Come again? Ullage?
More likely the downcomer (the big pipe that transfers propellant from the upper tank down to the engines). If there was a bubble in the pipe somewhere due to negative acceleration, then it could have slammed downward and ruptured the pipe once the acceleration ramped up. Because liquid is (basically) incompressible, this can result in a huge pressure shock (often called water hammer, even with liquids that aren’t water). This is all highly speculative, though.
Official Starship update:
Deets about the booster failure:
Following stage separation, Super Heavy initiated its boostback burn, which sends commands to 13 of the vehicle’s 33 Raptor engines to propel the rocket toward its intended landing location. During this burn, several engines began shutting down before one engine failed energetically, quickly cascading to a rapid unscheduled disassembly (RUD) of the booster. The vehicle breakup occurred more than three and a half minutes into the flight at an altitude of ~90 km over the Gulf of Mexico.
The most likely root cause for the booster RUD was determined to be filter blockage where liquid oxygen is supplied to the engines, leading to a loss of inlet pressure in engine oxidizer turbopumps that eventually resulted in one engine failing in a way that resulted in loss of the vehicle. SpaceX has since implemented hardware changes inside future booster oxidizer tanks to improve propellant filtration capabilities and refined operations to increase reliability.
Energetic(ally) is one of my favorite words in aerospace. It carries such promise, both positive and negative.
Cool, thanks for the update! Lots to unpack at your link. Much appreciated.
(Why the f was it necessary to vent O2? It was a simulation of… something. And F9 vents shittons of O2 and doesn’t esplode. Sounds like the aft tank was also venting? The text is quite dense.)
Do we really say “outer space” these days? It didn’t reach orbit. And 93 miles is a stretch anyway, a few km below Karmen. But I’ll give them that, shrug.
In retrospect, they probably should have done something other than that O2 vent, because it’s not a realistic scenario–but they had to do something.
Basically, they had too much propellant, because even if overall performance isn’t quite 100% yet, it still did very well and it had no payload. The vehicle wasn’t “supposed” to bring that much O2 to orbit, and it wouldn’t have completed the deorbit/reentry if it was still there. So they vented it.
They could have used a mass simulator instead, as if it were actually carrying satellites to orbit. That would have gotten rid of the excess prop as well. But then you’re running on slimmer margins. What if there had been a greater performance shortfall (say, because some engines went out)? Then, they’d have liked to still make it to orbit by burning that excess prop.
So there really wasn’t a great solution. And this isn’t a normal venting action–they were getting rid of dozens, maybe hundreds of tons of excess O2. Not just bleeding off a little excess pressure. Unfortunately, all those tons of O2 went where they weren’t supposed to and caused a fire.
The Kármán line is at 100 km (62 miles), not 100 miles. Though some American sources prefer 50 miles since it includes a few more early astronauts.
Dunno about “outer” space. But anyway, speed is more important. It’s easy to go 100 km straight up. Much harder to go 100 km up and 24,000 km/h sideways.
OK, thanks. My sources or reading comprehension failed me. Seems a bit grandiose to claim “outer space” but I’ll give them that one. And your point stands re velocity. They kicked some ass.
So the point was they wanted to do a “fully loaded” launch with no payload? Then adjust for reentry? I’m really not following the logic here. Perhaps I’m being dense. Quite likely.
I was hoping the Outer Space Treaty would shed some insight, but it doesn’t seem to have a proper definition. It does assume that Earth orbit is outer space, though, so I’d suggest that the lowest altitude where you can maintain a stable orbit is a good guideline.
But that’s not a fixed boundary, either; no one denies that satellites can orbit at 500 km, but at that altitude satellites don’t last more than a few years without thrust. If we assume thrust, though, you can go way lower. 150 km is probably fairly marginal but 167 km has been tested in the real world, so 150 km is probably possible.
Something like that. If they launch without payload, they gain a bunch of extra margin that they can use to make up for unintended performance shortfall. But then they need the vent if that shortfall doesn’t happen. They could have added a payload (or mass simulator), but then what if a few engines go out? Then the test is wasted because it doesn’t even come close to orbit.
I mean, given infinite impulse, you could “orbit” at 1km, right? 
OK. Still not grokking the mission profile, but thanks.
I assume a “mass simulator” is basically “mass”. Bricks. Steel plates. Stuff they can jettison.
I suspect that after the tests of the upper stage (now YEARS ago) yielded barely encouraging results (one finally made a soft landing before catching fire), they realized that success of Falcon 9 or no, this was going to be a lot harder than they thought. They did have to completely design a brand-new version of the engines, and the sheer scale of Starship introduced problems they couldn’t fully anticipate (like how rough launches were on the launch pad). Starship hasn’t failed yet but it hasn’t succeeded yet either. I’m afraid the days when we confidently assumed mass space transportation was just around the corner are gone. My worry now is that we’re going to get another Space Transportation System (the Shuttle): something that flies a few times a year after endless tinkering to fix yet something else that’s gone wrong. I’m bummed.
I might remind you that the Falcon 9 reusability test program only ever made a few low-altitude hops at first, with their final test vehicle making a nice RUD:
They didn’t even build another test vehicle (it got cancelled). Was that a sign of giving up? No–it meant they learned everything they needed to know from the test program.
They actually landed a rocket about a year and a half later and made steady progress until today.
Starship is undoubtedly hard–harder than they’d have liked, I’m sure–but it’s already clear that it won’t be an SLS/STS boondoggle. They’re producing new boosters and engines are far too high a rate for that.
I do expect reentry to be hard, though. But even without second-stage reuse, we still have a fine partially reusable superheavy lift vehicle.
Starship looks awesome at this point. I wish we could up the cadence though. Fail faster.
Yeah. But only jettison once in orbit (like a satellite). No way to jettison it halfway into flight when you decide that your engines are underperforming. But if you leave off the mass simulator, then you get the option of burning the extra propellant (in the case of underperformance), or just venting it (in the case of nominal performance). Unless the venting itself fails…
It’s gonna look like this:
Interminably slow at first, and then so fast that you can barely keep track. Yeah, I want my rockets now, but this is already a pretty quick program…
BTW, I aim to be around Starbase on April 6. I’ll be in the (general) area for the eclipse and decided to make a trip down there. Not expecting a launch or anything, but the infrastructure should be impressive.
The FAA has also closed their investigation:
They still need a license, but it’s obviously getting close since there’s no environmental review this time.
Also, 17 corrective actions in IFT-2 is way better than the 63 from IFT-1. By the magic of geometric series, I can conclude that IFT-3 will only require 5 corrective actions, IFT-4 will have one, and IFT-5 will have zero. Yay!
“Mass simulator” is such a silly term, especially for the industry that brought us “rapid unscheduled disassembly”. It doesn’t simulate mass, it is mass. If we could simulate mass without any actual mass, we wouldn’t need chemical rockets.
What’s wrong with “test mass” or a “payload simulator”?
We can call it a Useful Mass Simulator instead. It doesn’t simulate any old mass; it simulates useful mass.
Or just call it boilerplate. It’s not that either, but it does make for better imagery.
Another FAA statement with some minor extra details:
So there was extra onboard camera footage. Maybe we’ll get to see that someday.
. . . a sports car . . .
Beware the Ides-of-March-eve!
Starship IFT-3 tentatively scheduled for March 14. No official FAA approval yet, but I suspect it’s pretty well confirmed within SpaceX at this point.