Elon and SpaceX Taking Twitter Suggestions to Fix Starship Landing Problem

by

Elon Musk is taking some Twitter suggestions to light all three engines on the Starship landing. Two Starship prototypes (SN8 and SN9) have crashed on landing.

By default, engine with least lever arm would shut down if all 3 are good.

Starship SN10 is on the launch pad and is days away from its first tests. Elon Musk and SpaceX will make adjustments to prevent engine failures and to make the Starship safer in case of engine problems.

SOURCES – Spacex, Twitter, Elon Musk
Written by Brian Wang, Nextbigfuture.com

42 thoughts on “Elon and SpaceX Taking Twitter Suggestions to Fix Starship Landing Problem”

  1. Depends. If the crash breaks the rich chocolate layer and exposes the creamy nougat at the heart of a Raptor engine then the crash is the way to go.

  3. It seems odd to me that the "suicide flip" maneuver has to go past making the ship vertical and then twist back.

    I believe I've read that this is done to kill the extra velocity added by the horitzonal thrust component of the engines as they do the flip.

    But it should be possible to have just enough horizontal velocity that flipping to vertical would cancel that velocity just as it arrives over the landing pad?

  4. I've read a suggestion that maybe the igniter for the second engine – basically a sparkplug – didn't work for the re-ignite.

    So having redundant instances of that might make sense, without adding a lot of mass or massive ship redesign. But there'd be no time to use one and then try another if the first didn't work – you'd want to use both in the first place.

  6. They have previously considered a very similar problem – i.e. they've flown hoppers with simulated mass added to make them more realistic.

    It would be pretty trivial for them to have just included extra fuel and burned off just enough while hovering to achieve a particular mass at landing. In fact, a thermal video of the explosion video showed a lot of left-over liquid oxygen on board SN9 – which is what you'd probably want to use for ballast to avoid making any crash extra boom-y.

    Also, they'll always have to deal with some degree of mass variance on landing, as they'll have more or less fuel remaining. What they probably shouldn't have relied on yet was having all engines working at landing. As Elon said, trying to only light 2 was a "dumb" oversight.

  7. I'm surprised they're using cold gas thrusters. Those have terrible performance. I suppose it's because they're using the helium to pressurize the tanks, too?

    What ever happened to that helium/hydrogen/oxygen gas thruster?

  8. He's hardly anonymous on twitter, you know. Blue check means pretty much the opposite of anonymous.

    Strikes me as the sort of boss who's OK with being contradicted by employees if they can prove they're right. But don't contradict him without proof.

    Well, that's fine: If you're going to contradict the boss when you can't prove your case, you're an idiot anyway. Better have a good reason if you're going to be contradicting the guy signing the checks.

    I've had at least one boss who'd be annoyed if you proved him wrong, it isn't fun.

  10. you could be right, but that would mean he's a much more humble on twitter as seen above, usually anonymity does the opposite.

    my monies on they got 99 problems and only 24 engineers that can only conceptualize and deal with 4 potential problems at a time.

  11. right now they don't have any intact failed engines to study, if they manage to save one landing with a failed engine, they could learn a lot.

  12. this would be a problem, except the landing burn is powered by separate header tanks that are FULL at the time of the start of the landing burn.

    you can't shake a gas into a stored fluid if there is no gas in the container.

  13. good points, allow me to add to them, it's also lacking the weight of thermal shielding and "hot" gas thrusters, as well as a host of other items used to secure and move cargo.

    those "hot" gas thrusters will produce 5 times the thrust per weight of the nitrogen thrusters they're using now, this could be used to make more efficient or more powerful thrusters, the latter which could aid in flipping and keeping the the rocket upright when engine troubles strike.

  14. Likely not, unless they were to add gas fueled engines. Turbopumps aren't that forgiving, and the raptors aren't all that throttleable.

  16. If the Raptor engines can throttle down to 40% of thrust, then re-light all 3 at 67% thrust instead of only relighting 2 engines.
    I'm sorta-expecting the Starship SN-10 mission to do some "pogo-stick" and "anti slosh" tests.
    Load as much methalox as the 3 engines can lift off with, then take the rocket up (cue The Byrds '8 miles high') and then at 0 degrees from vertical cut the engines, let SN10 go weightless and drop tail down, do a ullage 'burp burn" using the header tanks to restart all 3 engines at 67% throttle, hot swap to main methalox tanks and climb back up to 8 miles. Repeat with SN10 dropping at 10 degrees from vertical, then 20, 30, 40, 50, 60, 70, 80, 90 until you find the angle where the ullage burn becomes ineffective.
    I'm curious if the vapors from a self-pressurized methalox tank can power a ullage burn to push the remaining liquid down to run the Raptors.

  18. There has been research on throttling solid rocket engines, using ultrasound to alter the rate of combustion, but it never got very far.

    Solid rocket engines have inferior ISP, and the rocket already has engines, they don't want to add the weight of redundant engines.

    They just need to solve the unreliable restart problem.

  19. There are maneuvering thrusters in the nose. Basically they are the same type of cold gas thrusters used for orientation on the F9 booster. The reason they don't use more powerful rockets is because they don't think they need to. Judging by these first two attempts they are correct.

    SN8 reoriented itself just fine. It crashed because the engines didn't provide enough thrust to land due to propellant flow issues. But it was oriented correctly.

    SN9 actually overshot vertical. This happened due again to engine thrust problems. The initial restart provided enough thrust to rotate SN9 to vertical, but almost immediately upon restart one of the engines failed and apparently there was not enough thrust available to halt the rotation and it kept rotating past vertical.

    SN9's thrust failure looks more like an engine malfunction than a propellant flow problem like SN8 experienced. I'm sure before long SpaceX will let us know.

  20. 1 million or less per engine is damn cheap, not expensive. Losing a few engines from financial standpoint is nothing catastrophic, it is testing. A potential design flaw would be much harder to fix.

    As I said before they would get more data if they succeed with 3 of them as they intend to do now(using 3 of them to land).

  21. Starting all three engines gives you a chance to collect data without trashing expensive engines, that's all. It isn't a long term solution, just a temporary fix to make the tests cheaper.

    Which would tell Musk more about why an engine didn't restart? Looking at the intact engine after a successful landing, or looking at the fragments after a crash?

  22. Unpopular opinion: I've the impression Musk is a challenging and hard to convince boss, with a risk for those dissenting of no longer work there, and that may result in growing an unhealthy yesmen mindset.

    If one man's opinion overrules the rest, and the organization weeds out dissenters, it will end up producing undesirable results.

    It has been mentioned before that contrary arguments can convince Elon, but if failing to do so marks you as incompetent or subpar, that's basically the same.

  23. Starting all 3 engines won't fix the main problem. Why the engine did not start, when it should? On the other hand they could at least get more data, intact engines if they succeed with 3 of them.

    I hope that it is not some design flaw and they won't need to redesign the engine, that would be time consumining and expensive to do.

  24. That's actually planned for the lunar version, to solve the two issues of the exhaust plume throwing dirt around, and because the main engines have too much thrust for a lunar landing.

    But they need the flaps anyway for their sky dive reentry, and would rather not add parts if they don't have to.

  25. Why doesn't Musk add retrorockets at the top of the spaceship to provide stability and steering? Its much heavier than the earlier designs.

  26. Wouldn't work on Mars, parachutes have to be enormous there to accomplish anything, and it would add weight. The maneuver is already feasible if they get reliable engine start, they just have to get past that.

  27. I wonder about the institutional environment at SpaceX, that they didn't originally do this. Maybe just a little TOO far on the take chances side of the scale?

    If this is a matter of fuel starvation, rather than unreliable start, starting 3 engines will only make it worse. That they're accepting the suggestion indicates that (they think!) they've gotten past the fuel starvation issue, and it actually IS unreliable starting.

    As I remarked on the other article, at T+1:10-1:15, on the SpaceX video, you can see that the turbo-pump on one of the engines is actually on fire. I've seen indications before that SpaceX's pumps leak; You can see on any of their Falcon launches that the protective wrap on the engine plumbing is inflating and letting out puffs of fuel and/or Oxygen.

    I wonder if the engine electronics or some other component was damaged by the fire, or the leak got bad enough to compromise starting? It would be interesting to know if the engine that caught fire on the way up was the same one that failed restart.

  28. You now have a re-entry shield that has minimal area and effect like dramatically reducing the size and effectiveness of brake pads AND it’s also dependent on mechanical deployment rather than being passive.

  29. The best part is no part. The best flip is NO flip. I believe the mixing that occurs durIng the flips, between the gasses and liquids, is what’s causing the turbo-starvation, and that’s the root cause of the problem. But If Starship had 8 Falcon 9 type legs and they were deployed downward 60 degrees they would approximate the shape of the IRVE-3 inflatable drouge cone. Then the canards and body flaps and flip maneuvers are no longer required. But what if each of the 8 legs had rigid panels mounted to each side that fanned out to complete the cone shape? Then Starship would have a reusable re-deployable re-entry shield, while REMAINING VERTICAL. No flips, no mixing, no RUDS!

  32. Isn't this a temporary problem or have they added ballast to the Starship test vehicles?
    The final version will have three more engines, which will add weight. That will make it easier to throttle with all 3 engines.

  35. What do you think they have been doing all these years and what do you think these low altitude flights are? Testing is exactly what it is.

  36. I think they might not be able to throttle a solid rocket engine at all. They just go until they are spent or permanently stopped. It might be called something like "blow out", eliminating the usable propellant.

  38. Starting 3 creates a larger acceptable failure envelope for each engine, which will then experience the Titanic lessons. Or the Shuttle lessons.

  40. You have to carry extra fuel anyway as the landing will require at least a minimum thrust so why not just use a reliable solid rocket jato

  41. The reason the rocket twists out of plumb is the thrust of th eengines slows the bottom of the rocket down while the top is still going fast. The low tech solution to this is a small parachute deployed when the deceleration begins. Just enough to keep the top from outrunning the rear. Pete

Comments are closed.