T-5 minutes!
And it’s off! It’s gonna come down now, one way or another…
This is amazing!
Now that’s what I call a suicide burn. Might want to work on the timing just a little…
Soo many flights in KSP that ended exactly that way!
But still, that was a heck of a successful test. Nice stable ascent, bellyflop all the way down, even the flipover worked well. Just need to get the final timing right. Maybe the engine restart didn’t work quite right, or the rotation wasn’t quite fast enough… but still, they hit a lot of milestones here.
Hey that’s a pretty big chunk of the nosecone semi-intact. Jeb may have survived!
Jeb is pretty durable… I think he must have made it!
Worth pointing out that they crashed dead-center on their landing pad. They seem to have pretty good control during the bellyflop.
Rewatching the flight–definitely some engine-rich exhaust at the end! The exhaust turns a bright green at the end, which means it’s burning the copper combustion chamber liner. It’s impressive that the engine is still going at all, though it must have lost a lot of power.
That was impressive as hell. I’ve watched it three times now. Tons of excellent data produced today, I’m sure.
I thought the flip and belly time was particularly impressive as it seemed to be under control and controllable. It actually looked like a plane with poorly placed stubby wings for a minute or two there. And when the Raptors re-ignited… well, I lost my shit. I hooted and hollered pretty good, threw my hands in the air with a big “woot!” and then half-laughed through my pitiable “awwwwww” as it, um, landed with prejudice. Is that the technical term?
Musk says:
I’m thinking the engine went way oxygen-rich (due to lack of methane pressure), and that both eroded the combustion chamber (copper green hue) and caused the lack of thrust. Seems like it should be a relatively straightforward fix.
Aye; I mean: we know SpaceX knows how to land a rocket.
Yeah–although it’s not that obvious, if they didn’t have fine control from the start, the entire descent would have been a failure. That stable horizontal descent is only possible because their flaps were doing their job. Otherwise it would have been flipping end over end or some other crazy rotation. There’s nothing inherently stable about their configuration.
Given that the flaps were a big part of the risk retirement they were hoping for, this is a big win. As always, their control system team deserves mucho props.
A very, very cool feat. Rocketry v2 is really happening now, isn’t it?
Rapid unscheduled disassembly.
Also, “putting the crater in the right spot”:
That whole thing was impressive as hell from end to end. But this is was what really impressed me the most:
They totally dominated the aerodynamics of that very unaerodynamic machine. And it wasn’t like there wasn’t propellant rattling around in those mostly empty tanks. There was massive CG sloshing as well as CP instability.
The sheer Hz rate of their control loops and the quality of the filtering and response tuning is just insane. Especially for something that’s never flown in this config. From the vids there was negligible hunting of the vehicle attitude; it just sat there falling sideways as if it was sitting in a horizontal cradle on the pad. Wow.
An interesting Q about the low fuel pressure is whether this was due to a valve freezing during the time the engine wasn’t running making heat, or was due to some sloshing issues like tank outlet uncovering, or a loss of gas pressure head (assuming they’re using an external pusher gas, not just residual ullage pressure. We’ll know soon enough.
As you say, ought to be a straightforward fix in any case.
The density of telemetry these days would make a 1950s Atlas engineer cry. It almost works the first time and whatever didn’t work is fully documented in actionable detail. So much informed guesswork went into the post-mortems of the early iterations of 1950s into 1960s rocketry.
Yeah, the flaps were super impressive. That’s 100% turbulent flow behind them, so for it to be as stable as it was is pretty amazing. Not to mention the sheer power requirements: I’ll have to do the math at some point, but the actuator power must be immense to give it the control authority it has. Apparently they’re using Tesla motors/batteries, so they may well be using megawatts of power across the four surfaces.
Your suggestions sound like definite possibilities for the header tank failure. I wonder if they’re doing anything special for ensuring the fuel makes its way to the outlet. Various rockets have used different solutions here for the problem of feeding propellant in weightless conditions; this is sorta the same thing. One possibility is to have a flexible diaphragm inside the tank, pressurized on one side, that directs all the fluid out the right hole. But maybe they’re just depending on the deceleration happening in the right direction.
As you say, the telemetry available now must be a godsend. It’s amazing they figured out anything from the data they had in the 50s and 60s; they were clever folk. Still, it’s nice to not have to make so many clever guesses.
The SN9 prototype is basically complete; hopefully they can retrofit it with whatever fixes are needed and fly it soon.
I like this video:
Just a normal, relatively wide-angle shot. The bellyflop comes down so gently. Part of it is that Starship is big, and so looks like it’s moving slowly. But part is that it actually is going pretty slowly, since it’s got so much surface area.
This first viewpoint doesn’t look real. Unbelievable!
