Images of Falcon 9 first stage crash


Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right or below. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.

Elon Musk has released images showing the final seconds of the Falcon 9 first stage as it attempted to land vertically on a floating barge last week.

The rocket can be seen tilted as it falls and explodes. I am more amazed that they got this close to a success on their very first try.

Share

11 comments

  • BSJ

    Are we sure the rocket was tilted, or was it the boat?

    How in the world could fins have had anything to do with this? Fins only work with airflow. What sort of control could they have given to a slow moving object like a descending rocket?

  • PeterF

    They said it “ran out of hydraulic fluid”. What kind of system would they use that exhausts its operating fluid? I’m more inclined to believe it went gigo fuel twenty feet off the deck. I’m sure they could tell us exactly what went wrong and anything we imagine is mere conjecture until they release the video.

  • PeterF

    ok so please disregard everything I said after “duhhh”. perhaps thrusters near the top would work better than fins at the bottom?

  • PeterF

    Obviously it had enough fuel to arrest its vertical motion. It kind of looks like the rocket just missed the mark and pitched over in an attempt to move back over the platform. The control programming should limit it from deviating so far from the vertical once it is in close proximity to the ground. The engines probably can’t gimbal 45 degrees to provide the proper thrust vector so they have to tilt the entire rocket. I doubt there was enough fuel to recover once it was so far from vertical so close to touchdown. It was moving pretty quickly horizontally.
    It probably would have been a perfect landing had it had a larger target to aim for. (Like a 1000 acre spaceport)

  • BSJ

    You seem to be assuming that the platform was motionless, and exactly horizontal. Neither of those two things are possible on a small floating platform in the middle of the ocean. Looks can be deceiving; with no horizon visible to the camera there is no way to judge how much the deck was actually moving.

    Aircraft carriers are BIG. Watch a carrier landing video and pay attention to how much the deck pitches and rolls when the planes comes in…

  • Tom Billings

    OK, guys. It’s usually best to read a bit before you write.

    1.) The fins *are* used from hypersonic to subsonic speeds, and help to adjust attitude. They worked fine. They are on an *open*loop* hydraulic system, because that is far lighter than a closed loop system. So, it worked well for about 240 seconds, and then ran out of fluid for about the last 24 seconds of the descent.

    2.) That last 24 seconds is more crucial in these landings, because the Merlin1D engine only throttles down to 70% thrust. That is much greater than the empty weight of the first stage, so it cannot hover. Thus, it comes down fast and has to kiss the deck as the engine stops. When the fins stopped working, the attitude depended entirely on the cold gas RCS and the engine to maintain attitude for more of the last 24 seconds than had been assumed in the software limits. The attitude was thus wrong as the stage came into”kissing distance” of the deck, and the result was seen in the video.

    3.) The barge has precision position and attitude control from water thrusters, so the deck was stable there. The only instability was its height changes. Those rates of change could be measured, and input to the guidance software, which computes the results as part of “kissing” the deck with a vertical rocket stage. Except, it wasn’t vertical, this time, and we saw the results in the video.

  • Pzatchok

    The hydraulic system uses pressurized fuel as the hydro fluid and a nitrogen can as the pressuring agent.

    It either ran out of nitrogen or the fuel tank went low.

    But for what ever reason it could not stay vertical.

    My best guess would a combination of a little wind and out of position fins dragging its top off to the side as it fell.
    The engines trying to keep the craft vertical would then start the whole craft sliding sideways, and I bet the software fought itself trying to stay over the ship.

    It might have went perfectly fine all the way down and then a side wind could have blown it off target and since the fins do nothing at slow speed they couldn’t help anyways. The engines can hover the craft but they might have had to much wind to compensate for and eventually ran out of gas trying to stay over the target.

    The fact it hit the ship at all was great.

  • BSJ

    How could those fins impart any control to a rapidly decelerating rocket? The slower it goes, the less effect they have. Even if they had been working the last 24 seconds what difference could it have made?

  • mpthompson

    I think you are under estimating how fast things must occur just before the stage lands. At 24 seconds before landing the stage is still very high up and moving downward at a rapid velocity. The fins are still providing a significant amount of control over the attitude of the stage until just a few seconds before landing. Because of the very narrow throttling range described above, the final application of thrust to bring vertical velocity to near zero only occurs in the final few seconds. If attitude control is lost via the fins before these final few seconds and the rocket drifts too far from its target, say more than 20 or 30 meters, there is insufficient time and altitude for the engine under thrust to correct the error.

    I believe this failure scenario is what we saw in the video. There was simply too much accumulated horizontal error to correct when the engines were finally fired up to reduce vertical speed. We saw the control algorithms valiantly try to reduce the horizontal error, but error was too great. The rocket was already doomed before the engines were ignited for the final burn as it had to literally dive for the platform to reach it.

    If there is sufficient hydrolic fluid to control the fins and rocket attitude until a few seconds before landing, I believe the results will be much better.

  • Edward

    I am in agreement with mpthompson. Please note in this video of the Falcon9R that the fins are deployed and active all the way to the pad, including the last 24-seconds.
    https://www.youtube.com/watch?v=xh9wCq-RJ8U#t=108

  • Edward

    On further reflection, It seems to me that there is more tweaking to be done to the software. The rocket seemed to have waited until far too late to try to correct for being well off of the target, “panicked,” and over-steered. Whether they were not prepared for a fin-failure mode, a gust of wind, or whatever the problem was, SpaceX needs the rocket to make better corrections in a more timely manner, no matter what problems arise. (I am reminded of Google’s self-driving cars, which are still under test after all these years.)

    But this is the nature of test engineering during a development program, and why Musk and SpaceX were pessimistic that this test would be a success. I believe that they came closer than anyone before in reaching a specific target. Usually we see proposed landing sites as large ovals on a map, but SpaceX was able to leave debris on a 1-acre(ish) landing pad.

    I am glad that they are willing to share their progress (read: failures) with us, even when it could be embarrassing. Some companies are more secretive. Hooray for Musk in helping the world to know that it is difficult — that rocket science is still rocket science — even though some of their Grasshopper tests have made it look easy.

Leave a Reply

Your email address will not be published. Required fields are marked *