low quality work has not been SX SOP in the past but you may have a point here. Or it may be some combo of bad luck and low quality. We'll see how it shakes out.

Without the benefit of hindsight, how so you dial how fast should you go?

And even with hindsight, how do you evaluate the alternative path of having fewer mishaps but also going slower?

Now you need not only hindsight but also clairvoyance...

The only thing you can compare to is alternatives that were actually taken, but then other variables come into play.

So this will always remain an open issue, which is good for the owners of the forums

[steveleach]

It seems to me that the recent delays in Starship development were often caused by low quality manufacturing and fast track development.

There may be an element of this, as they figure out how to transition the industry to low-cost mass production, but I don't think it is the main cause.

I think failures are more likely to be due to the "if you are not occasionally adding things back in, then you are not deleting enough" philosophy, and that is a very deliberate leadership decision.

[JIS]

C'mon. Where exactly is the SpaceX lead of 2-4 years? New Glen 7x2 is de facto finished product.

@ JIS, In your opinion, when will NG launch for the fourth time? All modern LVs have taken about two years. (Except Starship, which is too weird.)
Why do you think NG will do any better? Note that BO does not consider it to be a finished product. They have announced important changes even for NG-3.

NG is launching commercial payload and clearly demonstrated the mission profile. The next launch is lunar landing demo which takes yet another step. The capability is clearly there. The fact they are planning some small upgrades for NG-3 or NG-4 is certainly relevant and increases uncertainty but still NG is further along than Starship.

Starship hasn't demonstrated full mission profile or capability to do lunar landing yet. They are also in the middle of major upgrade of their launcher. The difference between Starship V2 and V3 is much bigger than between NG-2 and NG-3 missions. No one knows how many launches it will take for Starship to even achieve LEO insertion.

Upgrading to Starship v4 of NG 9x4 will be a major step for both companies - I'm not talking about this. I'm just trying to estimate chances for near term crewed lunar mission and from this point of view BO has bigger chances right now. IMO the reason is they have more Lunar optimised architecture while SpaceX makes little if any effort for such optimisation.   

I see. You prefer to not answer my question, which was:
  "@ JIS, In your opinion, when will NG launch for the fourth time?"

I know very little about NG manufacturing schedule. I'm truly shocked anyone would ask for my opinion on this topic. But I have no problem to answer if you insist. They plan the third launch early in 2026, 2 months after the 2nd launch. So I would expect the fourth launch will follow few months after the 3rd one. Pending the status of recovered boosters and outcome of 3rd launch.

My impression is that all necessary manufacturing and launch infrastructure is up and running, they have healthy human and financial resources, good enough product, big backlog, so it makes sense to ramp up the production.       

[envy887]

Fast track development is part of the SX secret sauce. Low quality work has not been SX SOP in the past but you may have a point here. Or it may be some combo of bad luck and low quality. We'll see how it shakes out.

SpaceX is simultaneously doing component development and qualification, all-up system validation, and manufacturing and operations cadence ramp. These processes are normally completed sequentially, trading lower development cost for a longer (and less spectacular) overall process.

The downsides of doing development this way is that it costs more to blow full articles (but SpaceX operates relatively cheaply anyway), and it airs all the dirty secrets of development. The upsides are that the overall process goes quicker as tasks are completed in parallel, and, more importantly, they will come out of development with manufacturing and operations already cranking away near maximum cadence and capacity.

And that means the steamroller is already gathering momentum - even though it's not completely designed yet, it's already gone through the period of several years of cadence ramp that large launch vehicles go through to reach 10+ flights a year. 2025 is probably the last year that Starship will have fewer than 10 flights. 2026 is probably the last that it will have fewer than 40.

[DanClemmensen]

I see. You prefer to not answer my question, which was:
  "@ JIS, In your opinion, when will NG launch for the fourth time?"

I know very little about NG manufacturing schedule. I'm truly shocked anyone would ask for my opinion on this topic. But I have no problem to answer if you insist. They plan the third launch early in 2026, 2 months after the 2nd launch. So I would expect the fourth launch will follow few months after the 3rd one. Pending the status of recovered boosters and outcome of 3rd launch.

My impression is that all necessary manufacturing and launch infrastructure is up and running, they have healthy human and financial resources, good enough product, big backlog, so it makes sense to ramp up the production.     

Thanks. The reason I asked is that you seemed to be making predictions about NG based on the schedule. There is nothing particularly wrong with this, and I tend to do the same thing about other rockets. I began to be intensely interested in launches about 5 years ago, and I innocently believed the schedule estimates, like Artemis III on 2024.

You think the fourth NG will launch in 18 to 23 months after the first one. If they do, I will be very impressed. That will beat every medium or larger new orbital rocket in the last 30 years (except Starship, depending on how you count it). Vulcan is a good recent example. Basically, Stuff happens and causes schedule slips.

I think NG1 underperformed and BO intends to make modifications, as happened with Starship. This takes time. I think this will push NG3 to NET Q2 2026 and NG4 out past January 2027.

[JIS]

I see. You prefer to not answer my question, which was:
  "@ JIS, In your opinion, when will NG launch for the fourth time?"

I know very little about NG manufacturing schedule. I'm truly shocked anyone would ask for my opinion on this topic. But I have no problem to answer if you insist. They plan the third launch early in 2026, 2 months after the 2nd launch. So I would expect the fourth launch will follow few months after the 3rd one. Pending the status of recovered boosters and outcome of 3rd launch.

My impression is that all necessary manufacturing and launch infrastructure is up and running, they have healthy human and financial resources, good enough product, big backlog, so it makes sense to ramp up the production.     

Thanks. The reason I asked is that you seemed to be making predictions about NG based on the schedule. There is nothing particularly wrong with this, and I tend to do the same thing about other rockets. I began to be intensely interested in launches about 5 years ago, and I innocently believed the schedule estimates, like Artemis III on 2024.

You think the fourth NG will launch in 18 to 23 months after the first one. If they do, I will be very impressed. That will beat every medium or larger new orbital rocket in the last 30 years (except Starship, depending on how you count it). Vulcan is a good recent example. Basically, Stuff happens and causes schedule slips.

I think NG1 underperformed and BO intends to make modifications, as happened with Starship. This takes time. I think this will push NG3 to NET Q2 2026 and NG4 out past January 2027.

Vulcan had a problem with solid motors and its next launch is for DoD which is very paranoid client. So perhaps this can explain slow ramp up.

NG had two "flawless" missions and the next one is their private launch. Yes, there could be some delays or problems which push out NG4. But underperforming NG makes no sense. They probably derated BE little bit for the first few launches but I haven't seen any indication it should hold them up.

In contrast SpaceX is having some issues all the time and they just launch anyway. In my opinion they should have ramped up infrastructure faster and should have launched even obsolete boosters and ships in parallel with building new ones. They keep claiming ramping up production for years but where is the hardware? Scrapped. 

For example if they kept launching V2 starships it would allow them to test on-orbit loiter and refueling much sooner. Now they have delays due to switching to V3 and Raptor 3 architecture. I think the reason for this step is that SpaceX is keen to discontinue Raptor 2 with its dirty autogenous pressurisation which makes on-orbit refueling problematic. Maybe they found out that solid H2O and CO2 makes refueling too hard? Still, they could have at least tested different aspects of on-orbit operation. I have to admit that I never understood their decision to implement this feature at first place. They lost several ships because of this and retired V2 immediately after they managed to make it somewhat work at last.

But anyway, the decision was made and it is their money they are spending. Doing it just for the sake of Artemis 3 mission schedule will not pay their bills. 

   

[DanClemmensen]

You think the fourth NG will launch in 18 to 23 months after the first one. If they do, I will be very impressed. That will beat every medium or larger new orbital rocket in the last 30 years (except Starship, depending on how you count it). Vulcan is a good recent example. Basically, Stuff happens and causes schedule slips.

Vulcan had a problem with solid motors and its next launch is for DoD which is very paranoid client. So perhaps this can explain slow ramp up.

That's the point. Five out of the last five new big rockets "had a problem", all different. It's the unknown or unexpected failures that get you. ULA could choose to launch Kuiper (LEO) instead of that GPS satellite.

NG had two "flawless" missions...

No. NG1 failed to land successfully, even though BO said it was necessary for their business plan. We can all hope that NG does not encounter an unknown unknown.

[sstli2]

This is an interesting discussion, but the wrong thread for it. Try the New Glenn thread? In fact, I'll quote you and respond there: https://forum.nasaspaceflight.com/index.php?topic=61201.msg2739532#msg2739532

[OTV Booster]

"When you get to the end zone, try to act as if you've been there before".

Apply that to the alleged current "moon race" with China.

The race is silly and promotes using out of date tech.  The new tech will be ready when it's ready.

Aren't delays on SX HLS caused because they moon proposal is low priority offshoot of the Mars plans?
 

Broadly speaking, delays in SS in general are due to just how far beyond state of the art it is.

Detractors try to spin it differently, but nothing on anyone's drawing boards today is remotely comparable.

Specifically for HLS, yes, SpaceX is prioritizing the 2026 Mars campaign over HLS demo, but that's just common sense.  Launch windows should take precedence, and HLS could fly a few months later, there's really no months level urgency.

It seems to me that the recent delays in Starship development were often caused by low quality manufacturing and fast track development.

For example delays caused by development of thermal protection or Raptor 3 are fully justifiable. Those are pushing the boundaries. But having the streak of COP vessels failures and leaks (IFT9) point to quality issues.   

So yes, they are making the progress but let's not be naive and expect those kind of failures will go away. No, they are the feature of SpaceX work and they will manifest across the development phase for years to come.

On the other side they seem to be doing well once the product development is truly finished (Falcon 9). I hope this allows them to achieve Booster and ground infrastructure high reliability soon. It would be really bad if they manage to destroy their only working launchpad. Even if they have more launchpads soon it would still be a major setback.

Blowing up Starship is just business as usual. I can imagine it makes NASA nervous about crewed HLS prospects.

Fast track development is part of the SX secret sauce. Low quality work has not been SX SOP in the past but you may have a point here. Or it may be some combo of bad luck and low quality. We'll see how it shakes out.

Without the benefit of hindsight, how so you dial how fast should you go?

And even with hindsight, how do you evaluate the alternative path of having fewer mishaps but also going slower?

Now you need not only hindsight but also clairvoyance...

The only thing you can compare to is alternatives that were actually taken, but then other variables come into play.

So this will always remain an open issue, which is good for the owners of the forums

It's always a call shot. Weaving a path through the future is the most important responsibility of leaders.

Nothing is ever 100%. Mistakes will happen. It's when the mistakes pile up that leadership absolutely must display the qualities that make them leaders, reassess their own performance and not push blame downward. If they don't they are no longer leaders and have become managers at best. This is true even if mistakes end up each being a one chance in a million frown of the gods that absolutely could not have been foreseen.

I have zero insight either way as to how SpaceX is handling this.

On reread, this comes across as an anti-Elon screed. That was not my intention but if the shoe fits...

For further reading on leadership failures check out recent Boeing history and Space Shuttle LOC.

[OTV Booster]

You think the fourth NG will launch in 18 to 23 months after the first one. If they do, I will be very impressed. That will beat every medium or larger new orbital rocket in the last 30 years (except Starship, depending on how you count it). Vulcan is a good recent example. Basically, Stuff happens and causes schedule slips.

Vulcan had a problem with solid motors and its next launch is for DoD which is very paranoid client. So perhaps this can explain slow ramp up.

That's the point. Five out of the last five new big rockets "had a problem", all different. It's the unknown or unexpected failures that get you. ULA could choose to launch Kuiper (LEO) instead of that GPS satellite.

NG had two "flawless" missions...

No. NG1 failed to land successfully, even though BO said it was necessary for their business plan. We can all hope that NG does not encounter an unknown unknown.

That's a bit unfair. SX lost quite a few boosters before they got a landing to work.  The booster was always lost *after* the payload was delivered so in that sense the launch was successful and the loss was a post mission development hiccup.


Blue nailed landing on the second try which is commendable, as is a successful first launch.


Blue has been the butt if many jokes about their development speed - all of them well deserved. It's dicy to extrapolate a trend from two data points but maybe they've turned a corner with their new leadership.

[envy887]

In my opinion they should have ramped up infrastructure faster and should have launched even obsolete boosters and ships in parallel with building new ones. They keep claiming ramping up production for years but where is the hardware? Scrapped. 

For example if they kept launching V2 starships it would allow them to test on-orbit loiter and refueling much sooner.

They are ramping up hardware production, as they rolled out 5 new boosters and 6 new ships in the last year. Only 1 of those was scrapped, and that was B17 which had no ships to boost because SpaceX succeeded at reusing two other boosters.

They could have built and launched more v2 ships, but to what point? The v2 ships don't have docking hardware, so they can't test refueling. There's no reason to believe that adding docking hardware to v2 would be faster than simply moving on to v3.

[xvel]

B17 was scrapped because S36 blow up, stop making things up

[uhuznaa]

In my opinion they should have ramped up infrastructure faster and should have launched even obsolete boosters and ships in parallel with building new ones. They keep claiming ramping up production for years but where is the hardware? Scrapped. 

For example if they kept launching V2 starships it would allow them to test on-orbit loiter and refueling much sooner.

They are ramping up hardware production, as they rolled out 5 new boosters and 6 new ships in the last year. Only 1 of those was scrapped, and that was B17 which had no ships to boost because SpaceX succeeded at reusing two other boosters.

They could have built and launched more v2 ships, but to what point? The v2 ships don't have docking hardware, so they can't test refueling. There's no reason to believe that adding docking hardware to v2 would be faster than simply moving on to v3.

Well, the main point is that they're pressing for reusability.

They easily could have continued with v2 to develop and test orbital operations, tankers, docking and prop transfer, if needed with expendable ships and tankers if there would not have been sufficient payload with all the reusability mass penalties.

Of course this would have just pushed the unavoidable switch to v3 into the future but getting further along sooner with all the other objectives would have looked better. That's what many people seem to be concerned with especially in comparison to NG. Which flew two successful missions, but of course with an expended upper stage this is much simpler to do. Starship v2 with no landing propellants, header tanks, heat shield, flaps etc. could get close to 100 tonnes to orbit I guess.

Or in other words: SpaceX could be much further along if they would just have continued with v2 and an expended second stage, but this is just not what they want to do anyway and they aren't especially concerned with how going for v3 without even getting into a full orbit with v2 may look to others.

[DanClemmensen]

B17 was scrapped because S36 blow up, stop making things up

That's valid but perhaps oversimplified. The Massey's mess was a lot worse than the loss of S36. It caused SpaceX to decide to stop V2 production and shift everything to V3, perhaps earlier than they really wanted to. That in turn made B17 obsolete and unusable.

If S36 had just destroyed itself but with minimal collateral damage (like B18), SpaceX might have just built another V2 Ship and flown it with B17 from Pad 1.

[JIS]

B17 was scrapped because S36 blow up, stop making things up

That's valid but perhaps oversimplified. The Massey's mess was a lot worse than the loss of S36. It caused SpaceX to decide to stop V2 production and shift everything to V3, perhaps earlier than they really wanted to. That in turn made B17 obsolete and unusable.

If S36 had just destroyed itself but with minimal collateral damage (like B18), SpaceX might have just built another V2 Ship and flown it with B17 from Pad 1.

Correct. They made decision to stop recovering and reusing V2 boosters and to scrap B17. They made decision to stop producing V2 starships. Even though they claimed they can do 25 launches in 2025, apparently without reuse of Starship, probably just booster reuse.

Why Musk made such decision? They had nothing more to learn from flying V2? C'mon, they haven't gone to orbit yet. They haven't attempted any orbital loiter longer than 30 minutes. They haven't done any real payload deploy, any docking test, starship recovery, further heat shield testing, recovery and reuse. All of that could have been done with V2. They threw it away for the sake of a new, shiny and "perfect" V3, which struggles to get produced, tested and launched.

Was it all because they blew up Masseys test site? I can remember that someone here tried to convince me that having more redundant and higher quality (resilient) infrastructure would not speed up starship development. 

Maybe there was different reason for the switch. Maybe it was all just the mantra to iterate fast. Do some quick design, build the test article, test it, find out what is wrong, improve the design little bit, build new test article and test it.

Musk doesn't really has any elaborate plan for testing and development. He wants to put a big cargo to Mars and decided to build a big rocket in many iterations. He is not wasting time by developing the details. He seems to be doer, not thinker. That would probably explain what we see happening in SpaceX from Falcon1, Falcon5, Falcon9 of many blocks, FH and loads of Starship versions. Maybe Falcon 9 is actually very lucky that Musk find a new toy for himself (Starship).