This is not a NASA Website. You might learn something. It's YOUR space agency. Get involved. Take it back. Make it work - for YOU.
Commercialization

SpaceX Had Something Go Wrong With A Crew Dragon Test

By Keith Cowing
NASA Watch
April 21, 2019
Filed under

SpaceX Statement on Crew Dragon Test Stand Anomaly
“Earlier today, SpaceX conducted a series of engine tests on a Crew Dragon test vehicle on our test stand at Landing Zone 1 in Cape Canaveral, Florida. The initial tests completed successfully but the final test resulted in an anomaly on the test stand. Ensuring that our systems meet rigorous safety standards and detecting anomalies like this prior to flight are the main reasons why we test. Our teams are investigating and working closely with our NASA partners.”

NASA Watch founder, Explorers Club Fellow, ex-NASA, Away Teams, Journalist, Space & Astrobiology, Lapsed climber.

123 responses to “SpaceX Had Something Go Wrong With A Crew Dragon Test”

  1. rktsci says:
    0
    0

    This may be footage of what happened:
    https://twitter.com/Astronu

  2. ThomasLMatula says:
    0
    0

    Love the way engineers do understatements like this in these press releases. Yep, having a huge explosion during a test is definitely an anomaly …

    The question is how many months/years does this put back Commercial Crew? If NASA decides that having a hyperbolic abort system built into the capsule is too dangerous it could require both Boeing and SpaceX to redesign their systems.

    Needless to say they are smiling in Russia today as it seems NASA might well need to buy a few more Soyuz flights. And of course the price just went up.

    • Shaw_Bob says:
      0
      0

      Space is hard.

      • fcrary says:
        0
        0

        Space is big
        Space is dark
        It’s hard to find
        A place to park.
        Burma Shave.

        Sorry, but the catch phrases get on my nerves sometimes.

    • fcrary says:
      0
      0

      When it comes to stilted language, my favorites are: “The post-facto probability of failure is unity,” when an investigation into a spacecraft failure could only identify half a dozen possible but unlikely potential causes, and “this aircraft has developed a condition which de-certifies it for trans-Atlantic flight,” when I was on a Dallas to London flight which diverted to Chicago (he was trying to explain the divert without actually saying it was, as I learned later, a fuel leak.)

      But as far as SpaceX is concerned, I can’t see NASA requiring a fundamental change in the abort system. Not something like shifting to solids and moving it up to a tractor on top of the vehicle. The spacecraft will have hypergolic propellent right next to the capsule. That’s necessary for in space maneuvering, and the difference between a Draco for that, and a Super Draco for aborts is just a matter of degree. In either case, an uncontrolled mix of fuel and oxidized would not be survivable, and how unsurvivable isn’t a useful question to ask. But I wouldn’t be shocked if some people within NASA wanted detailed and lengthy analysis to confirm that. Over this and just about everything else involved in this test, there will be an incident investigation, and that could take some time.

      The other issue is the hardware to use for the in-flight abort test and for Demo-2. Since the Dragon 2 planned for use on Demo-2 has to be a full-up flight article, and carefully tested, I suspect they will stick to the plan and not use it for the abort test instead. The abort test can be done with a less than flight-like Dragon 2. It would not, for example, need complete and fully tested life support systems. I don’t know what SpaceX production schedule for Dragon 2’s is like.

      If everyone is very, very lucky, getting a suitable Dragon 2 for the abort test and completing the incident investigation to NASA’s satisfaction could only take three or four months, and Demo-2 could still happen by the end of the year. But I’d say the chances of that can be described by words typically used to describe a person’t weight: Slim or fat. But I’m optimistic that the delay won’t be a full year.

      • hikingmike says:
        0
        0

        Upvote for that fuel leak quote. Did you confirm that’s what it was?

        • fcrary says:
          0
          0

          I couldn’t really confirm the fuel leak, but it seemed to be from a reliable source. After the pilot told us we were diverting, the next bit of bad news were the fire trucks with flashing lights along the runway when we landed (not a good sign.) When we got off the plane, they said they would probably start boarding us on another plane in about 30 to 45 minutes. And handed out vouchers to cover diner while we were waiting. That was at about 4 PM. So I had some doubts about the 30 to 45 minute part… I noticed we had the same crew (or at least flight attendants; the pilots may have hit time limits and been replaced. I don’t know about that.) So, while we were over the Atlantic, I very quietly and politely asked a flight attendant about the problem with the original aircraft. She was the one who told me it was a fuel leak. That’s all I know, but I can’t see why she would have made that up.

          By the way, it’s amazing how far being polite and reasonable can improve you flights. I was once on another diverted flight, when the plane was over Maine when it turned around and went back to Chicago (Chicago. Again. I just never have good luck going through ORD.) When everyone else was complaining, I asked if, even though were were landing where we took off, would I get any frequent flyer miles? That got a laugh, a free drink, and (ironically) an offer from the airline’s customer service people to give me a bunch of miles as an apology.

          • hikingmike says:
            0
            0

            If you got it from a flight attendant, that works for me.

            “de-certifies it for trans-Atlantic flight” – my other guess would be that you had the plague on board or something 🙂

            From Maine back to Chicago, geez.

    • Jeff2Space says:
      0
      0

      You don’t want to call it a failure, because we do not yet know what failed. If you start calling it a failure early on, people using sloppy wording can easily start attributing that failure to some cause. So, yes, it’s an anomaly until the engineers pinpoint the actual failure which caused the anomaly.

    • Michael Spencer says:
      0
      0

      Hypothetical Example #1: The torque spec on a nut someplace was out of spec. As the test proceeded, expected vibration led that nut astray. Vibration led to bigger vibration. The test stand shook itself apart. Result: A stupendously huge explosion. Engineer: “Anomalous”.

      Hypothetical Example #2: Cross-bracing provided nearly all resistance to shear; though the braces met spec, someone goofed up the calculations. Result: A stupendously huge explosion.
      Engineer: “Oh, F**k!!”

  3. Rick Smith says:
    0
    0

    Hmm. How’s that SLS looking now?

    • Skinny_Lu says:
      0
      0

      Not any better than yesterday. A failure for SpaceX does not change the fact that SLS is a turd no one wants, but Congress.

    • fcrary says:
      0
      0

      If you can tell me what zero failures out of zero launches means, I might have an answer. Computers generally regard that ratio as “Not a Number” and comparisons between an actual number and NaN give twitchy results.

    • ed2291 says:
      0
      0

      SLS still looks like an expensive failure with no clear purpose that sucked money for decades on a poor design that was also unnecessarily expensive. The only thing SLS does well is make excuses for failure and announce delays.

    • Jack says:
      0
      0

      Behind schedule and way over budget.

    • richard_schumacher says:
      0
      0

      It hasn’t directly killed anybody yet! Woo!

      • Jeff2Space says:
        0
        0

        That’s some damning with faint praise, but it still beats killing astronauts or aerospace workers on the ground.

    • PsiSquared says:
      0
      0

      Well, let’s see. SpaceX already has one successful mission of its Crew Dragon/Falcon Block V to the ISS. How many successful missions has SLS flown? Hell, how many times has SLS flown?

      • motorhead9999 says:
        0
        0

        How many unexpected anomalies has SpaceX had?

        • fcrary says:
          0
          0

          How can anyone have an expected anomaly?

          • motorhead9999 says:
            0
            0

            They should have been in quotes. My point being though is that as much as space x has done, they do have a tendency to have things unexpectedly happen to them. They’ve blown up a pad (along with a very expensive payload), and their launch record is good, but not perfect, along with other issues that have been caused.

            My point is that while SLS has its issues, SpaceX isn’t perfect either, and in this business, having issues like these appear is a problem, despite many people here who sweep it under the rug and excuse it as “testing”. There’s a difference between something failing a test, and something causing a catastrophic failure.

          • Steve Pemberton says:
            0
            0

            Kind of hard to sweep under the rug a destroyed payload, so I’m not sure who you think is hiding what from whom. Everyone knows about all of the incidents, and that certainly includes the people who purchase launches which includes the USAF. And yet they continue to purchase. That’s because they aren’t just looking at newspaper headlines, they are quite familiar with SpaceX, their capabilities, their processes, and their recovery methods after incidents. They know statistically there is a slightly higher risk than say Atlas, but the cost savings make it worth it when launching a typical satellite, which is covered by insurance and can be replaced, albeit at some inconvenience to their schedules. One of a kind scientific satellites they still seem to lean towards the more proven rockets even with the higher cost, which is understandable.

            Flights with astronauts have a different dynamic in the decision since LAS is available, something not feasible with satellites. Of course the dependability of the LAS is part of the equation, which is why this current situation is a bit of a nail biter.

          • Not Invented Here says:
            0
            0

            SpaceX has a lot of unexpected things happened to them because they are doing a lot of things, statistically the more you do something the more likely it blows up in your face, even if the probability of blowing up is the same.

            SpaceX is doing tons of testing and flights, so naturally bad things tend to happen more frequently to them, unlike SLS which doesn’t test so often (and NASA is trying to delete more testing from it) and hasn’t even flown once.

          • Michael Spencer says:
            0
            0

            What would that difference be?

          • fcrary says:
            0
            0

            Since I’ve already complained about “unexpected anomalies”, I’ll dive in and ask what you consider a “catastrophic” failure. It can’t automatically mean destruction of the item being tested, since destructive testing is a common practice. I’m not even sure I’d call unintended destruction of the vehicle “catastrophic.” I’d reserve that for something which could or would end the whole project. I.e. base the definition on the consequences rather than the size of the smoke cloud.

            By that standard, I don’t think SpaceX has had any catastrophic failures. They blew up a Falcon and Amos-6, but the company continued to fly satellites on Falcon (admittedly, SpaceX did give them a free launch in compensation and to encourage continued business…) They lost a Falcon in flight with a Dragon cargo spacecraft. But NASA didn’t cancel their commercial cargo contract. Nor was did they seriously consider doing so. From the sound of it, NASA is still, after this event, still behind commercial crew and using Dragon 2.

    • Michael Spencer says:
      0
      0

      Is there something about being slapped around, and predictably, that is appealing to you, Mr. Smith? :-). There’s one hell of a gauntlet in these parts.

      After removing the unhelpful barbs, SLS looks about the same.

      SLS is “looking” exactly like what it is: a very capable space system that in many respects is worthy of a great nation. In many other respects, not so much. Here’s why:

      When viewed from the outside (which is apparently where most folks around here are standing, including myself), SLS has some “‘splainin” to do. And I don’t mean explanations like we so often see, explanations largely summarized by “feed the machine.”

      Nope. Comments by those feeling sufficiently knowledgable to equate opinion and explication just aren’t helpful. Similarly, too-loud views like “it’s a jobs program!” are to be dismissed.
      At some point, cooler, and more thoughtful heads will be brought to bear. Those heads, in a sane world, are likely to be separated from their life support system.

      High costs and inexplicable scheduling will at some point need answering. questions will begin like this, from Senator #1: “Can you help us understand the utility of comparing development and production costs of SLS with Falcon 9? What can be learned from that sort of comparison, do you think?”

      Were I responsible for supplying those answers, I wouldn’t even know where to start.

    • Bill Housley says:
      0
      0

      SpaceX will get past this and then still end up flying missions in place of Congress’s overly late, overly expensive, big, white, dodo bird.

  4. Steve Pemberton says:
    0
    0

    The one positive in this could be the timing. I’m sure some will think that it came at a terrible time, happening just months away from the first planned test flight with astronauts. However that is exactly why it is good that it happened now, instead of later with astronauts on board. Also it happened after lots of testing, including several successful tests just prior to the explosion, which increases the chances that the problem is not one of fundamental design, but instead a particular and uncommon combination of variables that occurred during this specific test, or perhaps unexpected wear or fatigue of a particular piece of hardware. Either of which can hopefully be corrected and not require a complete redesign as would be the case with an issue with the fundamental design.

    • fcrary says:
      0
      0

      That all makes sense, unless the event was a reuse issue. In that case, the investigation wouldn’t be relevant to crew launches. NASA wants all new vehicles for that. If (repeat if, it’s far too soon to say) that’s the case, the whole thing is simply an unfortunate delay for commercial crew.

      • BigTedd says:
        0
        0

        OR possible no delay depending on what SpaceX have in the pipe already

        • fcrary says:
          0
          0

          I’ll admit Tesla makes me think Mr. Musk’s companies need some help with production and flow management, but I don’t think having another Dragon 2 ready for flight would make sense. First, even without this anomaly, it wouldn’t be needed until late fall or winter. Why overproduce and then leave it sitting in a warehouse? Second, I’d expect the production rate to be matched to the expected flight rate. Why spend money on a production line which can build them at twice the rate they are needed, and then have to shut down and later restart the line? Third, they presumably learn something from every test and flight, and would like to incorporate minor fixes and changes. That’s harder to do on a fully built vehicle in storage, compared to one that you’ve just started building.

          • BigTedd says:
            0
            0

            Im not sure they need a fully built version just a boiler plate for the abort test most likely and then a modified vehicle for crew launch. Its hard to say as we have no idea what caused the RUD to happen

  5. MarcNBarrett says:
    0
    0

    Scott Manley has concluded that it was a COPV failure. He has a history of being right about things.

    • fcrary says:
      0
      0

      I’m not convinced. It feels like the old habit of fighting the last war. It’s a preflight test failure, and the last SpaceX preflight test failure was a COPV issue. Even experts have an unconscious bias towards attributing the current event to the cause of the last, vaguely similar one. We’ve got virtually zero information and there are tons of things which can go wrong with hypergolic engines. Jumping to conclusions just muddies the waters and drags out the incident investigation.

    • Steve Pemberton says:
      0
      0

      Are you sure whatever you are reading is a direct quote from Scott Manley? He is usually the last person to jump to conclusions on things. I suspect that someone got confused with an older post of his about the Amos-6 explosion which also occurred as part of a static fire test.

    • Bill Housley says:
      0
      0

      These are Scott Manley’s YouTube remarks. He does pour a little bit of cold water on the likelihood that a SuperDraco itself failed, he does not mention the COPVs and declines to speculate on the root cause overall. Maybe he did on Reddit or something.

      https://youtu.be/Fl3Jcczz5PY

      • Steve Pemberton says:
        0
        0

        “Maybe he did on Reddit or something”

        I don’t think so. No source given for the alleged comment, I don’t believe he said it.

  6. fcrary says:
    0
    0

    Does anyone know where that video of the explosion came from? The resolution and frame rate are very poor for any sort of actual, official test monitoring equipment. Was that off an employee’s cell phone? Or, not to start a conspiracy theory, is it even real? At that resolution, one hour of CGI work by an incompetent could produce something similar. I know it’s on the internet, but I don’t automatically believe everything posted to YouTube.

    • Steve Pemberton says:
      0
      0

      The camera shakes just under a second after the explosion so I would estimate it was about a thousand feet away.

    • Not Invented Here says:
      0
      0

      It’s a cellphone recording of a computer monitor showing the event, which may explain its poor quality. Not certain this comes from employee though, could be from the government side who is observing the test.

  7. cb450sc says:
    0
    0

    I guess what amazes me in that video is the sheer speed with which it went sour. I guess I was expecting to see a test burn in progress, then an explosion.

    • BigTedd says:
      0
      0

      I also found it interesting that in that short vid no real flame as such seems more like a pressurized item exploded causing the other

  8. Lawrence Wild says:
    0
    0

    The audio makes it hard to tell but it sounds to me like the actual test fire hadn’t actually been triggered when the explosion occurred. I’m clearly hearing the word “Nine” right before the kaboom. As if they were starting to count down 9,8,7… to the actual firing. This suggests a spontaneous hypergolic reaction in the tanks, possibly from atmospheric intrusion from the earlier tests? I suppose in a few weeks we’ll get some sort of report. I really hope this doesn’t lead to a very long delay. I really want to see SpaceX succeed and American astronauts being launched from American launch sites again.

  9. ed2291 says:
    0
    0

    No, because they had the same problem. We will see.

    • Skinny_Lu says:
      0
      0

      Not the “same” problem. I read Boeing had a leak in the propulsion system. As much as I like SX…. blowing up a flight capsule, not the same. However, SX is still leading the way to space, everyone else is playing for 2nd place.
      If you are not blowing up stuff every once in a while, you are not innovating. Keep going!

  10. ThomasLMatula says:
    0
    0

    Maybe not. Boeing is also using hyperbolic fuel for its abort system and they did have a leak with it.

    • Michael Spencer says:
      0
      0

      I wondered, too, about the wisdom of hypergolic, but the more I learned the more I realized exactly why it is preferred in situations like this. I also learned that the handling of these materials is well understood with many decades’ experience.

      So, I am thinking that IF this explosion was the result of inadvertent mixing, it’s not likely to be a design problem. More likely some sort of human mistake.

      And there you have the Monday morning quarterback analysis from someone hardly qualified to play football. Touch football.

      • BigTedd says:
        0
        0

        It appears more like a COPV overpressure issue!

      • fcrary says:
        0
        0

        I’m not sure if I’d say I actually like hypergolic propellants, especially toxic ones. But for many applications, there aren’t any good alternatives, and they are the best available choice. So I’m all for using them. Of course, I have all sorts of unreasonable prejudices, such as a strong dislike for using liquid sodium as a nuclear reactor coolant.

    • BigTedd says:
      0
      0

      yeah nah , boeing has problem all their own which is requiring some redesign !

  11. tutiger87 says:
    0
    0

    Meh…How much stuff got blown up in the 60s? I love that montage in ‘The Right Stuff’ where there’s explosion after explosion, with the funniest one being the test with no explosion and the chute popping out. The difference now is that its on social media in seconds.

    They’ll fix it and get it done.

    • mfwright says:
      0
      0

      Yes, I remember reading an article about that incident with the Mercury capsule on the Redstone where engine shutdown prior to launch then escape tower flew off followed by the chute popping. “The engineers could explain what happened but the public had their doubts.” Yes back in those days there were lots of explosions. Some articles by Dennis Wingo listed the amount of money spent in 1950s on Atlas and Titan, in the order of $40B in 1960 dollars. Yeow, you don’t have to do the math to get the feel of how much resources (and priority) was put into those launch vehicles. And there were other rockets i.e. Redstone, Thor, etc. probably had their expenses.

      • BigTedd says:
        0
        0

        Kinda of a different methodology , they didn’t know as much so somethings just needed to be flown to prove they worked

  12. Daniel Woodard says:
    0
    0

    NTO does not detonate. The COPVs are at ambient temperature and static pressure at this time. That leaves MMH, which can detonate when subject to a rapid increase in pressure in a confined space. This could have been triggered by a pressure regulator failure overpressurizing a line or tank, or by contamination (from sea water intrusion?) in a line, or a small leak that could have ignited and heated MMh in a fuel line. A leak could result from a crack in a line or the MMH tank, or the failure of a valve. However a valve failure would have probably vented the MMH through the nozzle rather than internally. In either case the precipitating event would have been pressurization of a fuel line.

    • james w barnard says:
      0
      0

      The COPV failure was in the booster (2nd stage???). Does the Crew Dragon even have any COPV’s? The COPV failure occurred when super-cold LHe was used to super-cool the LO2 tank. But the Draco’s and Super Draco’s are NOT cryogenic. Whatever the failure mode, it not only occurs with no apparent firing of the Super Draco engines. It appears to have just been sitting there and WHAM! The plume of “smoke” everybody saw looks like fuming nitric acid fumes…what you get when N2O4 is released into the moisture-laden atmosphere. But that could have resulted when whatever caused the explosion ruptured the N2O4 tank(s).

      • Daniel Woodard says:
        0
        0

        I agree. The red smoke is NTO which breaks down into NO2 and boils off when released at ambient pressure.

      • fcrary says:
        0
        0

        I guess the Dragon 2 would have COPVs. The Super Draco does use helium to pressurize the system, and liquid helium is a hard cryogenic liquid. COPV for any sort of cryogenic tank is relatively common. But that would be a very, very different item from the one involved in the Amos 6 event.

        • james w barnard says:
          0
          0

          Well, at this point, until we get more and reliable information from either SpaceX or NASA or both, it’s all just speculation. Hopefully, it will be something more or less easily fixed, rather than a basic design problem.

          • ThomasLMatula says:
            0
            0

            The big question is that after that video will NASA really be comfortable with having 3,000 lbs of hyperbolic fuel so close to astronauts. Yes, that was the case with the LEM, but that was a different NASA.

          • james w barnard says:
            0
            0

            What other choice is there? You need some hypergols for the reaction control system (RCS). I’m not sure just what the Shuttle OAMS thrusters had, but it had to be at least that. Whatever caused the “anomaly” will be analyzed and corrected. I almost think it is tougher for us speculating than it will be for the engineers to get a handle on and fix the problem.

          • ThomasLMatula says:
            0
            0

            Yes, but the OMS was located in pods on the tail of the Shuttle Orbiter. Of course if it blew the tail would be lost at the very least, leading to the orbiter being lost.

            Still, this may be a Hindenburg moment. It is one thing to know in theory what happens when 3,000 lbs of hypergolic fuel has a system failure, another to actually see it. The percentage of the crew/passengers that lost their life in the Hindenburg were far less than in other accidents involving dirigibles, but those movies of it ensured the emergence of an irrational fear that put an end to them after decades of use. Interestingly enough modern theories are the fire wasn’t from the Hydrogen, but the chemical that coated the fabric on the airframe, an easy fix.

          • james w barnard says:
            0
            0

            Whether the images that have been shown will scare some of the general public has nothing to do with the continuation of the use of hupergolic propellants in the Crew Dragon. That will be the decision to be made after the investigation and analysis of the malfunction. I doubt seriously that SpaceX would shift to a tractor-type “escape tower” at this point.
            Sure that was spectacular, but the explosion of the Titan II at Damascus, Arkansas, makes this looks like a small explosion. Yes, hypergolics can be dangerous, but so can ANY rocket propellants, solids or liquids. We lost a Titan III due to a case bond separation in one of its SRM’s at Vandenberg a while back.
            When I was about 15, I CHOSE aniline and RFNA for a rocket engine I designed and static tested at two different university facilities, because my contemporaries were messing around with zinc dust and sulfur in hardware store aluminum tubing and making a lot of flying shrapnel!
            So, SpaceX will find the problem fix it, and…PRESS ON!

          • fcrary says:
            0
            0

            The Hindenburg-like problem wouldn’t be the public’s reaction to the images. They won’t pay much attention or remember. Aerospace managers and some executives or politicians might. And they might decide that hypergolic propellants mean a week or two of very bad press when something goes wrong. That could create an unfair bias against them. Something similar happened to hydrogen peroxide, since early attempts used had problems with spontaneous detonations. But that was due to contaminants in the peroxide which acted as a catalyst. That’s no longer an issue, and hasn’t been for some time, but it still has a bad reputation.

          • ThomasLMatula says:
            0
            0

            I was not thinking about the general public, but the NASA management that is risk adverse in general and very risk adverse when using commercial options. They might decide they would not like to go before Congress if they ever lose a crew in a similar accident no matter how low the risk.

            Keep in mind that the risks to the Shuttle Orbiter from tile damage was know since the first flight of Columbia in 1981 but the Columbia accident resulted in those risks going from a probability number to reality, and it ended the Shuttle Program. The only reason Shuttle flew as long as it did afterward was because they needed to finish the ISS. If not for that the Shuttle Program would have never returned to flight.

          • chuckc192000 says:
            0
            0

            In the video it looks like the capsule was blown to smithereens. What if there aren’t enough pieces left or instrumentation to say what caused it? What then? Hope for the best?

          • fcrary says:
            0
            0

            A lack of information can always be a problem of the incident investigation. But in this case, I think SpaceX would have reasonably good data. It was a ground test. That means they could make all sorts of measurements which aren’t practical in flight, record all those measurements at a very high rate, and also get high resolution video from multiple cameras at multiple locations. That video we’ve seen is total junk compared to what almost certainly has available.

            Measurement of things like temperatures, pressures, voltages, etc. at many points within a spacecraft are collected. But in a ground test, you can receive all of them, at a rate of many samples per second. In flight, you can only send down a fraction of those measurements, and only one measurement per minute is typical cadence for robotic spacecraft.

            Compared to some spacecraft incident investigations, SpaceX has a wealth of data. Even if the physical evidence is a mess. And SpaceX has filed to make a near-coast ship landing for their next Falcon 9 flight. It would have been a ground landing, but that would have been at or close to the site of this event, and they didn’t want to compromise the physical evidence.

          • space1999 says:
            0
            0

            Assuming this “anomaly” had to do with the propellants, would using the propellant being demoed in the GPIM mission have made any difference? Just curious, I know nothing about propulsion. I suppose the potential energy is there, and if it goes in the right way it’s going to be bad…

          • fcrary says:
            0
            0

            I don’t think so. I believe that mission is testing a monopropellant, and the problem here seems to be a hypergolic bipropellant. The whole idea of “green” propellants is to find ones which are less toxic and easier to handle and store. They’re trying to do that without reducing the energy content and (in the case of bipropellant) still having a hypergolic mix.

          • Jeff2Space says:
            0
            0

            There is lots of hypergolic propellants on ISS or docked/berthed to ISS. Every single visiting spacecraft at ISS has used hypergolics. Two of the Russian modules have hypergolics built into them which are used for the service module’s engines (for control moment gyro desaturation and reboosts when a visiting craft isn’t docked at its aft port).

            Even all of the proposals for US Propulsion Modules would have used hypergolic propellants (lots of propellant actually).

    • richard_schumacher says:
      0
      0

      Is this the first set of Super Dracos and their associated tankage, plumbing, and wiring to have been operated after a dunking in sea water?

      • fcrary says:
        0
        0

        These engines were the first Super Dracos ever flown. The cargo version of the Dragon doesn’t use them. So, unless SpaceX decided to include a salt water dunking as part of their ground tests, I guess that would be a first. Of course, for all we know, SpaceX might have included a dunking in the ground test program.

        • ThomasLMatula says:
          0
          0

          Actually the first flown and never fired before being inserted into salt water. The 2015 abort test resulted in a salt water dunking, but only after the Super Draco engines fired. Also with SpaceX four years means that there have been a lot of revisions since. So it was definitely operating in an unknown part of the envelope as you note.

          • Jeff2Space says:
            0
            0

            True, but I also believe the abort test capsule was recycled into the Dragon Fly test article. It would be interesting to know what refurbishment was done on Dragon Fly and how that differed when compared to the DM-1 capsule’s refurbishment before the failed Super Draco test.

          • ThomasLMatula says:
            0
            0

            Yes, and did they also drain and refuel the abort system? I would assume so, but if not that creates a new set of possibilies. I hope the report on this is made public.

      • Not Invented Here says:
        0
        0

        Note the SuperDracos are not supposed to be dunking in sea water, they’re fairly high up in the capsule and the nozzles have protective covers on them to prevent water intrusion. They specifically designed Dragon 2 to avoid as much sea water intrusion as possible, this is why they were able to recycle the vehicle so quickly after landing.

        • richard_schumacher says:
          0
          0

          This capsule visited ISS and landed in sea water, so it was indeed dunked. Salt spray gets into everything. If there is any aluminum in the plumbing (for example, valve bodies), then Oopsie.

    • savuporo says:
      0
      0

      NTO might not detonate but with the right organic contaminants its a bad day quickly

      • Daniel Woodard says:
        0
        0

        Sorry I was wrong. Apparently nitrogen tetroxide mixed with seawater was the source of the initial explosion in the loss of the K-219. I remain unclear on the exact reaction. If the Dragon explosion was related to refilling the NTO tanks then it is specific to reuse, but the tanks on the Shuttle were purged and refilled. so there must be a way to do it safely,

        • Spaceronin says:
          0
          0

          Point of order those missiles used IRFNA. Not MON. Even nastier stuff.

        • SpaceRonin says:
          0
          0

          Also FWIW, IIRC the OMS tank was a big unit that could be disassembled and inspected. Can’t find my old published shuttle handbook ATM. Not sure a light weight spacecraft style tank is in any way comparable to it. Of course I am just speculating here and have no idea what Spacex have in there.

        • SpaceRonin says:
          0
          0

          Just thinking about it MON and water give you heat and nitric/nitrous acid. Confine that in any way and you get a rupture. I doubt there was any water in the tanks or it would have gone during loading. If there was water in the feedlines and there was a pseudo priming sequence during the countdown. Then it could have gone off that way. They would need to isolate the feed system from the tanks to fill them. I have seen propulsion systems with measured dew points of -60C pouring out water when cut open. Only sure fire way of avoiding this is never get water into the MON system at all. Preferred simulant for vibration testing is HFE. Even an IPA azeotrope drying approach with water is unlikely to work.

    • SpaceRonin says:
      0
      0

      NTO/MMH reaction is quite quick (~5ms) so they don’t tend to ‘detonate’ You get combustion rates limited by the ability to mix the two hypergolic components. MON leads in engines are there to prevent over pressures as the first step in any reaction is stripping the hydrogen. Too much hydrogen present leads to massive pressure rises, engine life limiting in some cases. This is not a detonation as such. FORP, partially combusted products, typically salts, are explosive as they are unstable and condensed. Any internal isolation failure in a feed system tends to lead to propellant vapor mixing which can lead to an over pressure failure. however I would have thought that such an event would have been detectable and the test aborted. Vapor combustion is kinda slow. Another possible failure is in the pressure regulator. Not sure what they are using here mechanical or electrical. The mechanical things are hard seats. If there was a common mode source of contaminants, say some weld splatter, this could hold the two seats open sufficiently to overburden the prop tank. Electronic regulator fail open scenario is a bit harder to envisage, perhaps a short on the prop card but typically they are design to specifically avoid such scenarios. I would have thought that such data indications would have been automatic test abort in any case. They are not all that fast. I wonder if there was a stress corrosion issue in play on the MON tanks? Reusing MON tanks is a very tricky business. The first order of business is an off load and decontamination that minimizes the SCC. Not a trivial undertaking. Then you would have to ensure the tank status minimizes any likelyhood of trapping MON in micro cracks during handling (high storage pressure). It is such a risk that ESA didn’t defuel Rosetta when the launch was aborted and opted instead to leave it fully charged in CSG for a year before the next launch window.

  13. Bad Horse says:
    0
    0

    bet it was a crack in a pressure line or tank (on one propulsion string only) resulting from landing. If so, it’s just a new check and test to add before you ignite the engines. These things happen. Go look at the data on a S-IVB that exploded during a test. This is just part of spaceflight. Bet they still launch on time.

  14. SJG_2010 says:
    0
    0

    Most likely cause? We don’t typically load, pressurize, FIRE, and then recycle bi-propellant systems. Flight systems get loaded and possibly recycled (drained and filled again), but rarely fired in between those steps. I would guess that the process of recycling the system allowed one of the two (fuel or oxidizer) to migrate to somewhere that is was not supposed to. Then on the next cycle the two (fuel/oxidizer) get mixed and “BOOM”. Contamination is another issue. WHen fueling Mars Global Surveyor, the hydrazine was contaminated by a spider crawling up inside an unprotected port o the fuel loading cart and having a buffet of various insects dragged inside. The hydrazine came out the color of iced-tea when we sampled it right before loading. The Prop-Engineer had quite the look (white face) when I mentioned “Good thing that didn’t happen while loading the NTO” (it would have exploded when the oxidizer reacted with the organics in the insect carcasses)

    • fcrary says:
      0
      0

      I’ll agree that full fuel, fire, refuel cycles aren’t common in flight articles. But the are common in ground tests. The R-4D is rated for up to 20,000 firings (!), or so the Wikipedia article claims. I know for a fact that Cassini fired its R-4D over 200 times, and that wasn’t close to the rated limit. It didn’t get that way without ground tests involving multiple refuelings. The Super Draco is also rated for multiple restarts, and the testing would have required refueling.

      So I don’t’ see the cycling in and of itself as a likely problem. What makes this one unique is cycling after reentry (I think that’s unique to the Shuttle’s OMS and RCS thrusters, and the smaller Draco’s on the Dragon (1 or cargo version.) I think it’s also the first cycling of a Super Draco which involved dunking in salt water. But, as you note, there are all sorts of things that could go wrong.

      • SJG_2010 says:
        0
        0

        I disagree that Flights systems are tested as normal practice. I have helped build almost a dozen spacecraft, and not a single one of them fired the prop system after completion. They DID test the thrusters individually, but did not ever hot-fire them at the system level (except with helium as a working fluid). I even worked in the cleanroom where the Cassini Prop system was built and proof pressure tested. We then shipped that straight to JPL where they do not have the capability to hot fire the system. So I question your assertion that the Cassini prop system was ever hot fired after completion.

        • fcrary says:
          0
          0

          I didn’t say the flight articles are put through many, or even any, firings before launch. But flight-like test articles are. That’s how they qualify the engines for however many restarts in flight. So engines essentially identical the the Super Draco’s on the DM-1 Dragon have been put through many cycles. Not those engines, but ones that are essentially identical. In the process of those tests, they would have done refuelings. So the cycles and the refueling aren’t anything new for a Super Draco. It probably was for _those_ Super Dracos involved in this event. But a more critical difference could be going through launch, reentry and exposure to salt water. Or something else. We really don’t have any information to inform even a guess as to the causes.

          Later edit: When it comes to the Cassini main engine, it was most definitely fired after completion and delivery to JPL. All those firings were after launch, but they were firings, to the tune of over 200 times. I get confused when people involved with hardware development focus on the pre-launch work and apparently don’t consider the flight use of the hardware in the same way.

          • ThomasLMatula says:
            0
            0

            Yes, any one of which could have created a flaw somewhere, one that resulted in a chain reaction leading to the capsule being scattered around the test site.

            But at least it happened on the test site where they will be able to recover the pieces and not on the inflight abort which would have made doing so a lot more difficult.

          • SJG_2010 says:
            0
            0

            ” But flight-like test articles are.” Sure for some designs. But in all of my Spacecraft: Cassini, Mars Global Surveyor, Stardust, Genesis, Mars98 Orb & Lander, Deep Impact, Kepler, and JPSS-1 there was no flight-like pathfinder unit. And the prop systems were loaded exactly once, never recycled, and never hot-fired until in space. Now most of those systems were mono-prop and didn’t have oxidizer migration issues. Trivia – Mars Observer is suspected to have been lost because oxidizer leaked into places where it wasn’t supposed to and when they prepped the system for Mars orbital Insertion, the fuel & oxidizer mixed and the vehicle was never heard from again. More Trivia – Mars Global Surveyor was a Bi-Prop system where the intended to use ALL of the oxidizer during Mars Orbital Insertion and “burn to depletion” where the system automatically became just a mono-prop (Hydrazine only) system for the remained of it’s mission. This is a tricky proposition because the hydrazine MIGHT have been pushed up the oxidizer lines from the combustion chamber at the end of the burn. IF it had found a tiny remnant of NTO “puddling” somewhere in the lines, it could have gone badly.

          • fcrary says:
            0
            0

            I intentionally did not say “flight-identical” test articles. For unique spacecraft designs like Cassini or Mars Global Surveyor, the flight article of almost every system is customized to some extent or another. But, by writing “flight-like” test articles, I meant things which were very similar in all significant ways. Of course, that opens up the unending and unendable debate over where you draw the line, and say the customization for the flight model created significant differences.

            Also, I’m glad you wrote about the “suspected” cause of the Mars Observer loss. They actually identified half a dozen possible causes, all believed to be unlikely, and didn’t have any data to draw further conclusions. They ended up pointing to a leaky valve because they thought it was probably a little bit less unlikely than the other possibilities.

  15. Spaceronin says:
    0
    0

    Repeated atmospheric tests of hypergols? Who’s running the book on this? I want ZOT.

    • Terry Stetler says:
      0
      0

      How about spiders, bugs, etc. crawling into the plumbing? Strong oxidizers + biological material can get…exciting.

      • SpaceRonin says:
        0
        0

        Don’t think it feasible. Engines have blow out covers once out of the clean room. Once fired the residuals are toxic to everything. No bug is getting near them. End of day procedure would require a venting and covering. Presumably by SCAPE suited personnel. So your posited scenario is a double failure: The engine left exposed and a bug getting in.

  16. Bill Housley says:
    0
    0

    I really hate to say this again. I know that it poopoos on the parade. But I still say that the SuperDracos, without their other use for powered landings, have now become an over-engineered reinvention of an already very effective wheel. While they were SpaceX’s foot in the door of the Commercial Crew Program, the complexity that they add to the system have now made them a risk.

    I said that in another thread here last month. This accident might have confirmed it.

    Sorry. I imagined real people inside that thing when I saw that explosion and cringed.

    Oh, speaking of which, where was Ripley? Lounging in a hanger somewhere I hope.

    • Steve Pemberton says:
      0
      0

      I think what is upsetting to everyone is seeing what at least appears with our limited information to be a failure related to an engine test firing (I realize the engine apparently hadn’t yet fired) and having that failure occur while the engine was attached to a spaceship. Engine test failures are nothing new, but if an engine blows up on a test stand it is quickly forgotten, no matter how dramatic the video of the failure might be, and no matter how potentially catastrophic the failure would have been if the engine had been attached to the vehicle.

      I’m not sure if an AJ10 was ever hot fired while attached to an Apollo SM and capsule. Imagine if it was and it blew up a flight article Apollo capsule, we would still be talking about it to this day. They hot fired the integrated SSME’s on the launchpad prior to STS-1 and prior to the initial launch of each new orbiter, but as far as I know all other SSME test firings were done at Stennis. I don’t know if any SSME’s blew up at Stennis, if so it has been long forgotten. But imagine the enduring fallout if an SSME had failed during a hot fire test and destroyed an orbiter.

      So to your point I think maybe SpaceX has created more of a PR complication for themselves rather than a technical one with the SuperDracos by setting up a scenario like occurred on Saturday. If in fact the SuperDracos had anything to do with the explosion which we don’t know yet.

      • Bill Housley says:
        0
        0

        Nice sentiment, but a lot more folks have been killed by hypergolics through the years than by malfunctioning solids. The escape tower is solidly tested technology and without powered landings the Super Dracos seem to be a dangerous step backwards on complexity and risk without a big enough step forward in envelope expansion.

        It’s very hard for me to say that, since I’ve always been a huge fan of the Super Dracos.

        • Steve Pemberton says:
          0
          0

          I think it’s a valid question whether they would have used solids if from the beginning they had never intended to do propulsive landing. The same argument about proven technology is used to prefer parachutes. I’m not against the argument, but at some point when capability is restricted there will be decisions leading to being willing to trade a certain amount of safety, to a degree. We would be safer if airliners only flew in perfect weather, but when done carefully the somewhat increased risk of flying through rain clouds or landing in strong crosswinds is seen as worth the benefit in added capability.

          Your point if I understand correctly is that without propulsive landing the benefits aren’t there when used just as LAS, compared to possible increased risk. That may be true in hindsight, but at the time the goal was quick turnaround, which propulsive landing offered. But that is something that is probably not going to be necessary for Dragon, so in hindsight, if Musk had foreseen that Dragon was probably going to just be a once a year taxi to ISS he may very well have stayed more traditional.

          • Bill Housley says:
            0
            0

            I know. He’s been developing and testing technology for landing on Mars, and Mars has no oceans to splash down in. I get that. But hearing him talk about uncertainty in flight characteristics due to the shape from the built-in Super Dracos and knowing that all that hypergolic fuel is just a few feet away from where the passengers would sit.

            Then watching that same capsule abruptly vanish, with its test stand, in a very loud puff of orange smoke…

            The hypergolics are there to escape a kerosene explosion in the heavily tested and flown booster, but where is the escape from the hypergolics? There exists only the reassurance that the end would come as quick as a gunshot.

            Also, crew having to load and unload the nasty stuff.

        • Leonard McCoy says:
          0
          0

          Can you cite numbers? Seems that Challenger was all about a failed SRB joint.

          Also, how do you know that there will not be a future use for propulsive landings to support cargo or other missions>

          • Bill Housley says:
            0
            0

            With solids I think of Challenger…a quick wiki check shows some less obvious ones: test facility accidents in 1964 and 1990, and factory incidents in 1988 and 2004.

            Didn’t hypergolics kill an entire Russian ground crew? And I think I recall reading about a large Chinese launch support incident as well.

          • fcrary says:
            0
            0

            You may be thinking of the preflight test of a R-16 prototype, known as the Nedelin disaster, after the most senior idiot responsible. That did involve hypergolics and almost the same sorts as SpaceX uses. But that wasn’t the initial problem. Once the second stage fired due to an electrical glitch, while on top of a fueled first stage, it was pretty much guaranteed that something really bad would happen. The presence of hypergolics just made a major disaster even more disastrous.

          • Daniel Woodard says:
            0
            0

            And a Brazilian disaster involving a solid
            https://en.wikipedia.org/wi
            As well as an incident at Cape Canaveral when a satellite kick stage was ignited by static electricity.

          • Bill Housley says:
            0
            0

            As for future landings usage plans for the SuperDracos, I only know about the cancelled ones. The future of that seems to be owned by Starship.

          • ThomasLMatula says:
            0
            0

            Based on what is out there on the Starship specs it looks like it will not use hypergolic fuel for its RCS. Instead it appears it will be using either cold-gas pressurized methane or possibly oxygen. The reason was those could be refilled on Mars, but I expect this accident will reinforce those design decisions.

        • james w barnard says:
          0
          0

          What folks are you referring to? if you are including the ones in the two Damascus, AR, Titan II incidents, that is comparing grapes to water melons. If you are referring to Soviet submarine K-219, that is a different scenario, and if reports are true, there was some human error involved in not pumping out the launch tube when the thing leaked sea water. No offense.

          • Daniel Woodard says:
            0
            0

            I believe the Titan silo fire was caused by hydraulic fluid ignited by a welding torch. The Soviet submarine disaster is a puzzle. What exactly was the reaction between seawater and the oxidizer, apparently nitrogen tetroxide?

          • james w barnard says:
            0
            0

            I can’t find my copy of the book, “Titan II”, by David Stumpf, but I believe you are correct about the first explosion, which killed quite a number of workers in the silo. The second incident occurred when one of the maintenance troops dropped a socket wrench down the launch tube, and punctured the Aerozene 50 (fuel) tank on the bird. An attempt to flood the launcher resulted in the explosion, which killed one or two (can’t find my book right now), and blew the warhead about 100 yards, as well as the launcher closure door!
            The Soviet submarine K-219 had had problems before, resulting in welding shut one launch tube door! The second incident was caused when sea water leaked around a seal in the door of that tube. The crewman failed to pump it out, and IIRC (can’t find that book right now, either…have a lot of books), there was a leak in the N2O4 tank of the missile, and also in the MMH tank. At any rate, the explosion ruptured the launch tube, and NO2 combined with the sea water to produce concentrated HNO3 (nitric acid) which ate through the rubber seals all over the sub! They lost control of the reactor, and a very brave, heroic seaman went in and “scrammed” the reactor by hand, dying shortly thereafter. Eventually, the sub sank.
            There is no question that hypergolics can be dangerous…IF mishandled. But, if you recall, the entire Gemini program spacecrafts sat on top of Titan II boosters, and never had a problem.

          • SJG_2010 says:
            0
            0

            james, Re. the titan that exploded due to dropping the wrench: My boss at Lockheed was the first member of the “broken-arrow” team to enter the silo after it blew. He tells the story about the female medical officer in charge of the operation. He was told before entry that he was to remain in the silo for NO MORE THAN FIVE minutes regardless of radiation levels measured (if any). After 3 minutes in (while in full SCAPE suit) he measured zero radiation, but the medical officer insisted that he exit per the plan. He argued at first because the radiation meter said it was OK. The medical officer reminded him that IF he failed to exit the silo in the agreed to 5 minutes, a biopsy from his left testicle was mandatory to check for radiation. He said he has NEVER moved so quickly in his life…. He also said that the explosion was due to the fact that the missile had already sustained a hydrazine leak, and the MMH was puddling near a floor drain when the techs went in to repair the leak. When the tech dropped the wrench, it fell down the hole, bounced off an object and punctured the NTO tank. Once that leak hit the MMH, it was over. He said it blew the lid off the silo, and they eventually found the warhead about a half mile away intact.

        • Daniel Woodard says:
          0
          0

          If you have a reference on people killed by hypergolic propellants I would be interested in seeing it, I do not have that data.

          The escape tower added so much weight to the Orion that it was impossible to launch on the Ares I. It required a complete blast shield around the spacecraft with a second hatch. It was expended on every launch. The Super Dracos have an unanticipated failure mode. It may be as simple as contamination or a leak in an MMH line as a result of the ocean recovery. Obviously landing the spacecraft on land with powered lift would have avoided that particular failure mode, but NASA nixed the idea.

  17. Ben Russell-Gough says:
    0
    0

    Woah! Can you say ‘tank overpress?’

    • BigTedd says:
      0
      0

      I thought same thing , wondered if they had been doing a test to see how much fuel they could squeeze into that COPV

  18. Robert Jones says:
    0
    0

    Makes more sense for these to be solid rocket motors unless you’re going to use them every flight (i.e., during landing).

  19. ThomasLMatula says:
    0
    0

    Looks like NASA is using the anomaly with the Dragon2 to justify doing the “Green Run” on the SLS ending any hope of it flying next year. Guess they believe in not letting the anomaly go to waste…

    https://twitter.com/NASASpa

  20. Leonard McCoy says:
    0
    0

    a relevant paper
    https://ntrs.nasa.gov/archi

  21. Robert Jones says:
    0
    0

    What would happen to Spacex and Dragon 2 if Tesla and Musk went bankrupt in the next couple years? Could Spacex be nationalized given that the US has a GOP government?

    • Bulldog says:
      0
      0

      I think nationalization would be highly unlikely. As with any bankruptcy, the assets of the company would likely be liquidated through a competitive bidding process administered by a bankruptcy court. The resulting funds would be used to pay secured creditors. That being said, I do think it would be very likely that Boeing, Northrop Grumman Innovation Systems or one of the other legacy space companies would be the successful bidder for the assets. They would also likely cherry pick many of the former SpaceX key personnel. Undoubtedly Falcon 9 would continue to fly albeit under new ownership.

      • ThomasLMatula says:
        0
        0

        And fly at a higher price. Plus they would basically freeze the designs and innovation to maximize profits.

        As for Tesla, they would probably be bought up by an old auto firm. Merceded Benz would be the logical one given that have already worked together. Solar City would be dumped and innovation would also be slowed.

    • Not Invented Here says:
      0
      0

      Nothing will happen to SpaceX, it’s a completely separate company from Tesla, the only thing it shares with Tesla is the boss and some R&D resources.

      Also, just because a company went bankrupt doesn’t mean it’s gone, both GM and Chrysler went through bankrupt 10 years ago, they’re still here.

    • MarcNBarrett says:
      0
      0

      This is like asking “What would have happened to Pixar if Apple had gone bankrupt?” (Because Steve Jobs was president of both) The answer being “nothing”.

    • BigTedd says:
      0
      0

      Private company in fact i think Musk has sworn off ever doing a share offering for anything he owns again. Tesla is unlikely to go bust its at 300 Dollars a share that’s a lot of wiggle room and they are about to ramp through into profit !