It looks like the rocket ran out of hydraulic fluid to control the grid fins. It uses an open loop system rather than a closed loop to save weight and complexity, and ran out 24 seconds before touchdown. Upcoming launches will carry more fluid, so the problem shouldn’t re-occur.
SpaceX plans to make another landing attempt when they launch the Deep Space Climate Observatory, scheduled for January 29.
I was a little surprised at this at first but it makes sense. The system doesn’t operate for very long, so the weight of the excess fluid is probably less than the extra piping and stuff needed to recover it.
The tweets mentioned “hydraulic fluid,” which I initially assumed actually meant RP1, since the Merlin engines themselves use a fueldraulic system for gimbaling. But I don’t think that actually makes sense–the fins have to operate even when the engines are off. It could use a hydraulic accumulator, but all the weight of the fluid and everything would have to be there anyway. It doesn’t make a lot of sense to pressurize that via the engines in flight.
So I’m guessing that they really just have a fixed reservoir, possibly pressurized via the same helium tanks that provide the tank ullage pressurization. They can then use proper hydraulic fluid, which has slightly better properties than RP1, and probably means they can use cheaper and “relatively” off-the-shelf components (fueldraulic systems need special design). They have only limited maneuvering time, but that’s ok since the vehicle is coming down no matter what…
That’s funny; the damage doesn’t look as bad out here.
From the charred containers, it looks like the stage just didn’t hit the center of the target, as one might expect given the hydraulic fluid running out. There doesn’t seem to be a ton of debris under the tarps, so unless they already moved most of it, I’d suppose that the remaining bits sank into the ocean.
Now I really wish there was video! I’ll bet that barge was rocking.
There is this amusing animated gif that I found on Reddit.
Despite the crudity of the animation, that’s pretty much also my best prediction as to what happened. The exhaust must have charred those containers, and clearly something impacted some of them. Probably a few bits fell off the rocket when it hit, the rest of it falling off the side.
If the barge is rockin; don’t come knockin’
No videos, but a few sweet pictures (originally from Twitter).
Looks like when the fins ran out of juice, they “safed” in a full-displacement orientation. Rocket wasn’t moving that fast at that point so it wasn’t immediately catastrophic (like it would be if it were still hypersonic), but still enough to overwhelm the vectoring on the main engines. Rocket hits at 45 degree angle and… boom. Jebediah Kerman is pleased.
Brian
Yeah, that phrase made me laugh too!
Thanks for the update and links, Dr. Strangelove.
I think I’m naming my next ship in Kerbal Space Program the RUD 1
Just wondering about the landing navigation, is it blind (i.e. GPS coordinates) or does the rocket use sensors to position itself over the platform for landing?
Video of “landing”
Brian
So awesome. Oh, how many KSP landings have ended that same way. Good thing the stage is mostly empty at that point! As impressive as the explosion was, it’s nothing compared to a fully fueled rocket exploding, like with the Antares a month or so ago.
Can’t wait until the next attempt, just a few weeks away. Hopefully the hydraulic fluid really is the last fix they need.
As far as anyone can tell, it’s mostly GPS based. They might have a radar altimeter to get a precise height above the surface, but they don’t seem to be using cameras or the like for landing. Probably wouldn’t work all that well anyway, since they have to land in the dark, in fog, etc. Musk has said that the legs have sensors on them but they might just be contact/pressure sensors.
At one point during the webcast they say “Landing platform as received acquisition of signal”. So there’s at least some kind of communication between the rocket and platform, though I doubt the platform can do anything for the landing besides stay in place. Might just be a telemetry or video link. The platform may also have differential GPS transmitters for better precision.
A little piece of possibly interesting trivia: a Falcon 9 first stage, when empty, has an almost identical density as an aluminum soda can. That includes the engines, so the tankage is actually (relatively) much lighter than the can. Pretty astonishing that such a flimsy piece of machinery can withstand such tremendous forces.
Wow. That thing took off like a rocket when it exploded.
I wonder how far it traveled before it hit the water.
Well, it doesn’t have to be an optical guidance system. Airplanes use instrumental landing systems based on radar and/or radio beacons, for example, and they can be accurate enough for carrier landings.
Differential GPS is good to a couple centimeters. That’s all the close-in nav you need. US Navy is researching doing fully automated aerial refueling where the receiver will use D-GPS to find the basket waggling around on the end of the hose.
Here’s what it looks like being done manually: F-18 Air Refueling - YouTube
Damnit, Elon. I keep waiting for you to do something annoying so I can end my man-crush, but then you go and do this:
Repairs almost done on the spaceport drone ship and have given it the name “Just Read the Instructions”
West Coast droneship under construction will be named “Of Course I Still Love You”
(pours out bottle of RP-1 to mourn Iain M. Banks)
Brilliant!
“Whoops! Had the silly thing in reverse” would have been even better though.
There are so many fantastic options.
Just Testing
Only Slightly Bent
Funny, It Worked Last Time…
Ultimate Ship The Second
Death and Gravity
I’d have thought you’d want “Dear John” and “Hi There”.
