Which is why I asked upthread what makes Starship tiles different from Shuttle tiles.
As I’ve had occasion to remark in the past, the entire Starship concept relies on the upper stage being recoverable, without loading it with so much shielding that it can’t carry useful cargo. If that doesn’t work goodbye cheap routine launch, colonizing Mars, the whole works. Reentry that is not one-use or extensive refurbishment every flight is the single issue that has bedeviled space flight since the beginning. Everything hinges on a Starship upper stage making it back down intact.
I’d disagree with this. Take away upper stage reusability and you still have a rocket with better cost per kg than the Falcon 9, which is already the best available. Methane/LOX is still probably the best propellant combo even ignoring the ability to produce it on Mars. Programs like Artemis are still relatively cost effective. Starlink launches still get cheaper, if not quite to the same degree.
Mars colonization is out without upper stage reusability, but not necessarily a boots+flags mission.
At any rate, SpaceX may have to take a totally different approach. Maybe the transpiration cooling they looked at originally, or a metallic heatshield concept, or some combination.
To add a bit, the current tiles don’t seem to have any fundamental problems. They’re capable of surviving reentry under optimal conditions. It’s just that it seems hard to keep them attached, and they’re still fragile enough that they can get damaged, and they have poor resiliency to failure. The fixes to these problems have to be cheap and low weight.
As Musk said, though, these seem to be failures of execution. Engineering problems. The material is already tougher than the Shuttle tiles, but surely it can be made more so. And surely they can come up with a way to attach the tiles securely. Etc. These aren’t trivial problems but it does seem more like a matter of trying things until they find something that works, as opposed to anything fundamental.
To an extent this is due to the economy of scale of Starship being just plain bigger, there being fixed costs per launch irrespective of the rocket size. The question is whether Starship can be cheap enough to support SpaceX’s own plans for it; some observers have commented that without that mass utilization Starship would flood the launch market to an economically unsustainable point.
I could see a middle road if it turned out that at least at first Starship had crappy upper-stage reusability: refurbishment back in the hanger after every flight and scrapping the upper stage after two or three relaunches, rather than the airliner cadence Musk hopes for. But presumably that would allow a developmental approach of working the bugs out.
There are a few other factors. The autogenous pressurization eliminates the helium needs, which actually costs more than the kerosene+LOX on a Falcon 9 launch. Methane is cheaper than kerosene as well. The booster always lands back at the launch site, eliminating the somewhat expensive oceangoing support. It should require less refurbishment than an F9 booster due to less sooting and other things. There are no fairings, which do get reused on an F9 but still aren’t free to recover/reuse.
So the upper stage becomes even more dominant compared to the F9. An F9 upper stage is said to be around $10M, with the marginal cost of a flight being around $15M. A Starship upper stage could plausibly cost $30-40M (it’s bigger, but also uses cheaper materials and has simpler construction), while the other factors probably won’t change much. So plausibly around $40M a flight for >6x the payload mass. That’s “only” a ~2x improvement over F9, but that’s still nothing to sneeze at. And I think that’s actually a conservative estimate, since F9 actually costs more than the pure marginal cost, while Starship should be able to get a bit closer.
So, certain categories of failure laid out by SpaceX won’t warrant another mishap report. I wonder if this option was always available to SpaceX or is something new. Regardless, I’m hoping it means we can see a further increase in flight rate. Musk mentioned that he wanted to see at least 6 launches this year, which seemed improbable at the current rate, but maybe with the improved regulatory status it could actually happen.
Those missing tiles are clearly intentional. One of the three has some different underlayment that is perhaps more protective than the white fabric. All three overlay the engine bay, which is probably not too affected by burn-through. Worst-case, it fries the engines, but for now the soft landing isn’t as much a priority as figuring out the heat shield. So it’s worth a risk here to gather data on what happens with missing tiles.
It might just be a trick of the lighting, but it looks like the gray one is deeper than the other two. So that one might be missing the white fabric, too, as a further test. I’d be willing to bet that there’s also something different between the two white ones, that’s just not visible in the picture.
I’ve said before that the only way to make something safe is to first have lots and lots of accidents, so you can figure out everything that can go wrong, and how to fix it. It looks like SpaceX is committed to that philosophy.
It’s off! Down one engine at the start. But they still have 32 left .
ETA: All nicely nominal for now. Got a great view of the hot-stage jettison. This is a temporary thing, but for now they’re dumping a pretty heavy (~10 ton) piece of dead weight.
Absolutely. They are committed to blowing things up or otherwise if it gives them good data (assuming no risk to humans). What they especially like are experiments that don’t risk the main mission, even if they themselves are very risky.
One of the earliest examples of this was testing hypersonic retropropulsion of the booster in the early days of booster reuse. Everyone else just let the booster break up in the atmosphere. Why not at least try the experiment? It turned out that the problem was way easier than expected and is one of the easiest parts of booster reuse. But no one knew this until they tried.
Looks like the landing burn was largely successful, and I think I saw some waves through the cracked lens and everything else. Incredible. Can’t wait to hear the confirmed results later. I wonder if the Ship is actually intact in the ocean.
Didn’t think they were going to get that soft splashdown after the fin started burning through, but if the telemetry on screen was accurate they absolutely did.
Unbelievable that it survived the loss of that flap. And quite possibly the forward port flap as well–we lost the video feed from that one, so it very likely also took damage (not to mention being the same design).
Scott Manley had a good quick recap of the flight:
I was at the launch but unfortunately with the cloud cover you couldn’t see much beyond the first few seconds of flight. The sound was absolutely amazing though, my t-shirt was visibly vibrating while standing on the jetties on South Padre Island.
So many mind-blowing moments during the flight, but the final 10 minutes of Ship’s descent, being able to capture that on live video, was like a Hollywood script. Watching the escalating damage to the flap it was unbelievable seeing that the Ship remained in control, the flaps were still functional, and a relight + flip + soft landing was performed.
.
