Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Space

SpaceX Launch Abort Test Successful 54

An anonymous reader writes: As we discussed yesterday, SpaceX launched a prototype this morning to test its Dragon passenger capsule in an aborted launch. The test was a success — the capsule separated cleanly, propelled itself to a safe distance, deployed its parachutes, and lowered gently down to a water landing, where it remained floating. You can watch video of the test on SpaceX's website — skip to 15:40 to get right to it. Externally, everything seems to have gone fine. I'm sure we'll hear in the coming weeks whether the downrange distance was ideal, whether they hit their splashdown target, and how the crash test dummy inside the capsule weathered the abort!
This discussion has been archived. No new comments can be posted.

SpaceX Launch Abort Test Successful

Comments Filter:
  • Saw it in person (Score:4, Interesting)

    by trout007 ( 975317 ) on Wednesday May 06, 2015 @09:09AM (#49629081)

    It was like a big model rocket launch. Quick burn, coast to apogee, chutes deploy, and landed in neighbors yard.

    It was neat the engines shut off before I heard them start.

    • by TWX ( 665546 )
      My wife worked on the spin-motors for the NASA LDSD test platform. Lots and lots and lots of rocket engines do this. Ejector seat motors come to mind- they want the pilot up and out quickly and without burning him up inside of the cockpit. Quick puff and the chair is out. Admittedly the pilot is now almost three inches shorter from spinal compression, but he should recover at least half of that.
      • by Luckyo ( 1726890 )

        In space vehicle rescue systems, goal is typically to direct both lift and ejection acceleration force to the horizontal rather than vertical vector in relation to the human body. We have a far greater tolerance of such short term high G loads, one of the reasons being that it doesn't stress the spine in the way you describe.

  • Sort-of-worked. (Score:3, Informative)

    by queazocotal ( 915608 ) on Wednesday May 06, 2015 @09:28AM (#49629295)

    Close examination of the video shows that one of the near thrusters shut off. Look carefully and you see a puff of smoke, and one of the thruster clusters dims as one of the two superdracos has stopped thrusting.
    At the same moment, the vehicle begins to pitch.
    The thrust was perhaps then terminated early - the vehicle did not quite get nominal total velocity.

    • Re:Sort-of-worked. (Score:5, Informative)

      by bareman ( 60518 ) on Wednesday May 06, 2015 @09:39AM (#49629421) Homepage Journal

      Pitch is part of the plan. Escape systems need to move the capsule out of the path of the rocket below it.

      "T+.5s: After half a second of vertical flight, Crew Dragon pitches toward the ocean and continues its controlled burn. The SuperDraco engines throttle to control the trajectory based on real-time measurements from the vehicle’s sensors."

      • Re:Sort-of-worked. (Score:5, Insightful)

        by photonic ( 584757 ) on Wednesday May 06, 2015 @10:39AM (#49630027)
        No, OP's remark was correct, he is not discussing the planned pitch over towards the sea after half a second, but a visible puff of smoke around the 4 second mark and some subsequent 'wobble', as if one of the 8 thrusters is switched of. See the zoomed video here [youtube.com]. This proves the effectiveness of having 8 redundant thrusters, instead of having only 4. It is still not clear if this was a real engine failure (which might be verified via post-mortem examination of the motor), or if this was a deliberate, unannounced test of the 'engine out' capability. Credit: the discussion in the forum over at nasaspaceflight.com [nasaspaceflight.com].
        • by bondsbw ( 888959 )

          Just throwing out a guess, but perhaps that was part of the test? A mechanical flaw may have been introduced into the engine in order to test that the rocket can appropriately decide when to abort, and also how to recover.

          • They want to re-use this Dragon for an in-flight-abort test, using a Falcon 9 first stage launched from Vandenberg, so purposely introducing mechanical flaws seems risky. If it were a deliberate test, I would be simplest to just set the throttle to one of the motors to zero in software close to the end of the burn (to not ruin the overall test in case of problems), and then see if the control system can recover using the remaining 7 engines. Let's hope that Elon comments on this in the next days ...
        • Re:Sort-of-worked. (Score:4, Insightful)

          by Bruce Perens ( 3872 ) <bruce@perens.com> on Wednesday May 06, 2015 @04:03PM (#49633129) Homepage Journal

          What I am getting from the videos is that this test was a success but that there was indeed an engine failure and the system recovered from it successfully by throttling off the opposing engine. There was less Delta-V than expected, max altitude was lower than expected, downrange was lower than expected, and that tumble after trunk jettison and during drogue deploy looked like it would have been uncomfortable for crew.

          This is the second time that SpaceX has had an engine failure and recovered from it. They get points for not killing the theoretical crew either time. There will be work to do. It's to be expected, this is rocket science.

          It sounds to me like the launch engineers were rattled by the short downrange and the launch director had to rein them in.

    • by Anonymous Coward

      It was supposed to get clear then go sideways (to clear an rocket that is about to explode under it) - pretty sure that was intentional since it can only control it's direction by throttling engines asymmetrically.

    • thus the redundancy of the thruster arrangement, call it an additional successful test of a feature
    • Musk mentioned something about a thruster underperforming due to a poor propellant mixture ratio, but the Dragon was also controlling its thrust to direct its trajectory out to sea, so either could be the cause of the "puffs".

      It came in surprisingly close to shore, but was also being dragged quite quickly by the wind before the parachutes finally collapsed. The altitude/speed reached are probably better indications of the actual performance.

  • If I ever ended up in one of those, I'd imagine I'd be grateful and all it probably saved my life...


    ...but I'd probably also probably be sick as hell. That doesn't look like a smooth ride at all.
  • by mlosh ( 18885 ) on Wednesday May 06, 2015 @09:30AM (#49629319)

    The landing was a bit closer to shore than expected, but probably due to high on-shore winds, and splashdown was 8 or 9 seconds early. Video seems to show one of the "SuperDraco" engines shutting down a bit earlier than the others. Still, very successful overall!

  • by Anonymous Coward

    From space-x sub-reddit:

    https://www.reddit.com/r/spacex/comments/34yote/rspacex_dragon_2_pad_abort_live_discussion/cr073cj

    Looks like it landed just a bit further off-shore than the unmanned Mercury capsule from an Atlas 3 failure April 25, 1961: https://youtu.be/Vp9BnBDKa0s?t=5m55s Flight terminated after 43 seconds, LES tower ignited, pulling capsule free. Apogee of 7.2km, downrange only 1.8km. Capsule recovered and used again.

  • by SylvesterTheCat ( 321686 ) on Wednesday May 06, 2015 @11:09AM (#49630341)

    >> Wednesday's test was conducted at Cape Canaveral in Florida, and saw a test vehicle - carrying no humans, only a dummy - hurled skywards by a set of powerful in-built thrusters.

    Strange. I don't remember reading anything about there being a member of Congress on board..

    • >> Wednesday's test was conducted at Cape Canaveral in Florida, and saw a test vehicle - carrying no humans, only a dummy - hurled skywards by a set of powerful in-built thrusters.

      Strange. I don't remember reading anything about there being a member of Congress on board..

      Now that's just insulting to crash test dummies everywhere!

  • by glwtta ( 532858 ) on Wednesday May 06, 2015 @11:15AM (#49630395) Homepage
    If you fail a launch abort test, does that mean you had a successful launch?
    • No.

      The purpose of the launch abort systems/procedures is to save the crew if things go badly wrong during the launch. The purpose of the launch abort test is to test those systems and procedures so that any problems are sorted out before they have to use the procedure for real.

      A successful lanch abort system test is one that gets the crew pod away from the launch vehicle (or a standin for it) and brings the pod back to earth while keeping conditions inside the pod (g-force, temperature etc) within limits. A

  • >> the capsule separated cleanly, propelled itself to a safe distance, deployed its parachutes, and lowered gently down to a water landing, where it remained floating.

    Wait, the author of the article thinks the fact that it floats is the most amazing part?

    • I doubt they think this is the most important part, but would it not be a serious problem if it didn't float? It'd indicate some serious problems in the supposedly space-going capsule!
    • Well, I was kind of surprised to see how high the craft was riding in the water. Perhaps it is the higher wall angle of the dragon compared to the Apollo capsule,
    • by Agripa ( 139780 )

      Most people understand physics through Hollywood where cars sink immediately.

      • by JustNiz ( 692889 )

        yeah and if (even one of) their engines stop, planes immediately drop out the sky like a rock.

  • Are these the same 3D-printed engines they discussed recently? If so, you think they'd have mentioned that little first-flight detail.
  • So, they successfully tested not launching a rocket? That's basically what I do all day long!

    (Yes, I know. But it's funnier this way)

  • remember the Atari game? In one of the missions you had to launch from the surface. If you just went full throttle straight up (until the LM went offscreen), the launch profile looked just like this Dragon test.

"Your stupidity, Allen, is simply not up to par." -- Dave Mack (mack@inco.UUCP) "Yours is." -- Allen Gwinn (allen@sulaco.sigma.com), in alt.flame

Working...