What if Starship can't work as currently envisaged?

[Kaputnik]

Previous and related thread:
What if Starship fails? Possible Plan B's for SpaceX


Firstly, just to get it out of the way, this is hypothetical. I think Starship will eventually be made to work as currently planned (two stage, fully reusable, methalox, etc).

But let's say it can't be made to work. Full and rapid reuse eats all the payload. Or the vehicle can't be made robust enough for economic reuse and acceptable reliability.

What would you change about the design in order to arrive at a vehicle which comes closest to achieving the program goals?

[edit: zubenelgenubi]

[Tywin]

IMHO, go full in the Nova - Stoke solution...

[Starship Trooper]

First, just an aside, in a recent NYT interview, Peter Thiel incidentally let slip that Elon Musk told him in 2024 that he had given up on Mars.  If true, this is huge, and removes the incentive to develop a reusable upper stage ASAP, which is where all the problems lie.

IMO, SpaceX overreached in trying to solve too many new problems simultaneously.  This is the same error Lockheed Martin made 30 years ago with the Venture Star SSTO programme, which failed completely.  The next step after Falcon 9 to build a "BFR" (originally supposedly standing for Big Falcon Rocket) should have been exactly that.  Solve only three problems:
1) Upscale to building the most powerful first stage ever
2) Make this massive booster stage reusable with a new catch system
3) Design new methane based engines and build experience with them
If they had done only this, they would be near success already, with huge potential from a reliable reusable heavy lift booster.

The reusable second stage was always where all the gremlins lurked.  The elephant in the room always has been thermal protection for the second stage.  Either the TPS would be light but fragile ceramic tiles or perhaps heavy metallic tiles with active evaporative cooling system which would be complex, heavy and reduce payload mass.  There might never be an ideal solution to this problem and the developmental time could be long.

The stated goal of crewed missions to Mars added many other major challenges, on orbit refuelling, the requirement for landing legs, and maintenance of cryogenic fuels for months in transit requiring active cooling.  None of these problems are simple, and we haven't even gotten into the requirements for long duration human life support and health maintenance.

[DanClemmensen]

Booster has already achieved its goals. Ship is already fairly cheap to build due to high-volume manufacture. Thus, they can choose to go operational with an expendable Ship while they keep working on the reusable Ship.

I expect them to achieve the reusable Ship before they need to implement this fallback.

[Starship Trooper]

IMHO, go full in the Nova - Stoke solution...

I see Nova as a nice small scale test bed for a reusable upper stage.  The orbital testing of a robust TPS with active cooling needs to be done, and SpaceX has a lot of problems to sort out before they can even begin - like orbiting a Starship to begin with.  However Nova will only lift 3 Tonnes to LEO, the high cost of a reusable upper stage, as we saw with the Shuttle.

[Starship Trooper]

Booster has already achieved its goals. Ship is already fairly cheap to build due to high-volume manufacture. Thus, they can choose to go operational with an expendable Ship while they keep working on the reusable Ship.

I expect them to achieve the reusable Ship before they need to implement this fallback.

Ok, I'm comparing apples, oranges, and pears in one go, but it is far from clear that an actually reusable Starship will be anywhere close to the payload goals previously planned.  The planned fully reusable Nova rocket will be ~ 40% of the mass of Falcon 9.  Falcon 9 can put 17.5 tonnes in LEO (when landing on drone ship).  Minus the requirements for upper stage reusability, Nova should be able to put ~7 tonnes in LEO.  Instead, Nova will boost only 3 tonnes to LEO.  That's a 57% reduction in payload capacity.
If SpaceX does manage to perfect the Starship to reliability comparable to Falcon 9, (not prone to rapid unscheduled disassembly) and with rapid reusability, I would expect the payload fraction to be no more than half what was previously imagined.  I have seen older materials hoping for 200 tonnes to LEO with Starship V3.  More recently 100 tonnes to LEO with V3 was the number quoted.  Might be the price of not shaking itself to bits during the boost phase and not burning up during reentry.

[Kaputnik]

How much would a third stage help?
Assuming it was simply a shortened version of Superheavy, with the same style of return profile, there would be a limit to how high and fast staging could be. And it would slightly complicate stacking, whist not addressing the problems with the Ship. But it's one obvious option.

Similarly, parallel staging, via strap on boosters, would add complexity but allow reduced engine count and dry mass on Superheavy, which allows more mass margin for the Ship.

Much more radically, they could return to carbon fibre for the booster. This was only abandoned when they realised it wouldn't work for orbital entry, and they wanted to maintain commonality between both vehicles. It may be possible to fly the Superheavy profile with a carbon booster.

Given that the booster is the bit of the system that actually works, it does seem a shame to make such major changes to it. But I'm struggling to see what could be done to change the design of the Ship without cutting in to payload mass.

Recovering only the aft end of the ship and expending the tanks and payload bay could work... until you need to land on Mars.

[sdsds]

If the current Starship/SuperHeavy system were somehow intrinsically flawed, the next best approach would be to take the best LPRE the world has ever seen (Raptor 3), design a massive reusable booster with the SL version of that engine, and use that booster as a flying test stand to work out what's required for a fully reusable upper stage.

No, that's not identical to the current plan because it drops the the requirement that the Ship can return from Mars to Earth without refilling. Having a single-stage Ship designed to get from the Mars surface to low Mars orbit still allows eventual Mars settlement, albeit more slowly than will happen with the current Ship design.

[crandles57]

If you cannot get fully reusable reliable upper stage due to TPS not quite being good enough but still want to get as close to fully reusable as possible for Starlink and other commercial launches:

Might be more complex but can we plausibly imagine a 3 stage vehicle:

Super heavy reusable booster 33 raptors as now since it seems to work ok.
9 engine second stage. Doesn't quite get to orbit so heat shield, that isn't quite good enough to be reliable from orbit, copes. Lands a long way downrange I assume? Might be shorter than Starship 2nd stage as no payload volume reducing TPS area making TPS refurbishment less time consuming.
Single engine disposable 3rd stage.

Not sure you can do as much with this as there isn't the refuelling volume on the 3rd stage. So as alternative to above 2nd and 3rd stage, also have just a second stage disposable tanker for refuelling and a different second stage only that can be refuelled by the tanker, has a payload volume and when needed TPS to cope with Mars aerocapture and landing.

Compromises to try to make it work in this specific case of TPS not quite working:
Disposable 3rd stage but it is only a single engine stage
Disposable tankers
Mars ships may not manage Earth re-entry so have to transfer to another vehicle
Taller so may need taller towers with 3rd fuelling quick disconnect
Transport of second stage back to launch site is a pain as is TPS refurbishment but can still get rapid launches once you have a large enough fleet of second stages but these steps and 3rd stage do make it more expensive to launch.

Kaputnik beat me to this but since I have typed it ...

[Robotbeat]

Firstly, just to get it out of the way, this is hypothetical. I think Starship will eventually be made to work as currently planned (two stage, fully reusable, methalox, etc).

But let's say it can't be made to work. Full and rapid reuse eats all the payload. Or the vehicle can't be made robust enough for economic reuse and acceptable reliability.

What would you change about the design in order to arrive at a vehicle which comes closest to achieving the program goals?

Then they’ll change it.

[edkyle99]

If they can't make it work, at some point (when money becomes an issue) they will have to stop trying and scrap the program.  I don't see a middle ground with this outfit.  Here's hoping they make it work.

 - Ed Kyle

[Kaputnik]

Firstly, just to get it out of the way, this is hypothetical. I think Starship will eventually be made to work as currently planned (two stage, fully reusable, methalox, etc).

But let's say it can't be made to work. Full and rapid reuse eats all the payload. Or the vehicle can't be made robust enough for economic reuse and acceptable reliability.

What would you change about the design in order to arrive at a vehicle which comes closest to achieving the program goals?

Then they’ll change it.

Well, yes. Just like they changed from carbon to steel, sand how details about legs and number of engines have been altered over the years.

Don't you care to take a punt at *how* they might change it?

[envy887]

The elephant in the room always has been thermal protection for the second stage.  Either the TPS would be light but fragile ceramic tiles or perhaps heavy metallic tiles with active evaporative cooling system which would be complex, heavy and reduce payload mass.

The upper stage TPS has done its job every time it has had a chance. I don't see any reason why it can't get to at least Falcon-like reuse (down to about 5 days now) with only incremental improvements.

[Kaputnik]

If they can't make it work, at some point (when money becomes an issue) they will have to stop trying and scrap the program.  I don't see a middle ground with this outfit.  Here's hoping they make it work.

 - Ed Kyle

The hard limit seems to be that the system doesn't work unless the Ship has enough delta V to get from the Martian surface back to earth. So whilst we can discuss drop tanks, extra stages, etc, this fundamental requirement always remains. Like you say, no middle ground.

If they wanted to shelve the Mars plan, they would have all sorts of options open. They have, after all, developed the world's highest performance engine. There's a great launcher in there somewhere.

[Kaputnik]

The elephant in the room always has been thermal protection for the second stage.  Either the TPS would be light but fragile ceramic tiles or perhaps heavy metallic tiles with active evaporative cooling system which would be complex, heavy and reduce payload mass.

The upper stage TPS has done its job every time it has had a chance. I don't see any reason why it can't get to at least Falcon-like reuse (down to about 5 days now) with only incremental improvements.

It's been able to get the very robust V1 vehicle through entry.
What's yet to be proven is how good a job it did of that. We don't know if the more lightly built V2 would have been week enough protected.
They've had to add a secondary TPS under the tiles, which definitely shows that its not working as well as intended. The TPS is only really doing its job fully when there is little to no refurbishment required.

[sdsds]

The hard limit seems to be that the system doesn't work unless the Ship has enough delta V to get from the Martian surface back to earth.

No. They can put a depot in orbit around Mars.

[meekGee]

Firstly, just to get it out of the way, this is hypothetical. I think Starship will eventually be made to work as currently planned (two stage, fully reusable, methalox, etc).

But let's say it can't be made to work. Full and rapid reuse eats all the payload. Or the vehicle can't be made robust enough for economic reuse and acceptable reliability.

What would you change about the design in order to arrive at a vehicle which comes closest to achieving the program goals?

Then they’ll change it.

Well, yes. Just like they changed from carbon to steel, sand how details about legs and number of engines have been altered over the years.

Don't you care to take a punt at *how* they might change it?

Well it'd be helpful to know how it didn't work first, wouldn't it?

But generally speaking, Mars ships are one way anyway.  Ablative shielding.

Tankers, if can't be used, go with a three stage configuration, second stage runs slower and has to be hauled back.  Third stage is disposable.   

Starlink, sames.

Or, replaceable ablative shield.

Less sexy ideas, but they are reasonable fallbacks.

[catdlr]

Firstly, just to get it out of the way, this is hypothetical. I think Starship will eventually be made to work as currently planned (two stage, fully reusable, methalox, etc).

But let's say it can't be made to work. Full and rapid reuse eats all the payload. Or the vehicle can't be made robust enough for economic reuse and acceptable reliability.

What would you change about the design in order to arrive at a vehicle which comes closest to achieving the program goals?

Then they’ll change it.

Well, yes. Just like they changed from carbon to steel, sand how details about legs and number of engines have been altered over the years.

Don't you care to take a punt at *how* they might change it?

Well it'd be helpful to know how it didn't work first, wouldn't it?

But generally speaking, Mars ships are one way anyway.  Ablative shielding.

Tankers, if can't be used, go with a three stage configuration, second stage runs slower and has to be hauled back.  Third stage is disposable.   

Starlink, sames.

Or, replaceable ablative shield.

Less sexy ideas, but they are reasonable fallbacks.

It would be sad if Startship didn't pan out.  However, perhaps the legacy of this is the creation of a great engine, the Raptor, but alas, an engine without a rocket is still sad.

[Ioanspace]

IMHO, go full in the Nova - Stoke solution...

Stoke is trying to combine an expander cycle engine with a regeneratively cooled heat shield, for a startup, I think their vision is more ambitious and unlikely than Spacex's starship.

[mikelepage]

I'm fairly sure they'll succeed in enabling the launch of much larger payloads to LEO in the near term, even if the refuelling architecture takes a lot longer to figure out.

I personally find the idea of building much larger and capable space stations in LEO to be pretty exciting. Some of which could be rotational gravity stations, some of which will be microgravity research/manufacturing stations. And as a necessary corollary, a fleet of much more capable space tugs to go around cleaning up the hazardous space debris.

I know it's trendy to poo poo all the work that's been done on ISS, but there's a lot more that could be done on similar platforms. And the cost/availability of up- and down-mass only has improve by a factor of 8-10x or so for several of the business cases for in-space manufacturing to close (NASA CASIS currently funds 85-90% of the cost of all experimental payloads to ISS). Even a partially reusable starship that only launches a dozen times a year will still achieve that.