I think Randall Munroe has it well diagnosed.
[Bottom of spacecraft]: This end should point toward the ground if you want to go to space. If it starts pointing toward space you are having a bad problem and you will not go to space today.
Kind of a weird video. The commentators still talking about stage separation after it had obviously flipped for the 2nd or 3rd time. Cheers when it exploded. Although maybe just getting to the point it did was a big accomplishment. (I haven’t really kept up with what the expectations here were). One of the commenters said something about anything after clearing the tower being icing on the cake.
Yeah, successfully making it to orbit on the very first attempt with a new vehicle would be pretty close to a miracle. The commenter is correct, if you get off the ground and get some altitude off the pad, you have achieved a successful test. The vehicle may have technically failed, but from an engineering standpoint the exercise is not a failure. It’s an expensive not-failure, but that’s how space goes.
I’m sure we all noticed that a couple of the 33 first stage engines were not firing by fairly late into the still-nominal part of the flight. They were in no obvious pattern, so that’s almost certainly not a method of throttling the collective thrust, but represents deliberate or forced shutdowns in the face of individual engine faults.
As I commented & Strangelove responded somewhere in one of these threads, with that many engines a) having a failure(s) is quite likely statistically, and b) is no big deal unless the failure mode is “contagious” = AKA highly energetic. Evidently these weren’t. Bravo SpaceX!!
That lady commentator had an excellent sense of timing to deliver her “it’s icing on the cake” comment just moments before what was probably range safety termination.
It’s been done enough times before that I’m not sure “miracle” is the correct term.
I don’t blame SpaceX for putting a good face on this but the “icing on the cake” bit is obviously not true. They were planning for the vehicle to achieve orbit and clearly everyone expected first stage separation and more flight into orbit, and it failed to do that. Stuff fails, that’s why you do unmanned tests, but there’s no point denying it failed.
I counted 5, and first stage engine status display in the video crawl agrees.
(This is cued to a moment when the engine flares are clearly visible and the status bar shows the same info.)
they are calling it: “unplanned disassembly”. seriously! i’m having quite the giggle about that terminology.
It looked to me like the vehicle rotated to horizontal too quickly, even before it started cartwheeling. When this happens to me in Kerbal it’s usually inadequate aerodynamic stability and the solution is bolting more stabilizer fins to the bottom of the rocket. Here I suspect the issue was asymmetric thrust. The whole flight right from coming off the pad the exhaust flame looked asymmetric to me, and there seemed to be plumes of unburnt fuel igniting behind the rocket or something.
@rocking_chair two posts up …
That’s one of the many standard joke terms of rocketry dating back 70 years. But like any good joke, it’s still chuckle-worthy even if you know it’s coming.
Those predecessors were done under the old-school “go slowly and incrementally and test each component individually and rigorously” model. SpaceX uses a different model, inspired by current software mantras, to wit, “go fast, don’t be precious, build stuff and try it and see what happens.”
In that context, I really do think it would have been close to a miracle if absolutely everything had worked as planned first time up.
OTOH this is the sort of failure you “expect” in the sense of knowing it is one of the likely modes.
I was surprised to see the flip maneuver beginning at the relatively low altitude of ~35km ~= 100K ft. and to start before the vehicle staged. Which from the commentator’s description at least began fully nominally.
I don’t have any clear notion of how they intended to stage, but I’d have initially expected them to stage in the traditional way while the vehicle stack was closely aligned with the velocity vector. Then once apart, the first stage would commence whatever off-axis thrusting and maneuvering it needed to get where it was supposed to go. Meanwhile the upper stage would ignite and continue thrusting more or less along the velocity vector. Yes, they’re also somewhere in the area of flattening the velocity vector from mostly vertical to mostly horizontal. Which necessitates axis-off-vector operations. But usually at pretty low pitch/yaw angles. Which we emphatically did not see even before things went off-nominal.
Even at 35km altitude, air loads are not negligible at high speeds. Either I misunderstood what was supposed to happen, or they intended to separate with non-trivial aerodynamic side forces on the stack. We know the upper stage is designed for such, but overall color me a bit surprised and mostly thinking I missed something.
Obviously once staging did not happen and wasn’t going to they just had a slowly tumbling ballistic pipe of soon-to-be wreckage on their hands. One hopes they got a lot of good telemetry out of the agonizing time it was tumbling slowly before someone or something leaned on the big red button. And good on them for not having an itchy trigger finger there. They were probably learning something with every second of that telemetry. Even if they haven’t had time yet to analyze whatever it told them.
Bummer 
Me too. I truly would have expected the more typical “MECO, separation, rotate and boostback*” first stage sequence.
*“Typical” boostback. That tells you how much Falcon 9 has changed the discussion about spaceflight.
I’m pretty sure they’d completely lost control of the vehicle by that point. They’re using a standard gravity-turn ascent during which you should never have the vehicle pointed in a direction noticeably different from its velocity vector. At a guess, the cartwheeling started either because MECO didn’t happen correctly, or because the staging resulted in a major aerodynamic anomaly. Given the plumes of flame coming out the back of the rocket during the spinning, I’d expect the latter. Possibly related to the ~5 Raptors that weren’t firing?
Not sure why Insprucker was still talking like it was still going to stage when it was clearly tumbling out of control.
Looks like some guys minivan got wrecked by rocks, so maybe from the launch pad or something? Pretty wild video. I’m glad nobody was hurt. Jebus.
Watching the video of the launch makes me feel much better about all my failed Kerbal Space Program launches. Who knew getting a flying bomb into space was so hard?
“Rapid Unscheduled Disassembly”?
Definitely the Phrase of the Day. 
Eric Berger.
Good summary — thanks.