Business, Media Criticism, Space, Technology and Society

Successful Flyback

30 Comments

…and failed landing. That’s what flight test is about. They’ll learn from it, as they always do from a failed attempt.

I would note, though, that this does complicate their operations, if they plan to land down range every time, and can’t return to launch site. I suspect they’ll determine that the problem was crappy weather conditions, and their FLIR or whatever they were using for guidance wasn’t doing very well. That means that there’s a new condition imposed on a decision to fly — weather at recovery site. Shuttle often scrubbed with good weather at the Cape, due to unacceptable conditions at abort sites, and that was just for contingency. If SpaceX wants to recover down range, they may occasionally have to make a decision as to whether to risk the loss of a stage, or delay and arouse customer ire. It will depend on whether or not there’s a tight window (e.g., a planetary mission), and who the customer is.

[Update a couple minutes later]

Oh, I hadn’t read to the end. It sounds like it wasn’t a weather problem — they “ran out of hydraulic fluid” (not sure what that means — it’s not a closed system?). But that seems like good news, both for their chances of recovering next time, and for being able to operate in less-than-perfect conditions. Sounds like they only thing that might prevent a launch, in terms of barge conditions, would be sea state (or high winds), not weather per se.

[Update a few minutes later]

Here‘s what looks like a reasonable explanation from Jon Goff. I haven’t read the post itself yet, but I’m sure it’s worthwhile to do so.

[Update a few minutes later]

OK, Elon just tweeted that it was hydraulics for the control fins, and they came within 10%. So that means an excellent chance of success the next time, with the addition of a little bit more juice.

[Update a few minutes later]

30 thoughts on “Successful Flyback”

  2. Remember their first soft splashdown attempt? They didn’t have the landing legs and the nitrogen thrusters weren’t able to control the roll rate, leading to fuel starvation. They put on the legs and increased the capacity of the thrusters. Now they’ve added grid fins (built by my company!) and ran out of hydraulic fluid. They’ll increase the fluid capacity for the next attempt. One great thing about SpaceX is that they learn quickly and don’t seem to have the same problem twice.

      1. Some claim that there were separate problems with the helium leaks. I don’t know. We can lump helium leaks into one broad category and admit that it took them 2-3 tries to get them fixed, or if the leaks were in separate locations, see each one as a separate problem that was fixed in turn.

        As for the second stage actuator, I don’t recall that being an issue with any of their previous flights. We’ll know if they fixed it if it doesn’t recur on future flights.

  3. Cannot say that this was unexpected. Losing the rocket to a hard landing. Like I said here before it is not like the first stage structure is that robust to begin with. Still they did hit the landing platform so at least the telemetry, control and guidance worked reasonably well. This is probably just a matter of tweaking the system until it works. As for the cameras not working to capture a night landing they could put some IR cameras in there. At least the exhaust plume should be highly visible in IR. The difference between a visible light camera and an IR camera is basically zilch to begin with. It is simply a matter of not filtering the IR spectrum as all modern visible light camera sensors are sensitive to IR.

    The weather conditions are a problem but then again this is an unmanned vehicle so as long as vehicle losses are economically acceptable it does not matter.

  5. If the next mission nails the landing, I’m still not sure they’ll re-use the stage because it should go to the Smithsonian as the first rocket to fly into space and land under its own power.

    As a side note, I was extremely frustrated to watch the launch on NASA TV and not find out anything about the first stage landing attempt. As far as NASA’s website was concerned, there was no such attempt, probably because such a thing isn’t possible. At one point I felt like I was in the Soviet Union awaiting the non-news about the latest non-attempt of their attempt to beat the Americans at something.

    1. Well, actually, NASA (at least the part of NASA that wants to get stuff to ISS) just doesn’t care what kind of post-staging experiments Elon wants to perform, as long as their stuff gets to ISS.

  6. I didn’t expect success on this mission, because so much of it was based on modeling and assumptions; they’d never tried the grid fins and high speed, let alone the hypersonic regime they experienced on this flight (To the best of my knowledge, this was the only time grid fins have flown on an operational F9, and the only actual flight experience was with the now-lost F9r DEV-1 in texas – one very low speed test.) There were a great number of ways the real world could prove different, and it looks like one did; the usage of hydraulic fluid on the open system that controlled the grid fins.

    So, SpaceX, IMHO, attempted this landing facing a huge number of potential failure points, because so much of the flight regime had been modeled, rather than based on actual test data. They, perhaps, could have avoided this risk had they tried a few more water landings with the new configuration and flight profile, plus rushing F9R Dev-2 into service in New Mexico.

    And that’s the point; they didn’t, because they had no reason to; this F9 stage would have been destroyed by a water landing. Therefor, the only risk they took was to the ASDS (barge). Against that risk was a full test of the system. If they learn from it, it’s well worth it – that’s why you do do experiments.

    I think that cost-lowering re-usability would be the most important advance in the history of spaceflight, so I think what SpaceX is trying to do is profoundly important. I sincerely hope they succeed, and I think they’;re going about it the right way.

  7. their FLIR or whatever they were using for guidance

    SpaceX has not seen fit to reveal, AFAIK, just what kind of guidance system they’re using to bring the first stage back. I have a bet with myself that it’s one of the fairly standard forms of augmented GPS, but it would be interesting to know for sure.

    1. I suspect it is that and maybe some sort of optical tracking system on top of that if they need extra precision.

      I mean they use it in the Dragon capsule.

      1. Not sure what you mean by “GPS augmenters”. Assuming that the barge is well anchored, its position in Earth coordinates is known accurately, and a differential GPS system – homegrown or a commercial service such as Omnistar – will give a horizontal position accuracy of better than a meter (1-sigma). This of course assumes that the booster is using GPS; there are certainly antenna placement challenges but that should be solvable. I don’t see anything about this test that would require any navigation more sophisticated than C/A code + DGPS; it’s nowhere near as stressing as Northrop Grumman’s X-47B fully autonomous carrier landings. I don’t think they would need an optical sensor.

        What I’m most curious about is what the root cause of not enough hydraulic fluid will be; running out of hydraulic fluid implies more control usage than predicted, which could be from lots of reasons – everything from environmental factors to less control power than predicted to a previously unknown lightly damped mode. With all due respect to Elon, his engineering team has to understand the “why” before they just put more hydro fluid in the tank. if Elon were my boss, I would certainly get frustrated when he goes full cowboy like this.

        1. I basically meant DGPS. I suspect it doesn’t really matter what Elon said. They’ll do an analysis, and make whatever fixes are necessary, if they go beyond simply having more margin, before attempting it again. On the other hand, they have little to lose from repeated attempts, and everything to gain.

          1. Agree with what you are saying. I’m perhaps overly sensitive to upper management shooting their mouths off without all of the data because I’ve had to walk back their hasty statements too many times :-/

        1. Also this, from 2004:


          http://www.navsys.com/papers/04-04-001.pdf

          The SRGPS architecture provides a precision approach and landing system capability or shipboard operations equivalent to local differential GPS systems used ashore, such as the FAA’s Local Area Augmentation System (LAAS). A relative navigation approach is used for SRGPS with the “reference station” installed on a ship moving through the water and pitching, rolling, and yawing around its center of motion. In addition, the ship’s touchdown point may translate up/down (heave), side to side (sway), and fore and aft (surge).

          Since the shipboard landing environment is much more challenging than ashore, the
          SRGPS approach must use kinematic carrier phase tracking (KCPT) to achieve
          centimeter level positioning relative to the ship’s touchdown point.

          1. SRGPS is what the X-47B uses for fully autonomous carrier landings, yes. But I don’t see that you need that level of sophistication to land on a stationary barge. Or a site on-land.

  8. I only got a couple hours of sleep last night. I woke up well before the time my alarm was set and couldn’t get back to sleep. Luckily for me it was on a weekend, so I had the luxury of going back to bed afterwards.

    I’ve been wondering when they were going to build a new test vehicle to replace the F9R that was destroyed last year. I think we have our answer. They’re so close to achieving a landing with actual flight hardware that further testing may be unnecessary.

    Back in the 1960s, space launches were exciting because each mission built on the experience and lessons learned from the previous one. With each new flight they pushed the envelope a little further. That is why SpaceX launches are so exciting, to me at least, compared to the other mature launch systems. They are doing the same thing now.

    Every other launch provider: “Was the launch successful? Good. Let’s get ready for the next one.”

    SpaceX: “OK, that worked. Now let’s try this.“

    1. Fully agreed – in the early years, everything was incremental development out of necessity.
      And this also applied to the actual spacecraft, not just rockets. Surveyors, Luniks, Mariners and other series were all technology development first and foremost, science second.
      We have almost completely lost that, space technology development is at comparative standstill today.

    1. My guess; some will portray the failed landing as a failure of the mission (ignoring the Dragon on its way to ISS).

      What I find interesting is stunning change in attitude at ESA; they went from ridiculing the concept of reuse to embracing it, and did so over a span of a few months. That tells me that they think SpaceX has a good shot at making reuse work. (and by work, I mean economically.)

    2. Just look at the title of the article Rand linked to, at Spaceflightnow.com of all places, calling it a “botched landing”.

    3. It wasn’t a change of attitude. Arianespace pans what SpaceX is doing because they are commercial competitors. CNES is a France national space research center so they have a different perspective on it. I bet if you asked someone at ESA they would have a similar perspective to CNES.

    4. Andy Pasztor at the WSJ (or is it Boeing) is the source of the “botched” comment. There is actually some good stuff in his article — and the commenter are giving it to him for the unfair parts — but the headline (which I understand he didn’t write) and lede are ridiculous:

      SpaceX Stumbles in Test to Show Reusable Technology Advances

      Elon Musk’s space company executed a flawless launch of an unmanned cargo capsule headed for the international space station, but botched a bid to maneuver a used rocket booster to gently touch down on a floating platform.

  10. Thank God for Elon Musk and SpaceX. Without them, all I’d have to look forward to for human space exploration would be SLS. The SpaceX story is absolutely riveting, and I hope to be around long enough to see a SpaceX flag flying on Mars. (No offense to the suborbital folks, who are also doing great things; but I would love to see bootprints before I die.)

  11. Good Try and Kudos to Elon for making the barge landing attempt. I hope the mathematics for fuel vs. payload on the next trip really shows enough margin to make this work.

  12. Anyone have opinions on if SpaceX will release whatever video they have of the attempted landing? Sounds like there isn’t much, but all of us would love to see it. I think they need to play to their fans.

    I’m confused by the idea of “not enough hydraulic fluid for the waffle fins”. How does that lead to a “hard landing”? Are the fins actually used to slow the descent, or mostly for guidance? I would have thought it would lead to missing the platform.

    1. Musk said that it was too foggy at the landing site to get good video. All it may show is glare from the rocket engine. It might still be interesting to see if not overly informative.

      As for the grid fins, when they ran out of hydraulic fluid, they may have been stuck in a position that imparted roll or other inputs. The guidance computer would be trying to command the fins to move and perhaps having to fire nitrogen thrusters (assuming they’re still on the first stage) to compensate. The grid fins are primarily for guidance but they may also impart some drag. I know my company machines the grid fins for SpaceX but don’t know if we make the hydraulic actuators or other parts of the system.

  13. For landing guidance they employed a retired US Navy carrier LSO.

    Their report card read
    “Landed and taxied to the #1 arresting wire. No grade.”

Comments are closed.