You’re at least giving SpaceX the benefit of the doubt. On that Mastodon site I mentioned the chorus of “Starship is a fraud” is louder than ever.
I watched the entire flight with a few distractions because fucking work is pesky that way.
- Launch nominal.
- Just after MaxQ, they lost 2 of the 33 booster engines in fairly rapid succession.
- No problem, flight continued until hot staging which went flawlessly.
- Booster returned for a soft water landing, then the FTS was triggered on purpose.
- Starship continued on and somewhere along the way, 1 of the 3 Merlin Vacs shutdown unexpectedly.
- The Ship compensated and achieved sub-orbital flight.
- LOTS of discussion and after some diagnostics on the 3 inner (non-vac) Merlins, the decision was made to go for orbit.
- Orbit achieved!
- Starlink v3s launched. (kinda cool to watch)
- Something in the orbit burn appears to have made them nervous about the non-vac Merlins and they decided to deorbit early.
- Deorbit was, as far as I could tell, pretty smooth including entry and prep for landing burn.
- Landing burn appeared successful (no idea if it was accurate enough for a chopstick catch).
- Starship blew up. It’s possibly that was also done by FTS to avoid a giant floating nav hazard, but I’ve seen nothing official on that.
My personal takeaway:
The Merlin v3s are a finicky lot and they’ve got some work to do. I think if they can get those to at or near Falcon Merlin (v1D) engine reliability, they’ll be close. But until they catch the Starship and can evaluate the turn-around time (which will start long and progressively get shorter if F9 is any predictor), then they’re still a long away from a cadence that will support their plans for Artemis.
TLDR: We ain’t going to the Moon anytime soon.
You mean Raptors.
I get why they wouldn’t be too bothered by a failure on landing, if that’s not what they’re currently working on. But why not at least attempt an intact landing? Bring it back to the pad, if that’s what’s necessary to make the attempt. If it doesn’t work, it doesn’t work, but if it does, they can get a lot of data from the intact vehicle, and it’d be great PR.
Wouldn’t a crash at the launch site create a lot of damage? I could see their not wanting to risk it until everything’s as sure as it can be.
OK, then, at an empty concrete pad, where it won’t be dunked in thermally-shocking cold water as soon as it tips over.
They aren’t built to land on their own like a Falcon 9 booster. No landing gears. They have to be caught with chopsticks.
The moon-landing version of Starship will definitely have landing gear. Maybe they could put some of those on one of these orbital versions, if only to test that landing gear out. Of course, that landing gear probably doesn’t exist yet and when it does, it may be too fragile for an Earth landing.
In case you hadn’t noticed, SpaceX is very systematically limiting the number of untested systems they exercise in any specific mission. That’s just good engineeering discipline: reduce the number of potential failures.
There’s a limit to “go fast break stuff” and that’s beyond it.
For all the integrated full suborbital/orbital flights they’ve done, when the ship made it all the way through reentry intact, every last one of them has made a pretty well-controlled soft landing in the ocean, and once it gets that far, successful test by SpaceX’s standards. They expected the explosion after the ship tilts over and hits the water. That last time when it failed to do the R.E.D. (rapid expected disassembly) was an outlier. And then they had all sorts of trouble recovering it when they hadn’t planned that part.
Son of a…yep. Starship uses Raptors, not Merlins. 
As I understand it the lending burn is supplied from the header tanks.
Do they vent the main tanks before re-entry?
Srarship losing tiles at launch.
That’s an interesting point. ‘Tiles’. The shuttle needed a great deal of meticulous tile replacement between each launch. If we are to achieve reusable airline-like launch cadences, we can’t have that sort of delay.
It seems we need some better type of reentry shielding?
I think may be trying to make re-usable re-entry vehicles is not a good idea, re-usable boosters yes, they work perfectly now, but the Shuttle had a lot of problem with being a re-usable re-entry vehicle and is not clear that Starship has solved them
Why not simply use boosters to go up and one-use capsules to go down?, once in orbit transfer to a Starship-without-reentry-capability and fly away to wherever, come back, transfer to a one-use reentry vehicle and go down to earth.
The sci-fi concept of the single earth->orbit/orbit->outer space/outer-space->orbit/orbit-> earth ship is too complex.
That’s a very good question. It perhaps boils down to economics.
Re-use the booster and expend the upper stage… what are the pros and cons of that versus a fully reusable system? If the upper stage is cheap enough to manufacture from things like stainless steel, does it make sense to go to heroic efforts to recover and reuse it?
The upper stage would not be necessarily expendable, you use it to perform different missions in space, it just never returns to earth (we’ll have to make in-orbit refueling work, but that’s something we need to do anyway)
True, once you have it up there, it could have a lot of uses. They could (I think) have orbited the shuttle hydrogen tanks to provide a lot of potentially useful orbital structure, but chose not to…?
Actually once SF writers got beyond the idea of “atomic” powered spaceships with magically high thrust and high delta-v, splitting spaceflight into dedicated orbiters, dedicated orbital transfer vehicles and dedicated landers was ubiquitous. SpaceX’s Starship revamped the all-purpose “spaceship” concept because it was realized that a reusable orbiter didn’t have to be a rocket-powered airplane (something Philip Bono tried to tell people all the way back in the 1960s). The math says that an upper stage refueled in orbit has enough delta-v to travel throughout the inner solar system, and with in situ refueling chemical rockets are cheaper than nuclear-thermal. All this is possible IF the reentry heating problem is solved satisfactorily, something that has bedeviled aerospace engineering since the dawn of the Space Age back in the late 1950s.
Everything that you could have used Shuttle tanks boosted to orbit for required that the Shuttle fly frequently and inexpensively, which never happened.
Ref Shuttle tanks …
IIRC they realized the extra weight to put guidance & maneuvering on the tanks for later use destroyed the Shuttle’s already marginal payload performance to some essential orbits.
Zooming out a bit …
The Shuttle was an unfortunate mix of technological bleeding edge hubris and the mismatch between it taking ~15 years to design for a pair of bickering customers whose needs changed every 6 months.
Pretty quickly early design choices got baked in even though subsequent requirements no longer needed whatever drove that decision.