66 thoughts on “The Falcon Failure”

1. Capsule separation would have had a very good chance of working today; the problem didn’t seem to involve significant deflection or breakup of the vehicle until (presumably) the range safety charges broke it up ~10 seconds after the first visible anomaly.

   But yeah, all sorts of people are going to jump on this to advance their agendas. One hidden agenda to watch for: The people who’d prefer to shut down Station so they can divert that funding stream to an SLS-based exploration program may suddenly become overwhelmingly concerned with the safety of the next Station cargo launch.

   Can’t be risking an unmanned cargo carrier merely because a hundred-billion dollar asset is at risk, after all…

   1. Gwynne confirmed there was no destruct signal sent.

2. It looks as if something on the second stage exploded or at least violently ruptured shortly before staging, but wasn’t violent enough to take the first stage out immediately. I suspect that the launch abort system would have been able to handle it.

3. I saw the video first, then came here. It seemed clear to me the booster was fine and the event occurred in the upper stage, and apparently I was correct. Did it have an abort system on it that wasn’t there previously?

   1. Not that I’m aware of.

   2. Today’s launch was a Dragon 1 cargo-only capsule, which has no onboard propulsion sufficient to push it away from the booster in a hurry. The Dragon 2 crew capsule is the one that will have an abort system – 8 “superdraco” thrusters in four twin-packs around the outside of the capsule, about 130,000 lbs of thrust total, enough to push the crew capsule away from the booster at 6-7G’s or so.

      Gwynne Shotwell said during the press conference just now that if it had been a Dragon 2 with crew on board, chances are the abort system would have gotten them away from the booster just fine.

      1. Thanks Henry.

4. This is a second stage or stage separation issue. Not that uncommon Rand.

   The way the capsule seemed to get away suggests that a crew could have been saved by a LAS on this flight though. So yes this does indicate that spending time on the LAS is not a waste of time. But we already knew this from prior flight data. A LAS will save the crew in some failure modes.

5. Woke up to watch the launch live on NASA TV. I’m usually really nervous for the first few minutes (I remember the early Falcon 1 launches), but then relaxed as things appeared to be going well. Then BOOM! My wife heard my groan from across the house.

   The way I see it, such failures in this business are inevitable. What’s important is not that it happened, but how it is handled. I guess we shall see.

   BTW, 12 months ago would anyone have believed we would see three supply failures to the ISS from three different launch providers?

   1. Yeah. That’s why I said NASA needs three providers. Not two. Not one. Downselect is patent nonsense.

6. One other thing, this has to be devastating to ISS. I guess they still have ESA or JAXA for possible resupply. The normal NASA tendency is to ground operations until root cause is identified and resolved. I don’t think Progress is resolved, and now you have this with SpaceX.

   1. Forget ESA. ATV program is over. Either Russian Progress, ATV, or nothing.

      1. I mean Russian Progress or Japanese HTV.

   2. During the presser, the Station guy said they currently have four months supplies on board. He said that when it gets down to a month and a half, they start planning for crew evacuation. (He said they like to keep six months worth on board normally.)

      There’s a Progress scheduled to go up next Friday, plus a Japanese HTV (the last?) schedule for August sometime, so they have two more chances to resupply before things get tight.

      Also mentioned: The next Cygnus flight (on an Atlas 5) is currently scheduled for December, but they may be able to move that up to as early as October if there’s a need. (NASA people don’t tend to give such answers unless it’s already been looked at pretty thoroughly, FWIW.)

7. You can see what looks to be Dragon falling away. I’m surprised there was apparently no attempt to pop the chutes to at least recover the capsule and its cargo. Lots lost on this one, including a docking adapter, I believe. And to those celebrating, fetch off. Or something like that.

   1. I wonder how long it will take to replace that docking adapter?

      1. They probably have a back up. They planned to have at least two attached I think.

         1. OK, according to the press conference, there is a second one, and parts available for a third.

   2. If anyone is celebrating, and I don’t doubt some would, I would suggest they have more hate for commercial space than interest in a successful space program. Even if one thinks CEV, NASA, or Boeing’s CST is the best solution; you still have a short term emergency in keeping ISS manned. Unless someone believes the inability to keep a station in LEO supplied means we can move forward to lunar missions and beyond.

   3. Hope they have a backup adapter ready to go. That docking adapter is holding everything up.

      1. Press conference says they have a back-up adapter.

         1. Thanks, I saw some random guy mention that in this article.

            A backup is a big relief.

8. Would the launch abort system have worked for this event?

   I don’t see why it wouldn’t.

   Of course, we haven’t seen the Max-Q abort test yet.

   I guess it’s too bad they didn’t include it on this flight.

9. How does the bode for station re-supply?

   1. They have food supplies until the end of September and there’s supposed to be a Russian Progress flight in July next week I think.

10. space station would seem to be f’d up a tree now… all three active resupplyn systems have failed in rapid succession. Will probably have to abandon or go down to three crew at minimum very soon.

11. Video shows the Dragon appeared to fall away intact after the FTS triggered so this should have been abortable if it was carrying a V2 Dragon.

12. Watching the presser now…

    The Supersize Me guy covers spaceflight now?

13. There’s a problem with relying on a single crewed launch provider, whether it’s SpaceX or Soyuz. Even with a successful escape, the result would likely be a stand-down that required ISS to be mothballed for a period of time.

    If you believe ISS is doing something important, that is unacceptable. If it isn’t doing anything important, why is the US government spending $3-4 billion a year on it?

    1. Who said relying on a single provider was the right thing to do?

       1. Who said relying on a single provider was the right thing to do?

          Everyone who signed off on the Soyuz plan. And a lot of Congressmen who have been calling for another CCDev downselect.

14. I may be way off the deep end here, but to me, it looks as if the failure was in the second stage and near instantaneous (a tank rupture). That’s what I’d expect to see if the stage was unzipped (meaning by the FTS. That’s how FTS works; rupture the tanks).

    So, I’m wondering; could the known O2 overpressure (per SpaceX’s tweet) be a sign of S2 FTS triggering? I’m thinking of a hacker as a possible cause. (as we’ve seen lately, the government isn’t exactly competent in that area.).

    What it definitely wasn’t was a staging event; it couldn’t be, as it occurred pre-MECO. Related, perhaps, such as the MVac precooling, but it happened before MECO.

    I wonder how long the grounding will be? No way anyone can know yet (due to not knowing the cause) but I fear it’ll be a while unless it was something specific to a Dragon launch (thus allowing sat launches to carry on).

    1. Inadvertent FTS activation has happened before; in 1999, the second DARPA / Air Force RQ-4 Global Hawk was lost over China Lake because somebody at Nellis decided to test their 1500 watt flight termination equipment, and used the same frequency and tone sequence as the FTS receivers in the RQ-4, which was aloft at over 40,000 ft altitude at the time. The Western Range Commander’s Council is supposed to coordinate all frequency (including FTS) usage in the US western flight test ranges, but the Nellis team habitually didn’t do the right coordination because they didn’t want to let anybody know what they were doing. It’s all in the official AF report on the accident, which used to be available online (don’t know if it still is). Ended up costing the government in the neighborhood of $100 million when all was said and done.

       That said, it seems pretty unlikely in this event; the only other similar event of which I’m aware is the Lockheed Martin Polecat demonstrator in 2006/2007, and it’s my understanding that LM was partially to blame on that one…

    2. Possible loss of vehicle to malfunction of “safety” equipment? Who around here could have foreseen that a narrow minded obsession with safety might make a rocket less safe? 😉

       1. Range FTS is a necessary evil for some systems, not for all; for example, after the Global Hawk inadvertent termination accident, the Air Force ranges agreed to take out all of the independent FTS equipment, since the GH had multiple redundant command and control links, avionics, actuation, etc. Rockets are often single-string and don’t typically have production command and control links.

          But range FTS is an incredibly antiquated system, dating from the ’50s. There are only three approved frequencies (425, 428, and recently 431 MHz), there are only 16 tones and almost everybody uses the default sequence programmed into the range FTS receivers (that was a contributor to the GH accident – the FTS receivers were set to the default sequence and the Nellis boz^H^H^H team dialed up the same frequency and sent the default tones). There’s a newer system, designed in response to the GH accident, that adds a unique key to the tone sequence, but it took nearly 10 years to get it available and I don’t know if the eastern ranges use it or not.

          Anyway, moot point for this event, as SpaceX has confirmed no FTS signals sent.

          1. Thanks for the info, cuthulu.

             I just hope they find the cause, whatever it is.

    3. I’ve seen deliberate propellant tank failures before, during burst testing. On video, from outside the bunker, obviously. And even replaying the video frame-by-frame, it’s often difficult to capture the breach. It doesn’t take high explosives to make a tank unzip in 1/24th of a second or so; pressure alone will do the job. Think boiler explosion.

       And O2 tank overpressure, which doesn’t necessarily mean O2 overpressure. The tank is pressurized by helium stored in high-pressure bottles, and the helium side is more susceptible to rapid changes in pressure than the LOX. Helium under pressure flows real fast through small holes or cracks, and there’s definitely enough in the bottles to rupture the tank if you let it into the ullage volume faster than you can get the LOX out.

       1. Helium, especially with SpaceX’s past problems with it, would have been my prime suspect were it not for Musk’s “counterintuitive cause” comment.

          If, however, it is helium, my guess is that we’re in for a long grounding because they’ll probably redesign that system (which has given them many headaches in the past). I hope I’m wrong.

       2. I think the helium is primarily used to pressurize the kerosene tank and the O2 tank is self-pressurizing due to LOX boiloff.

15. The definition of insanity is doing the same thing trying to get a different result. Ever since the shuttle people have clung to the hope of cheap safe reliable space travel, with conventional chemical propulsion technology. I think spacex and space in general it at a kind of fork in the road with two options.

    One is going all in to break the tyrrany of the rocket equation via R&D on propulsion tech. Magnetic rail launch, air breathing like skylon, or maybe something exotic like the emdive.

    The more likely option is to frankly go russian. Accept your stuck with expendables and work within the system, with price and efficiency. Like angara or some big dumb booster concept. With drone and dirigable replacing sattilite for communications, and current political trends, space thru 2020 and beyond is probably going to be restricted to millitary no so cheap contracts ala ULA. I expect at least 50 percent chance by end of year, spacex will retool falcon and falcon heavy to be a competitive hopefully less expensive but expendable conventional launch system.

    1. There is no reason whatsoever to think they won’t stay the course.

    2. Seems to me this is a vast over-reaction. SpaceX has had quite a few successful launches in a row. Its main competitors have had even more. Presumably they will find and fix the issue that caused this failure. There will be other issues.
       It would be nice if space were as simple as clicking Refresh and fixing the Javascript if the web page doesn’t look right. Unfortunately, trials cost tens or hundreds of millions of dollars and you only get to do them every month or two, so it’s going to take a while and some of our blood pressures suffer. But that’s a silly reason to give up.

    3. Many people tried making a heavier than air aircraft in the late XIXth century. They all failed. Some actually had decent aerodynamics. The problem was that steam engines just did not have enough power to weight ratio to do it. It took lightweight gasoline powered engines to do it.

       The same thing might happen with rockets. But there are not a lot of viable options to increase the power to weight ratio and increase the ISP. There’s nuclear, beamed power propulsion, and little else.

       It might be that someone will make it work though.

    4. But then again, the recent rash of Progress failures show that no expendable system will ever be reliable at a reasonable cost over the long term. In that case, the rational thing to do is the develop a reusable system as fast as you can so you can get vehicles back, tear them down, and find out what almost broke – even if that meant adding risk to the primary mission to do so. There’s a non-negligible chance that if they had gotten the CRS-6 rocket back, they would have found evidence of wear and tear that could have prevented this.

16. There is a slow motion video over at Space Flight now that seems to show the Dragon capsule tumbling away at the 30 second mark. Curious if that is in fact the capsule and if so, was the chute able to deploy so that it could potentially be recovered.

    1. What I am hearing is the Falcon has an auto-destruct based on a wire running the length of the booster. If the wire (a pair or more most likely) breaks before 1st stage sep, it initiates the FTS.

17. Don’t boot me off the site for raising the issue, as all probablity is that the launch failure has conventional explanations. But the idea of sabatage has been mentioned by others, and as I watched the launch live on SpaceX, I noticed some other weird/anomalous things I have not seen in all the other launches.

    The events I saw are in this video of the launch:
    https://www.youtube.com/watch?t=1370&v=ZeiBFtkrZEw

    At 22:39 in the coverage I can what I think are the cold gas thrusters firing. I have not seen them operate on ascent before. Could be lighting angle and moisture content in the upper atmosphere, but nonetheless, I have never seen this in previous launches. Immediately after that event at 22:38-22:40 some very weird EM signal interference with the broadcast can be heard. Given the delay in broadcast vs. mission clock time ( 1-2 seconds) these events are nearly simultaneous. Additional EM interference can then be heard from 22:41 until about 22:51. The vehicle operates nominally until its failure during MVac chilldown. So…long story short, what is causing that EM interference. Unless it is innate to the broadcast equipment, it would have to be a pretty powerful signal overlapping into the broadcast band, or at some lower frequency resonance of the band. Could subsystems of the Falcon vehicle be hacked? The cold gas thruster was a test of the signal, the second signal from 22:41 to 22:51 was for the subsystems that time and execute the staging event, which is where the vehicle looks to have failed.

    1. I’m waiting for SpaceX to deem it sabotage.. but they won’t without ironclad proof.

18. Any experienced rocketeers have a good explanation of the sequence of events during failure from 23:44 where it appears the LOX ( or pressureized He) tank ruptures? At 23:47 it looks like S2 ignites while still attached. At 23:49 I can still see the S1 engines operating. Everything unzips at 23:53 and I also see Dragon separate first at 23:49 before everything RUD’s at 23:53. From that I take away that the top of S2 ( which is where the LOX tank is, yes?) loses mechanical integrity before the vehicle comes apart. Pretty tough vehicle. Too bad Dragon could not deploy it’s chutes. I’m guessing the launch sequence has no eventuality to deploy chutes on ascent.

    1. It’s also too bad the first stage couldn’t get away and at least attempt its landing. Though I’m not sure the software was set up to do that in a non-nominal flight.

    2. I wonder if it could have been due to a pressurization event related to firing the second stage. I wonder if it was a tank or if a valve or fitting or a line itself could have failed. I think they were getting near BECO and staging.

19. What kind of incompetent outfit has a failure due to a tank rupture?? It’s not like an engine with moving parts, or software that can change between flights. A tank! That’s something they should have dialed down by now. Jeez. With all the turnover and the best people leaving, Spacex is not what it used to be. It was time they got some humble pie.

    1. Software can change between flight? Does it self program? I wasn’t aware SpaceX had so much AI in their system. Interesting.

       1. Yeah yeah, smartass

          1. Oh, you didn’t want to be taken seriously. I should have known by the moniker.

    2. Senator Shelby, is that you?

    3. I’m not sure the problem isn’t in the unmoving parts of the tank; it’s the parts that keep the tank pressurized. Pressurized gas can be super tricky. I myself blew up a vessel in my gf’s had the other day because I turned up the feed too high by accident. (She’s fine. It was just pretty startling.)

       1. Considering the data feed was inconsistent with the results, it could well be a software bug, possibly from mismatched versions to equipment. That is after all what caused the Ariane V failure. For ESA, the thruster performed perfectly, as well the software, and they were mismatched. A improperly installed sensor, or software misread of the sensor, could cause an over pressurization of the tank.

          But let’s not take this jingles guy too seriously.

20. While this was disappointing, I’m not worried. 18 consecutive successful launches is unprecedented for a new launch vehicle. They had to roll snake-eyes sooner or later, and the Dragon 2 launch abort system would have certainly saved the crew in this case.

    I think SpaceX will figure this out quickly and be back to flight soon. The system itself works fine. There was probably some kind of issue with this particular vehicle.

21. I’m wondering if something as simple as a helium valve sticking open might have caused the overpressure event leading to failure.

    1. That thought also crossed my mind.

22. Since I’m not an engineer, but just a lowly history and international relations major, all I can say is cheer up people, the road to Nirvana was always going to have a few bumps in it.

    1. Just as long as there are no shotguns on the road to Nirvana. Shotguns and Nirvana don’t really mix.

       1. *groan*

          But still made me chuckle. 🙂

       2. No IEDs, either.

          1. I don’t think you got the reference Mike, my post may have been a bit too Curt for you to pick up on.

Comments are closed.