NASA's Artemis Program Updates and Discussion Thread 3

[spacenut]

Musk is more pragmatic than the NASA and can turn his company on a dime or a moments notice.  He is bound and determined go to Mars the cheapest with the least hassle possible.  Starship is much cheaper to build and probably to operate than Shuttle ever was.  Musk also is not doing nuclear to Mars because of the government hoops he would have to jump through. 

Starship, they way it is built, can do a lot.  Without re-entry, it can become an in space workhorse.  The engines are cheap to build.  One RS-25 or SSME is $100 million.  Raptor with similar thrust, Musk said is around $2 million each.  There are no heavy solids to deal with.  Starship can have stretched fuel tanks and become a tanker.  With habitats on board, becomes a human transporter.  The large fairing space, it becomes a heavy lifter for anything you want in space. 

There is really no comparison to Shuttle.  Shuttle had wings which limited its cargo carrying capacity to a certain width and length.  Shuttle had a disposable fuel tank.  Solids weren't as easy to refurbish as originally thought.  Falcon 9 has proven liquids are better for reusability with essentially the same payload capability.  Starship/Superheavy will be completely reusable and is designed from the get go to be easier to launch and re-use. 

Now, for the Artemis program, Starship can be the reusable lander and have tankers bring fuel for it.   It is large enough to be the habitat while astronauts are on the moon.  With several lunar Starships, tons of equipment and supplies could be brought to a moon base. 

[VSECOTSPE]

SS is developing variants on one... design to address different uses/markets.

Actually the opposite, SpaceX is developing ONE design... that will only have three variants with rather as 'minimal' of design changes as possible.

Not the opposite.  You just repeated what I had written.

that I find hard to assume will be 'cheap' when all we have to go by is non-operational, non-prototypes which have very little to do with actually working orbital launch vehicles

We don’t put TPS or life support on a test article that will only ever fly uncrewed to some tens of kilometers in the same way that we don’t put Bose speakers and a fine leather interior in a car with crash test dummies. It’s a useless waste.

Do the addition of TPS, life support, more redundancy, and other system improvements have the potential to increase SS costs?  Obviously.  Will they send SS costs through the roof?  Probably not given the SX record, which so far has always produced LVs, capsules, and satellites that cost a fraction of what’s come before.  Is it guaranteed that the operational SS vehicles will cost more than the current test articles?  No, the scale SS achieves in production and reusability may offset the additional cost of the operational systems.

SS has to have just about as many flights as the STS was requiring to meet similar economic goals.

Economic return or break-even on SS costs is not the goal.  If the SS program safely lands a bunch of people and their stuff on Mars, SX has achieved its reason for existence.  Full stop.  Musk can blow $10B (~1/15th) of his net worth on SS, never see a dime back, and still get what he wants out of SS.

SS could do nothing else but launch StarLink satellites more cheaply than F9, and SX, Musk, and their Mars goals would still benefit.

SS isn’t measured by STS economics (or lack thereof).  Even STS wasn’t ultimately measured by its economics.  As long as it didn’t kill too many astronauts and eventually got the ISS up, the USG was willing to pay for that national symbol.

As per usual neither Congress nor OMB were willing to buy enough Orbiters to come close to the needed flight rate

STS never had the operational tempo needed to meet the old Mitre study or any other economic projection.  Investing in more orbiters would have just made the economic case worse.

I've seen on these very forums someone 'crow' that ONE cargo/SS could fly off the entire planned annual launch inventory in a single month... And then not understand the follow of question of "Ok, then what?"
The idea that the launch market will expand to fill such capability lacks an understanding of the actual launch market and the economics involved

This is a straw man argument.  Unlike STS (or VentureStar), SS isn’t planned around driving every US payload onto it (or capturing the entire worldwide launch market).  SS just has to get people/stuff to Mars and, ideally, reduce StarLink costs.  Everything after that — DearMoon, Artemis, non-StarLink satellite launches, P2P, etc. — just helps SX recoup some of Musk’s investment in SS.

I actually don't see anything to support this view point as SS is based on much the same economic and launch market theory that the STS was

That’s just not true.  No one in the USG has claimed that every US payload and launch has to be driven onto SS.  No one is doing that like we did with STS.  That’s a false argument.

STS was expensive primarily because it was a government program spread over a large portion of the United States and a very ambitious first attempt at a reusable space vehicle. It was arguably too large and complex for the launch market it was aimed at, (which can be argued about Starship as well), and aimed at doing too many things with one system but it had a lot of potential that was also unrealized.

STS was a poorly conceived program, period.  As the CAIB report concluded, STS was an experimental program masquerading as an operational one.  That’s what made it so bloody intensive, expensive, and dangerous to operate.  Full stop.

In typical SX fashion, SS is getting tested to hell and back.  Whatever SS becomes, it will not be experimental like STS.

Starship at first glance does not appear to have similar flaws but in fact it has all the same flaws and is just as dependent on government financing

This is again just false.  SS development and testing has been, and continues to be, ongoing with no support from the USG.  When the HLS award eventually flows, it will only pay for a fraction of SS development.

People wonder why there are no 'Starship" competitors but that's like asking why there were originally no STS or EELV "competitors".

Apples and oranges.

There were no STS competitors because the USG dictated so and started phasing out the old Atlas/Delta/Titan programs.  Aside from HLS requirements, the USG is dictating nothing with respect to SS.

DOD based EELV on a LEO comsat constellation market that was outside its control.  StarLink is a LEO comsat constellation that is controlled by SX.

Similarly Starship is based on a lot of assumptions and numbers that COULD apply as easily as they could not and are just as valid as not.

Every engineering project is based on assumptions and numbers that could turn out to be wrong.  But SS is not based on assumptions and numbers that only exist in your head.

[Zed_Noir]

Wonder what NASA's response will be. If SpaceX just add the cosine thrusters to a crew Starship with a bit more propellants and offered that modified Starship design as an alternate HLS lander with a drop in price.


Presumed there will be hatch covers over the cosine thrusters for atmospheric reentry back to Earth.

[su27k]

This will essentially infuse a bunch of public money into a singular private space launch system which in and of itself then becomes perilously close to if not in effect a government subsidized monopoly. As such it stifles competition and since it is subsidized by the government there is vastly less incentive to do anything about it. Due to it's size Starship, like the Shuttle, requires a hefty segment if not domination of the overall launch market to be economic. It can't help it and reusability actually encourages this to keep the price down.

Gee, where have I seen this before, let me see... Does it sound awfully like a company called ULA???

This whole rant about SpaceX/Starship becoming a monopoly makes no sense (and also off-topic). First of all, the example of ULA shows: a. USG is not avert to creating a launch monopoly if it suited their goals; b. Even with a USG created monopoly, it didn't "stifles competition", because... SpaceX.

SpaceX is the perfect example that there can still be fierce competition even with a government established monopoly in place. They fought the monopoly and they won, so the concern about SpaceX themselves becoming a monopoly and stifle competition is not valid. It does require potential competitors to fight equally hard, if not more so, than SpaceX, as it should be, that's what competition is all about. If you want to dislodge SpaceX, you have to be more innovative and work harder than them. If you can't and lose, that's on you, not SpaceX.

Finally, USAF/USSF has a standing policy to support at least two launch providers, this alone ensures SpaceX wouldn't be a monopoly.

[su27k]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version, I mean NASA even has their own version for Shuttle transport, and for carrying telescope, and of course another version is being used by Virgin Orbit to launch rockets. It all worked out very well, probably because Boeing has a production line pumping these out like sausages, and each variant is specialized for its particular role without needing to worry about other roles, this disentangle the dependencies.

In other words, the difference between 747 variants and Shuttle is this: When you ask 747 to be Shuttle transporter, a flying telescope and a rocket launch platform, you bring in 3 separate 747 aircrafts and each is modified to do its thing and only its thing. Shuttle is like refitting a single 747 to carry Shuttle, carry telescope and launch rocket all at the same time, that's where it went south.

[soyuzu]

They are: Tanker, Cargo and mixed Passenger/Cargo. Developing anything ELSE gets expensive and as I said even if the customer demands it there's actually not a lot of incentive to go that way. Contracts or no, (and keep in mind the HLS is NOT as specific as people tend to think it is otherwise Starship, no matter how cheap, would not have even been considered) the monetary penalties are likely be trivial for 'over-performance' than 'under-performance'. Especially as there is likely to be no 'backup plan' in place.
...
I'd argue that what we'll see for "HLS" will simply be a "Starship" prototype, (and actual one) with a lot of low-complexity systems packed in to meet the minimum requirements. i don't know if people have been paying attention but SpaceX illustrations have already dumped the 'lunar landing engines' and I fully expect that SpaceX will argue they aren't needed so they won't have to 'compromise' the Starship design. I won't be at all surprised if SpaceX uses the NASA money to finally get serious about a actual working landing leg system but it will be 'compatible' with the "standard" Starship not something specific for the Moon.
time.
...

STS tried to do too much in one (very expensive) design.  SS is developing variants on one (relatively cheap) design to address different uses/markets.

Actually the opposite, SpaceX is developing ONE design, (that I find hard to assume will be 'cheap' when all we have to go by is non-operational, non-prototypes which have very little to do with actually working orbital launch vehicles) that will only have three variants with rather as 'minimal' of design changes as possible. SS has to have just about as many flights as the STS was requiring to meet similar economic goals. As per usual neither Congress nor OMB were willing to buy enough Orbiters to come close to the needed flight rate but in a similar fashion SpaceX has to have a certain 'minimum' number of SS/SH's AND fly them often enough to be economical.

Like every LV, SS will have problems.  But it’s not repeating that particular STS mistake.

I actually don't see anything to support this view point as SS is based on much the same economic and launch market theory that the STS was with arguably less basis to rest on. STS was expensive primarily because it was a government program spread over a large portion of the United States and a very ambitious first attempt at a reusable space vehicle. It was arguably too large and complex for the launch market it was aimed at, (which can be argued about Starship as well), and aimed at doing too many things with one system but it had a lot of potential that was also unrealized. Starship at first glance does not appear to have similar flaws but in fact it has all the same flaws and is just as dependent on government financing and support rather than the existing launch market and relevant economic factors.

To sum up, “Because I, as knowledgeable as a Rocket scientist, believe SpaceX will try to merge all requirements into basically one design, even though this is directly contradict with what they announced they will do at multiple circumstances, and because I misunderstood that SpaceX has ditched the landing engine of LSS and don’t see that SpaceX is building the first orbital vehicles using the same method as current “non-prototypes”, I believe SpaceX will repeat the same mistakes of STS for features SpaceX plans to avoid ”

Pretty satiric, as I have seen most people in Chinese forums believe LSS will be made out of Al-Li, while Reddit is discussing SpaceX will ditch the first Superheavys into ocean.

You are not talk about Starship, you are talking about an imaginary vehicle that you believe is Starship, which is actually not.

[RanulfC]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

SpaceX does not need to do everything. They can deliver a Starship cargo section to NASA. NASA builds a habitat to their own specs into it and SpaceX plants it on a propulsion section and sends it off.

You'd be correct for any 'standard' launch service but that's exactly the opposite of what SpaceX is aiming for. With Starship they plan from going from "Launch Services" to "Transport System" which is arguably a sea-change in that the 'service' will no longer be driven by customer requirements but by the transport system capability.
(Arguably btw that's a GOOD thing )

In the context here, SpaceX tells NASA what the capability is of the Starship cargo bay, NASA then builds a habitat that will fit into those requirements and ships it to the launch site. SpaceX puts it in the cargo bay and then delivers it to the destination. NASA can 'want' a bigger or more capable hab but if it doesn't "fit" it doesn't fly. Again this transforms space access from custom to mass production as it were. So far so good.

The downside here is that unlike standard space launch a transport system REQUIRES a certain steady volume of transported good because unused 'space' is lost money. You lose things like 'secondary' or 'ride along' payloads because instead of a 'single' customer ALL customers have to load share the price. Again all well and good but the load market has to be there in advance because it is very much not a "build it and they will come" type market.

Randy

[RanulfC]

Musk is more pragmatic than the NASA and can turn his company on a dime or a moments notice.

So? NASA is a government agency beholden to and answerable to Congress and the US public. SpaceX is beholden to Musk but also to the market and economic forces. Musk can 'want' to pivot on a dime but he's as bound by those factors as any other company.

He is bound and determined go to Mars the cheapest with the least hassle possible.

Yes he is and that tends to blind him and make him focus on factors that are NOT related to the market and economics. Having a vision is fine, but if it causes you to run your company into the ground or fling yourself down the wrong path...

Starship is much cheaper to build and probably to operate than Shuttle ever was.

Possible but we don't know that yet as it hasn't been flown or the actual economics figured out. We'll know more one we get to see and examine an actual Starship/Superheavy and their operations. The Shuttle never met it's proposed economic and operational goals due to a number of factors but what it did it actually did quite well and as a system it had a lot of untapped potential. Still you can't really compare a government and commercial system as they are designed to and for different goals.
 
[quoteStarship, they way it is built, can do a lot.[/quote]

Proposed, not built. It hasn't been built yet so we don't really know.

Without re-entry, it can become an in space workhorse.

Arguably but not really as currently proposed. And not as much as you might think as more optimized platform would probably be a better option. Starship is proposed to really be 'optimized' for Earth surface-to-orbit transport with a goal of eventually getting to Mars. The problem is that getting to and from Mars may not be as plausible with Starship given it needs a LOT of 'extras' that would hamper the surface-to-orbit mission.

The engines are cheap to build.  One RS-25 or SSME is $100 million.  Raptor with similar thrust, Musk said is around $2 million each.

"Musk said" There's arguments that the number doesn't actually work but we probably won't ever really know because Musk/SpaceX won't sell them just like they won't sell Merlin's.

Starship can have stretched fuel tanks and become a tanker.  With habitats on board, becomes a human transporter.  The large fairing space, it becomes a heavy lifter for anything you want in space.

Stretched propellant tanks are actually cheap, everything ELSE is expensive and has to be juggled with capacity and capability Standard (mixed passenger/cargo) has to be designed and built for that job so does the cargo version. That's going to mean compromise and complexity and that's JUST the Earth orbital version.

There is really no comparison to Shuttle.  Shuttle had wings which limited its cargo carrying capacity to a certain width and length.

Actually no, the wings allowed a specific mass for the cargo whereas length and width were by requirement during design. Starship ALSO is limited by payload bay length and width as well as mass so not dissimilar at all

Shuttle had a disposable fuel tank.  Solids weren't as easy to refurbish as originally thought.

Again the design choices are not comparable because the Shuttle was designed to fit "requirements" other than just directly mission related.

Falcon 9 has proven liquids are better for reusability with essentially the same payload capability.

Again not really 'comparable'. Arguably at the time the Shuttle was designed it was well understood that liquid rockets were better for reusability. That's why the Shuttle use them In fact the idea was to use liquids both in the orbiter and the booster but economics prevented this. This is why almost all 'reusable' concepts have used liquid propellant. Falcon 9 has had ZERO influence on this thinking it's been around since the 40s. Lets not over-blow accomplishments here

Starship/Superheavy will be completely reusable and is designed from the get go to be easier to launch and re-use.

And we'll see how that works out, but that's not proven in anyway and arguably there are some serious operational questions that appear to make that argument less likely rather than more. 

Now, for the Artemis program, Starship can be the reusable lander and have tankers bring fuel for it.   It is large enough to be the habitat while astronauts are on the moon.  With several lunar Starships, tons of equipment and supplies could be brought to a moon base.

In theory if Starship works mostly as advertised, yes. On the other hand it is quite arguably that you could do the same with a much smaller and cheaper fully reusable system than Starship/Superheavy flying more often. (Yes that actually helps reduce the cost) But, as well all know Starship/Superheavy isn't DESIGNED for the Moon or even going to Earth orbit on a regular basis The thing is at this point we don't KNOW any of this and therefore the 'comparison' is fully speculative.

It is likely cheaper than either of the other two lander concepts but as a whole that's actually NOT really a main consideration. ("Cost" is due to the limited budget but that also isn't 'economic' or market driven but politics which is all I'll say)

Randy

[rakaydos]

Musk is more pragmatic than the NASA and can turn his company on a dime or a moments notice.

So? NASA is a government agency beholden to and answerable to Congress and the US public. SpaceX is beholden to Musk but also to the market and economic forces. Musk can 'want' to pivot on a dime but he's as bound by those factors as any other company.

Wrong. SpaceX is privately held, not publicly held. And all his private investors are on board with his mars goal, so there's no threat of "rights of the shareholders" actions if he pursues mars related goals.

[Yggdrasill]

i don't know if people have been paying attention but SpaceX illustrations have already dumped the 'lunar landing engines' and I fully expect that SpaceX will argue they aren't needed so they won't have to 'compromise' the Starship design.

Where do you have that from? The most recent information seems to be that they have moved from a small number of larger landing engines to a larger number of smaller landing engines. They can be seen right under the solar panels:

Actually the opposite, SpaceX is developing ONE design, (that I find hard to assume will be 'cheap' when all we have to go by is non-operational, non-prototypes which have very little to do with actually working orbital launch vehicles) that will only have three variants with rather as 'minimal' of design changes as possible.

SN20 is basically a working orbital launch prototype. The only thing it may or may not have that impacts that assessment is a payload door.

I'm pretty sure there will be five variants of Starship. Cargo, tanker, lunar mixed passenger/cargo, mars/earth mixed passenger/cargo and passenger (earth point-to-point). And I expect they will be developed in that order.

With some luck they will be able to put a cargo variant on a Mars trajectory in 2022.

[MaNaeSWolf]

Actually the opposite, SpaceX is developing ONE design,. . . . . . that will only have three variants with rather as 'minimal' of design changes as possible.

Modern rockets are not designed like this. Rockets are made from sets of components. Those components are designed to work together, but you can often chose to add, remove or change components depending on the mission criteria.

The Delta IV has variants with different side booster, upper stages, fairings ext. SpaceX is designing an extended F9 fairing. You can fly the F9 with or without legs. This ability to change as customers needs do are baked in all launch vehicle architectures, which are made of different components. The variants are the end result.

We already know that SS will have 2 fairing variants. We know that there will be some variants that can land on rough terrain, and some that cant. To remove re-entry elements is trivial. Starship is not one thing, its an Architecture, that allows for many things, just like all rockets. The biggest difference between starship and other rockets, is one of its "components" allows for 2nd stage recovery, which is baked into the initial design.

The components for a starship can be seen as something like this
- Raptor engines (SL + Vac + ??)
- Smaller methane thrusters (which will be needed for orbital refueling and lunar landing)
- 9m diameter 1st stage
- 9m diameter 2nd stage
- Fairing (17m + 22m)
- - Chomper
- - Tanker
- - Sealed fairing for human occupation (this will be different for LEO, Mars, Lunar, Science, Space Station)
- - What ever anyone wants to pay for
- Recovery elements
- - Flaps (could have increased size for Mars landing, or none at all)
- - TPS (make them thicker for different re-entry requirements, or remove them if not needed)
- - Header tanks (not needed without a flip maneuver)
- - Avionics
- - all the other bells and whistles
- Electrical generation (solar PV, fuel cells, ??)
- Heat rejection (Radiators, necessary for cryo-storage and ECLSS)
- ext, ext, ext


Its not really a question of how many variants there will be, it more a question of how the components play together. You design components that work together to achieve various goals, they end up being these different variants. Same as how ULA is developing SMART to put on the Vulcan, they don't design a new rocket or "variant", its just a component.
If SpaceX feels there is a demand to have a

SpaceX will absolutely make a dedicated Lunar lander because
A) the cost of doing so are not that great, as they are already working on the components, that make a dedicated lunar lander necessary. HLC pays for some of these development costs. Its a win, win.

B) A lunar lander is not compatible with either a LEO human transporter or a Mars rocket.
Thermal and electrical management on the moon will be vastly different than a Mars lander.
The lunar landers are only needed for 2 weeks, or, just long enough to survive a lunar day in full sunlight. Mars lander will be in full sunlight for 4-5 months, then experience day/night cycles. Just the difference in life support requirements is immense.
Heat shields and flaps are not needed for the moon, but they are on Mars. Landing on the moon teaches you about nothing of landing on Mars, except maybe the last 50m.

SS has to have just about as many flights as the STS was requiring to meet similar economic goals. As per usual neither Congress nor OMB were willing to buy enough Orbiters to come close to the needed flight rate but in a similar fashion SpaceX has to have a certain 'minimum' number of SS/SH's AND fly them often enough to be economical.

SS will have about 10-20 flights a year just to keep starlink going. This means the flight rate is already high enough to keep the cost down. The Lunar lander shares so many common components to the SS, that you could see it as the same vehicle, even though it looks different. The only reasons why a lunar lander would cost substantially more is if there are specialized manufacturing elements that cant share a production line with the other components.

[Khadgars]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version, I mean NASA even has their own version for Shuttle transport, and for carrying telescope, and of course another version is being used by Virgin Orbit to launch rockets. It all worked out very well, probably because Boeing has a production line pumping these out like sausages, and each variant is specialized for its particular role without needing to worry about other roles, this disentangle the dependencies.

In other words, the difference between 747 variants and Shuttle is this: When you ask 747 to be Shuttle transporter, a flying telescope and a rocket launch platform, you bring in 3 separate 747 aircrafts and each is modified to do its thing and only its thing. Shuttle is like refitting a single 747 to carry Shuttle, carry telescope and launch rocket all at the same time, that's where it went south.

While I understand the point you are trying to make with 747, I don't think that really applies.  You don't have to add thrusters or additional engines to carry the shuttle or transport telescopes and the take off and landing are essentially unchanged.

Moonship will have thrusters that Starship won't have and will operate in very different environment for extended amount of time on the moons surface.  Mars Starship will have an ECLSS beyond anything we have ever seen and will land in one of the most punishing environments humans will attempt in our life times.

All three of those variants are quite different from tanker starship which will be the largest re-fueling spacecraft ever produced. 

I am surprised so many people believe certifying all of these will be cheap and easy.  As a lowly program manager, I see extensive risk associated with all of these with most of the issues we won't even know exist until we're deep into the program.  SpaceX ethos of rapid build/test/repeat definitely helps in this regard but the task at hand is monumental.

[rakaydos]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version, I mean NASA even has their own version for Shuttle transport, and for carrying telescope, and of course another version is being used by Virgin Orbit to launch rockets. It all worked out very well, probably because Boeing has a production line pumping these out like sausages, and each variant is specialized for its particular role without needing to worry about other roles, this disentangle the dependencies.

In other words, the difference between 747 variants and Shuttle is this: When you ask 747 to be Shuttle transporter, a flying telescope and a rocket launch platform, you bring in 3 separate 747 aircrafts and each is modified to do its thing and only its thing. Shuttle is like refitting a single 747 to carry Shuttle, carry telescope and launch rocket all at the same time, that's where it went south.

While I understand the point you are trying to make with 747, I don't think that really applies.  You don't have to add thrusters or additional engines to carry the shuttle or transport telescopes and the take off and landing are essentially unchanged.

Moonship will have thrusters that Starship won't have and will operate in very different environment for extended amount of time on the moons surface.  Mars Starship will have an ECLSS beyond anything we have ever seen and will land in one of the most punishing environments humans will attempt in our life times.

All three of those variants are quite different from tanker starship which will be the largest re-fueling spacecraft ever produced. 

I am surprised so many people believe certifying all of these will be cheap and easy.  As a lowly program manager, I see extensive risk associated with all of these with most of the issues we won't even know exist until we're deep into the program.  SpaceX ethos of rapid build/test/repeat definitely helps in this regard but the task at hand is monumental.

Thrusters are going to be off the shelf, not something they are going to need an excessive amount of development for, at least not above developing the booster landing RCS.

The issue, with both starship and the 747, is the structrual hardpoints to mount stuff, which applies equally to the shuttle mount as the engine mount.

[Khadgars]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version, I mean NASA even has their own version for Shuttle transport, and for carrying telescope, and of course another version is being used by Virgin Orbit to launch rockets. It all worked out very well, probably because Boeing has a production line pumping these out like sausages, and each variant is specialized for its particular role without needing to worry about other roles, this disentangle the dependencies.

In other words, the difference between 747 variants and Shuttle is this: When you ask 747 to be Shuttle transporter, a flying telescope and a rocket launch platform, you bring in 3 separate 747 aircrafts and each is modified to do its thing and only its thing. Shuttle is like refitting a single 747 to carry Shuttle, carry telescope and launch rocket all at the same time, that's where it went south.

While I understand the point you are trying to make with 747, I don't think that really applies.  You don't have to add thrusters or additional engines to carry the shuttle or transport telescopes and the take off and landing are essentially unchanged.

Moonship will have thrusters that Starship won't have and will operate in very different environment for extended amount of time on the moons surface.  Mars Starship will have an ECLSS beyond anything we have ever seen and will land in one of the most punishing environments humans will attempt in our life times.

All three of those variants are quite different from tanker starship which will be the largest re-fueling spacecraft ever produced. 

I am surprised so many people believe certifying all of these will be cheap and easy.  As a lowly program manager, I see extensive risk associated with all of these with most of the issues we won't even know exist until we're deep into the program.  SpaceX ethos of rapid build/test/repeat definitely helps in this regard but the task at hand is monumental.

Thrusters are going to be off the shelf, not something they are going to need an excessive amount of development for, at least not above developing the booster landing RCS.

The issue, with both starship and the 747, is the structrual hardpoints to mount stuff, which applies equally to the shuttle mount as the engine mount.

Which off the shelf engines will moonship use?  Its not just the mounting but additional plumbing and failure modes on a crew vehicle that is an issue.  If this were solely a cargo vehicle things would be easier.

[rakaydos]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version, I mean NASA even has their own version for Shuttle transport, and for carrying telescope, and of course another version is being used by Virgin Orbit to launch rockets. It all worked out very well, probably because Boeing has a production line pumping these out like sausages, and each variant is specialized for its particular role without needing to worry about other roles, this disentangle the dependencies.

In other words, the difference between 747 variants and Shuttle is this: When you ask 747 to be Shuttle transporter, a flying telescope and a rocket launch platform, you bring in 3 separate 747 aircrafts and each is modified to do its thing and only its thing. Shuttle is like refitting a single 747 to carry Shuttle, carry telescope and launch rocket all at the same time, that's where it went south.

While I understand the point you are trying to make with 747, I don't think that really applies.  You don't have to add thrusters or additional engines to carry the shuttle or transport telescopes and the take off and landing are essentially unchanged.

Moonship will have thrusters that Starship won't have and will operate in very different environment for extended amount of time on the moons surface.  Mars Starship will have an ECLSS beyond anything we have ever seen and will land in one of the most punishing environments humans will attempt in our life times.

All three of those variants are quite different from tanker starship which will be the largest re-fueling spacecraft ever produced. 

I am surprised so many people believe certifying all of these will be cheap and easy.  As a lowly program manager, I see extensive risk associated with all of these with most of the issues we won't even know exist until we're deep into the program.  SpaceX ethos of rapid build/test/repeat definitely helps in this regard but the task at hand is monumental.

Thrusters are going to be off the shelf, not something they are going to need an excessive amount of development for, at least not above developing the booster landing RCS.

The issue, with both starship and the 747, is the structrual hardpoints to mount stuff, which applies equally to the shuttle mount as the engine mount.

Which off the shelf engines will moonship use?  Its not just the mounting but additional plumbing and failure modes on a crew vehicle that is an issue.  If this were solely a cargo vehicle things would be easier.

a large number of hot gas booster RCS thrusters, as I said. as for certifying for crew to NASA's satisfaction, that's what NASA's paying them 3 billion dollars for.

[Coastal Ron]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

You could ask the same question about variants of commercial jetliners like 747, how many 747 variants are there? Probably dozens if you count specialized version...

While I understand the point you are trying to make with 747, I don't think that really applies.

I would agree that each variant would need to be tested to the same degree a non-Starship version would, but I'm not sure the original premise about comparing Starship to Shuttle holds.

The Shuttle was perceived to be a failure from a cost standpoint because during the design phase it was designed to be a vehicle that could do everything. And the Shuttle was not modified into different variants, they were all the same.

With Starship it truly is like a 747, where each version will use the basic "airframe", but will be modified from there to fulfill each role very specifically. So purpose built, but not having to start from scratch.

And to your point, each variant would need to go through a full set of testing, but that analogy fits with the 747 too, where each 747 variant is put through a full set of tests before becoming operational. But since they are using the same "airframe", the testing for each variant doesn't have to duplicate the basic function of the Starship that is the same from variant to variant.

[ncb1397]

The Shuttle was perceived to be a failure from a cost standpoint because during the design phase it was designed to be a vehicle that could do everything. And the Shuttle was not modified into different variants, they were all the same.

They weren't all the same. Some had features the others lacked.

[spacenut]

I will add that Starship is far more modular than Shuttle.  The core engines, frame, and fuel tanks are all the same.  Then you add different legs, fill the nosecone fairing with whatever you need, cargo, crew, both.  The fuel tanks can also be stretched to make a tanker.  Shuttle couldn't do any of that because it was a fixed airframe. 

Starship is designed for space use from the ground up, not fly through the atmosphere. 

[Redclaws]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

I think this is entirely valid as a concern, but it’s important to take the situation on its specific merits as well.

Which of the roles actually require significant design changes?  The idea is that the problem with the Shuttle was that the payload bay and wing shape (for cross range) made it kind of messily untenable.  The basic design - vs the early dreams of people like Max Faget - was disrupted.

So, I think the question is: what about the basic starship design is disrupted to meet these different goals?  Perhaps surprisingly, I think the answer is “not much”.  The basic engineering equations still seem to close for the current general design, without ugly compromises that disrupt it.

I agree that lunar starship is a bit of a stretch in some ways, but even there I don’t think the basic design is harmed much.  The absurd payload capacity given by size and orbital refueling is used to solve the lunar starship problems in ways that do not require changing the base design to something which is highly optimized for this task.

[Slarty1080]

I have to admit, I am concerned with the nearly unlimited roles people are proscribing to Starship.  Each variant would have to be tested extensively and you introduce a whole gambit of failure modes.

Isn't that what doomed Space Shuttle trying to accomplish too many roles?

SpaceX does not need to do everything. They can deliver a Starship cargo section to NASA. NASA builds a habitat to their own specs into it and SpaceX plants it on a propulsion section and sends it off.

You'd be correct for any 'standard' launch service but that's exactly the opposite of what SpaceX is aiming for. With Starship they plan from going from "Launch Services" to "Transport System" which is arguably a sea-change in that the 'service' will no longer be driven by customer requirements but by the transport system capability.
(Arguably btw that's a GOOD thing )

In the context here, SpaceX tells NASA what the capability is of the Starship cargo bay, NASA then builds a habitat that will fit into those requirements and ships it to the launch site. SpaceX puts it in the cargo bay and then delivers it to the destination. NASA can 'want' a bigger or more capable hab but if it doesn't "fit" it doesn't fly. Again this transforms space access from custom to mass production as it were. So far so good.

The downside here is that unlike standard space launch a transport system REQUIRES a certain steady volume of transported good because unused 'space' is lost money. You lose things like 'secondary' or 'ride along' payloads because instead of a 'single' customer ALL customers have to load share the price. Again all well and good but the load market has to be there in advance because it is very much not a "build it and they will come" type market.

Randy

You're argument about volume is not without merit, but is based on what has happened in the past.

IF Starship turns out to be half as good as promised (and IMO that seems likely) the space launch business will be turned inside out and upside down. It will be profitable for SpaceX, it will expand the launch market and it may kill off some but not all completion.

Profitable for SpaceX because it should cost less to launch Starship than Falcon 9  and eventually they will use Starship instead of Falcon 9 to launch satellites commercially removing the not inconsiderable costs of expendable second stages and refurbishments. Starship will also be likely to win further contracts from NASA and fly missions such as DearMoon.

SpaceX is already expanding the launch market by way of of its own launches of Starlink and Starship will further cut the cost of maintaining the Starlink constellation.

The competition will undoubtedly take a real drubbing, but it will not be eliminated. Many foreign governments will not be willing to abandon their national launch capabilities and the US government   is very keen to have at least two suppliers for services as can be seen by the political bruhaha and frantic political arm waving following the SpaceX award (which to be frank the political critters have only themselves to blame for): https://forum.nasaspaceflight.com/index.php?topic=53602.100

Ultimately it will be SpaceX plus a small band of also rans supported by the US government for the publicly declared and sensible reason that competition is needed and the unspoken political reason to ensure jobs are maintained in key areas. Come on Blue and ULA try to keep up...