Looks like the landing wasn’t quite as gentle as it first appeared:
Landing legs didn’t fully lock, plus it seemed to have a few meters/sec of residual velocity. Probably didn’t do wonders for the internals–say, by rupturing the methane downcomer. Definitely some fine-tuning left to do. I’m guessing some big, springy legs would have saved it, but they can’t spend that kind of mass on an orbital vehicle. In any case, it seems like it’ll be relatively straightforward to fix. Maybe it just requires some tweaks to the control system.
If you look at the landing carefully, Starship seems to bounce a little after it landed. At first I thought it might be the landing legs compressing then rebounding, but if they weren’t working correctly my second guess would be that the Raptor didn’t quite shut off in time and some residual thrust bounced the rocket.
But the fact that they stuck the landing is tremendous. The remaining issues around the landing gear and explody hits can obviously be fixed.
After the last landing attempt I was really worried about a design problem in the engines having to do with precession loads on the turbopumps when it flips with engines running. But this time it looks like all three fired, so that’s a very good sign!
Did anyone notice that the engines weren’t burning the same color? One raptor’s exhaust was orange, the other’s blue. I don’t k ow if that"s supposed to happen there (maybe different power levels?) but it was interesting. Looks like one was running richer than the other or something.
Yeah, I noticed the orange flame as well. Hard to say if it’s intentional or not, but I do know that one important thing for rockets is that they run through each of their propellants at the expected rate, and will tweak the mixture ratio based on that. Could be we’re seeing some propellant balancing along those lines, although why it would be limited to a single engine I don’t know. Didn’t seem to have any negative effects. I suppose throttle level might be an effect as well.
BTW, lots of previous Falcon landings have had a bounce. Some is from residual downward velocity and springiness in the legs, and some from the the engines not shutting down in time and giving it a little upward velocity before settling down. The Falcon legs have a fair amount of give, and crush cores that will absorb a fair amount of kinetic energy, so it has pretty good margins in this respect. The dinky Starship legs aren’t nearly so good at the moment, and effectively only act as crush cores.
In this case, the bounce may have been from the structure itself–it is a pressurized balloon, after all. It looked like it was low enough that it was almost standing on the body flaps.
Looking closely at the bounce it was bigger than most & rather abrupt.
The thing didn’t exactly “explode” It looked more like accumulated leaked propellant and oxidizer inside the fuselage skirt / engine compartment cooked off. Almost like a natural gas explosion: more of a whoomp! than a boom! In that contained area the effect was a brief burst of sorta-thrust. Just enough to toss the now-empty vehicle up about 1 fuselage length then drop it.
I recall some issues the Space Shuttle had with inerting the OMS pods after landing. It never exploded, but that was because a lot of nitrogen flushing was applied quickly.
It wasn’t my fault this time, I swear!, I went for a nap after my last comment and didn’t wake up until the ship had already taken off, flown, landed, and exploded.
Re: The explosion. It looked to me like more than just gas accumulated under the skirt. My first thought was that the fire around the bottom may have weakened the lower bulkhead of the main tank above the thrust buck, and it gave way and released the residual fuel in the tank. Or a fuel line ruptured, or something like that.
Scott Manley again with probably the best analysis of the landing and explosion. Using frame by frame footage he concludes it’s a pressure failure of the oxygen tank which may (or may not) have been weakened by the fire.
It appears that both of their pressurization techniques have their own problems. Helium pressurization has the issue that it always remains a gas, and if it gets sucked into the engine then bad things will happen (loss of thrust at the very least, but possible engine damage). Autogenous pressurization however has the opposite problem–it is too easy for it to convert back to liquid. This means the engines are unlikely to ingest gases, but significant sloshing can cause the gasified propellant to prematurely liquify. The tank then loses pressurization (“ullage collapse”).
Well. It sounds like slosh is the fundamental problem here, and if they address that then they can go back to autogenous pressurization. One possible solution would be to have a bladder in the tank separating the pressurant gas from the liquid, but that poses its own problems and may not work for them. Could just be that they need some appropriately-designed baffles. Or maybe just some open-cell foam like what’s used in race car fuel tanks?
Yeah. The 90 degree rotation from horizontal to vertical very shortly before landing has to really thrash the propellant in the almost empty tanks. Reliably getting that last 10% out with no gas admixture is a very tall order.
I might be wrong but I think the header tanks are much more than 10% full for the landing burn–that’s the whole point of them after all, so that you don’t have a tiny amount of propellant bouncing around a huge volume. But it appears that’s still not enough.
That’s a new one to me but makes complete sense. Shake up a temperature-stratified volume and you’ll liquify the gas or at least decrease the pressure.
Maybe they need a bubbler or the like to prevent the stratification in the first place? Would help to both mix the liquid and cool the gas down to the boiling point. Though then you might be back to the helium problems…
I was thinking of main tanks feeding the engines directly and hadn’t considered header tanks. Of course there are header tanks. D’oh!
Ideally the header tanks are full or nearly so. And under constant pressure. And being replenished from the main tanks with clean liquid at a rate that ensures they’re still near-full at final shutdown.
Although I suppose you could size the header tanks for the entire landing burn, then ensure they get filled during freefall, then valve off the main tanks. Now you’re just fighting slosh, gasification, and pressurization problems in a much smaller space while starting from much closer to 100% full.
OTOH, you’re also radically emptying those small tanks in a big hurry. Keeping that small volume at constant pressure while it goes from 1% ullage to WAG 90% ullage is still tough. Unless you make the header tanks a bunch bigger and only plan to burn them down from 100% full to WAG 50% full at shutdown.
Lots of dimensions in that trade space. Glad it’s not my problem.
The city will be dog friendly and the ‘The Doge’ will rule it. Ok, is it ruled by The Doge coin or former rulers of Venice and Genoa? Boca Chica will become Starbase which is in Texas. Somehow I don’t see exactly how that will come to be. Any ideas.?
Eddie Treviño, judge for Cameron County, Texas, confirmed that SpaceX informed the authorities of Elon Musk’s intention: to incorporate Boca Chica into the city of Starbase . The official noted that the mogul and his company must comply with all state statutes of incorporation and clarified that the county will process any petition in accordance with the law.
The 49-year-old billionaire is determined to make Texas his main base of operations. In addition to its launch center in Boca Chica, SpaceX has a rocket test facility in the town of McGregor and they plan to build a manufacturing plant in Austin. In July 2020, it was announced that Tesla’s next gigafactory would be built outside of Austin.
Boca Chica is a name for a tract of land; it isn’t a city. It’s standard legal stuff to incorporate tracts into cities although each state does it slightly differently.
Musk has infinite money and Boca Chica is a shithole comprising 20 people and their ratty 1960s houses. I fail to see what you think is hard or complicated here.
Just like with Disney in Orlando, they picked a cheap nowhere on purpose. And then turned it into a company town. Not in the sinister sense, just in the economically efficient = “seriously cheap” sense.
Don’t forget though that they have to maintain pressurization in the main tanks–the whole thing would crumple like so many beer cans if they lost that! So at least in principle they have a large reservoir to work from. And at least in autogenous mode, it’s constantly being refilled by hot gas bleedoff from the engines.
As far as I know, the header tanks are sized just right for the landing burn. In fact I suspect the venting you see during their test flights is to ensure that there’s no remaining propellant in the main tanks to mess up the center of gravity during landing. Maintaining a CoG compatible with the bellyflop is half the reason for the header tanks in the first place.
Heh, that’s for sure. But it’s fun to see them iterate toward a solution. This is the kind of thing that seems pretty hard to model on a computer. That means lots of kabooms as they find out what doesn’t work.
Starbase is a terrible name for a city on the ground. He should have called it ‘Spaceport’ or ‘Space City’ or something.
As for the landing issue, The header tank in the nose already fills the space, and because it’s there for weight and balance it can’t be moved. It’s also just LOX, isn’t it?
There are probably many solutiins here, all with different tradeoffs. For instance, putting a secondary header tank right at the axis of rotation coild minimize sloshing. Or, they could take a hit on fuel and do the rotation to vertical earlier to allow time for the fuel to settle down.
Since you didn’t link to the article I have no idea what “the Doge” is. But at least now I know what the heck you’re talking about, so thank you for that.
