Sunday, January 17, 2016

SpaceX Falcon 9 Muffs Landing Attempt at Sea

After last month's successful return and landing of their Falcon9 booster on the Kennedy Space Center, SpaceX upped the ante today and attempted to land on a barge at sea after a launch from Vandenberg Air Force Base in California.  They got closer than they have before, but the rocket fell over on the barge.  Elon Musk went to twitter to explain that of the four legs required for landing, one failed to lock in position, so that when the rocket leaned onto that leg, it fell over.  Initially, it was thought that they landed too hard and broke the leg, but later said the leg had never locked properly. 
 "After further data review, stage landed softly but leg 3 didn't lockout. Was within 1.3 meters of droneship center," SpaceX confirmed later.
Later he posted this picture of the Falcon9 lying on its side on the barge, saying "Well, at least the pieces were bigger this time!"   (I think I'd wear that on a Tee shirt!)
One report said the seas were 10-13' , which seems like it has to affect the landing.  Think of the ship rising and smacking the legs of the booster just as its ready to gently touch down.  Or think of the booster calculating where it's going to land and the ship being 6' lower than the calculation because of a wave.  
"Definitely harder to land on a ship. Similar to an aircraft carrier vs land: much smaller target area, that's also translating & rotating," SpaceX CEO Elon Musk tweeted.
Each attempt at landing on a barge has gotten a bit closer to sticking the landing, so I fully expect them to get it figured out soon.  You can see the engineering getting better.  The drone ship landings are desirable because they allow the biggest portion of fuel in the booster to be used for the launch.  To return to the land at the launch site requires more fuel. I believe they'll get there. 



9 comments:

  1. Close, but no cigar this time.

    Well...maybe an exploding cigar....

    ReplyDelete
  2. I haven't found anyone who can answer whether or not they are using a continuous dynamic landing solution, or whether they take a snapshot of it and land. If it's dynamic, what is the update rate? The barge doesn't move all that quickly, and surely they are using a small radar or stereoscopic vision to calculate the relative orientation of the rocket body to the landing plane. But how fast is this loop? Is it at least twice the Nyquist frequency of the barge motion?

    Inquiring minds want to know...

    ReplyDelete
    Replies
    1. I'd like to know those things, too. Of course, I don't have any contacts inside SpaceX, I just read what's published.

      The motion of the barge is going to be almost "forever" compared to the speed of the cheapest electronics. We agonize over nanoseconds, the period of the waves has to be around 15 seconds.

      I don't know what speed they land at, but I suspect "a few" feet per second.

      Delete
  3. If they are smart, they do NOT want to land on the barge to "save fuel". They were forced to land on the barge this time because they didn't get permission for a land landing in time. Barge landing at the Eastern Test Range was the baseline for early launches due to concern about them dropping the booster in "downtown" Cape Canaveral. If they have a problem during the return, the booster has to be destroyed BEFORE its trajectory would dump it in the wrong place. And that is very difficult to do.

    Landing the booster on a platform at sea exposes the engines to significant salt spray. And that is not good for hot metal that has to be able to withstand the next launch.

    ReplyDelete
    Replies
    1. Huh, I hadn't thought of that. I wonder why they're so (relatively speaking) focused on successfully landing on the droneship (so much fun to say!) then. Weird. Slightly off topic, is it just me, or does Elon Musk make anyone else think of a less-badass Tony Stark?

      Delete
    2. I think RDJ says he based his version of Tony Stark on Elon Musk. Exaggeration is the essence of theater, so he took Elon and went over the top.

      When I was trying to bring the character of genius billionaire Tony Stark to the big screen in Iron Man, I had no idea how to make him seem real. Robert Downey Jr. said, ‘We need to sit down with Elon Musk'. From this

      Delete
  4. One of the big issues with aerospace engineering is "fault tolerance". When you're aloft and things "go south", you can't just pull over to the side of the road and wait for someone to tow you home. SpaceX just showed it has ZERO fault tolerance in its landing system. Now that failure only damaged THEIR toys, but how many OTHER zero fault tolerant systems do they have in their vehicle? What are the credible failure modes for each element of their launch system? What backup systems (if any) do they have to "save the day" if any individual credible failure occurs? How do they verify shortly before launch that those systems work properly? And what are the worst-case consequences possible if the failure occurs? On launch, one of the standard failures postulated is a "hard over" by any one engine controller on the vehicle. The vehicle destruct lines - where Range Safety sends a destruct signal to make the vehicle turn into tiny pieces - are set so that there is no credible situation when the vehicle reaches those limits where pieces parts (or fireball) would reach an area occupied by people who are NOT associated with the launch. That was my concern about the most recent launch from the Cape. The vehicle appeared to fly a much more vertical path than normal, and it appeared that a thrust vectoring failure during climb out would have the potential for putting pieces parts on T-Town even if Range had quickly destroyed the booster. Not to mention the risk to Cape Canaveral (including the Trident basin) if the vehicle got a hard-over command during flyback. Under the normal launch review process, SpaceX would have to convince Range that the risk was sufficiently low. Hopefully they did so. Hopefully there was NOT political direction for Range to let it go anyway. Not that anyone in this country's government would EVER consider doing any such thing...

    ReplyDelete
  5. Even if everything on the booster had worked exactly as it was supposed to work, expecting something of that size and shape to remain standing on a (relatively) small barge (as opposed, for example, the deck of an offshore oil platform) in 10'-12' seas is ridiculous. I don't know how many of you folks have been at sea on a boat under those conditions, but landing something like that on a vessel with the pitching and rolling that would be going on would be a miracle. Believing it could remain upright for more than a brief moment without a gantry arrangement of some sort is unrealistic.

    ReplyDelete
  6. Barge is 300' long by ~100' wide. I believe it is anchored so the bow slews into the wind, and therefor largely into the prevailing waves. Whether the vehicle can remain upright depends on where the center of gravity of the vehicle is located, and the effective sail area of the vehicle as well. As long as the vertically projected cg remains within the footprint of the landing gear, the vehicle will remain upright. Once it goes OUTSIDE that footprint, the vehicle will topple.

    I believe that the cg at landing is rather low. Most of the fuel and oxidizer should have been burned by then, which leaves the nine engines with their pumps as a significant fraction of the mass. I don't have inside contacts with SpaceX, but expect they would have designed the barge to be able to safely land the stage in any seas that would normally be expected from winds within their landing limits.

    ReplyDelete