back to article SpaceX: Breach in liquid oxygen tank caused Falcon 9 fireball ... probably

A large breach in the liquid oxygen tank of SpaceX’s Falcom 9 rocket likely caused the explosion during a test at the start of this month, investigators have said. A static firing test at the Cape Canaveral Air Force Station in Florida ended abruptly when the rockets burst into flames on the launchpad and disintegrated, along …

  1. JeffyPoooh
    Pint

    "...one tenth of a second..."

    You'd be wanting a low latency system to move the information out of harm's way.

    As opposed to buffering it locally. Where it'll be destroyed.

    1. Richard 12 Silver badge

      Re: "...one tenth of a second..."

      And very fast sample intervals!

      Most of the data stored in an aircraft "black box" is measured at 1, 10 or even 30 second intervals.

      It's only really the voice recorder that goes fast enough to have any data at all from an event like this.

      I wouldn't be surprised if most of their usable data was actually sound recordings.

      1. Anonymous Coward
        Anonymous Coward

        Re: "...one tenth of a second..."

        I wouldn't be surprised if most of their usable data was actually sound recordings.

        This one will make it into the next Roadrunner cartoons. "Meep meep ... BOOM".

        I wonder if it would not be possible to run a few one strand fibers for telemetry. They'll break as soon as the thing takes off (or explodes), but until then you can push a lot of data through it, safely away from things that can go awry. Or maybe they're doing that already - does anyone know?

        That said, for a company that didn't exist until 2002 it's done a frankly amazing job. It's in the nature of their business for things to occasionally go wrong, but they keep that count quite low. Plus, it was a Facebook satellite where exploding is possibly not the worst of two outcomes for that flight :).

        1. danR2

          Re: "...one tenth of a second..."

          "...it's done a frankly amazing job."

          They've thrown so much R&D and money at a booster that can land on its ass, they've neglected systems the approximate equivalent of which haven't so blowed up real good since 1983. Incidentally, it's not just SpaceX that's doing the, erm... 'anomaly' investigation. The accident team includes all kinds of gubmint types. A fact that Musk is making no great pains to explain fully to... taxpayers.

        2. Nolveys
          Flame

          Re: "...one tenth of a second..."

          Plus, it was a Facebook satellite where exploding is possibly not the worst of two outcomes for that flight :).

          I was expecting they would find that the cause of the explosion was methane related.

        3. imanidiot Silver badge

          Re: "...one tenth of a second..."

          Pretty sure most data connections when the rocket is on the pad will already be optical in nature. Plus a whole range of radio links that would be active during flight on top of that.

        4. JeffyPoooh
          Pint

          Re: "...one tenth of a second..."

          AC suggested "...run a few one strand fibers..."

          NOOOOOO!!!!

          Don't suggest using cheap strands of fiber. Especially to SpaceX.

          You'll be heavily downvoted.

      2. Bubba Von Braun

        Re: "...one tenth of a second..."

        Sampling is occurring at 200ms or faster. Telemetry comes off the rocket (ie: no black box) they will know what happened. right now they just don't know what caused the breach.

        Aircraft longer flight times, slower recording, though advances in solid state recorders have helped the future will be off aircraft recording via systems such as ACARS.

      3. ilmari

        Re: "...one tenth of a second..."

        If I remember correctly, the previous incident's cause was determined in part by comparing the time at which different microphones had picked up first signs of trouble.

    2. Bronek Kozicki

      Re: "...one tenth of a second..."

      Actually I think you will want low latency, high frequency sensors and fast connections. "Low latency system" to me implies data processing, but that does not have to be done in-place at all, as long as you have sufficient bandwidth to move the data away, to a secure place. As others noticed, a piece of fibre in the tower would go a long way.

    3. Steelted

      Re: "...one tenth of a second..."

      And the breach was caused by a UFO! I have the video.

    4. JeffyPoooh
      Pint

      Re: "...one tenth of a second..."

      This 'lesson learned' about latency is from an investigation into an air crash, where the investigators were initially perplexed by the Flight Quality Assurance Recorder (a maintenance aide, not a mandated system) memory having data that terminated almost a minute before the crash.

      They initially considered this timing sequence to be an important clue.

      Later they determined it was nothing more than this recording system simply buffering data in RAM before writing blocks to the non-volatile memory. So the final chunk of data disappeared from RAM at impact. Nothing complicated at all.

      That's why such systems, meaning the important and mandated ones, need to be explicitly designed to avoid latency. Lest your data disappear in the explosion.

      It's a concept worth filing away in the back of your mind, in case you're doing such system design or spec writing.

  2. Anonymous Coward
    Anonymous Coward

    What about the monkey?

    did it eject in time?

    1. Destroy All Monsters Silver badge

      Re: What about the monkey?

      Kerry has been spotted sound at safe at the UN, so everything's cool.

      1. Anonymous Coward
        Anonymous Coward

        Re: What about the monkey?

        we call him 'The Mummy' round our way.......swear to god he's wearing someone else's face on top of his own!

        1. Destroy All Monsters Silver badge

          Re: What about the monkey?

          It's another episode of "Obama West - Reanimator"

  3. Anonymous Coward
    Anonymous Coward

    too technical for me

    how much heat and carbon was released into the atmosphere by this unwelcome defect and how does SpaceX plan to 'Offset it'?

    or do they get a pass?

    1. Tom 7

      Re: too technical for me

      I'd guess its less than a Saturn5 which has the same energy as 0.000000138 seconds of the suns heating of the earth. I'd guess an olympic swimming pool of trees would do the job.

      1. Anonymous Coward
        Anonymous Coward

        'an olympic swimming pool of trees'

        is that an official El Reg metric?

        1. Anonymous Coward
          Boffin

          Re: 'an olympic swimming pool of trees'

          Measured in official El Reg units, 4 double-decker bus eqivalents of shrubbery would also do the trick.

      2. alisonken1

        Re: too technical for me

        Actually, probably a lot more. Saturn V used liquid hydrogen for it's flame - and hydrogen combines with oxygen creates water vapor.

        The only thing the Saturn V could be said is about all of the heat from the fire, hardly any CO2 or any other stuff comes out.

        1. ScissorHands

          Re: too technical for me

          Only on the upper stages were hydrolox. The first, massive stage, was querolox (RP-1 + liquid oxygen).

          1. Yet Another Anonymous coward Silver badge

            Re: too technical for me

            The falcon9 holds nearly 120,000kg of Kerosene.

            An AGA burns about 40litres of kerosene a week (according to their website), say 2000kg/year

            The Gruniad has a readership of around 150,000.

            So assuming that 100,000 Guardian readers have an AGA they burn 200,000 ton of Kerosene a year or nearly 1500 Falcon9 launches - I propose this as a new el'reg unit.

            1. This post has been deleted by its author

        2. Andy france Silver badge

          Re: too technical for me

          @alisonken1.

          Not quite. The first stage of the Saturn V was RP1/LOX with a couple of thousand tons of propellant and this would make a lot of C02,

          The comparatively small second and third stages were indeed hydrogen and LOX powered which do as you say just make water vapor. However industrial hyrdrogen is almost always made from steam and burning coal and the byproduct of this reaction this is C02 (electrolysis of water works with no C02 but is too expensive), so the net result is still lots of C02

          1. Black Betty

            Re: too technical for me

            I think you're referring to the old water gas/syngas method of passing steam over a bed of hot coal in an oxygen free environment. C + H2O --> CO + H2. It was the CO which both killed and left behind a pretty corpse when ovens were the go to suicide method for housewives.

            Today hydrogen is generally made from hydrocarbons for a much higher yield. CH4 + H2O --> CO + 3H2 being the simplest.

    2. Anonymous Coward
      Anonymous Coward

      Re: too technical for me

      Interesting question.

      If you do the maths it works out about as much heat and CO2 as five humans exhale in their lifetimes.

      Makes you think.

    3. allthecoolshortnamesweretaken

      Re: too technical for me

      I'd say the explosion released pretty much the same amount of 'heat and carbon' as a launch would have done. Only in seconds, not in minutes, and while staying put. Possibly less, depending on how much of the fuel was burnt in the explosion. Some of it (from the other tanks) may have been sprayed about before it could ignite.

    4. Anonymous Coward
      Anonymous Coward

      Re: too technical for me

      The combined power consumption of all the readers of this story -- their home machines, wifi, routers; all the network hardware between them and the server farm that hosts the site -- probably contributes the same amount of CO2 to the atmosphere as the explosion.

      And continues to do so 24/7/365.

      And the tea break after Strictly, 3 or 4 magnitudes more.

      1. dajames

        Re: too technical for me

        And continues to do so 24/7/365.

        (Off topic, but ...)

        "24/7/365" is that really a thing, now?

        24 hours a day, 7 days a week, 365 weeks a ... what?

        I would have thought that 24/7/52 or 24/365 would have made some sense, but it seems to me that unless a 7-year period has some special significance then 24/7/365 does not.

        1. Tempest8008

          Re: too technical for me

          24 hours a day

          7 days a week (which includes weekends, which are usually days off)

          365 days a year (which includes all the holidays etc, again usually days off)

          Generally that's how that is read.

          But as you're obviously intelligent, you knew that....you're just being nitpicky.

          Nothing wrong with being nitpicky, per se, though sometimes it can come across as just being smarmy.

          Engendering replies like mine. Which is a textbook example of being...

          Crap. Have I just been trolled?

          Dammit...

    5. Anonymous Coward
      Anonymous Coward

      Re: too technical for me

      SpaceX plans to offset the CO2...by giving you another Planet.

      1. Anonymous Coward
        Anonymous Coward

        Re: too technical for me

        A planet blanketed in CO2 which really isn't relevant since it ain't here. Too bad, so sad.

    6. imanidiot Silver badge

      Re: too technical for me

      Not much more than would have been released by the launch anyway.

    7. Anonymous Coward
      Anonymous Coward

      Re: too technical for me

      It was mostly hydrogen and oxgen that went foom, AC; very, very cold hydrogen and oxygen. So the end result will have been the same as usual - lots of water. The heat created by the hydrogen and oxygen mixing would be offset to some extent by the coldness of the fuel. The CO2 burden on the planet, then, was primarily from the processes used to create the liquid hydrogen and liquid oxygen, and the solid bits of the ship, not from the fuel burning/exploding.

      if you're worried about mankinds carbon footprint, you;re far better off looking at how wasteful our society is in general - space launches add only a miniscule amount compared to just about everything else that we do. It;s just that a rocket launch - or a rocket exploding - simply looks so much more spectacular than a few million burgers in buns, but the latter are likely much more damaging to the planet.

      1. cray74

        Re: too technical for me

        It was mostly hydrogen and oxgen that went foom, AC; very, very cold hydrogen and oxygen.

        Hydrogen attached to carbon. Both stages of the Falcon 9 run on RP-1 (kerosene) and liquid oxygen. The AMOS-6 payload used hydrazine and mixed oxides of nitrogen.

        Currently, SpaceX has no plans for developing hydrogen-fueled engines. The next generation engines like the Raptor are methane-oxygen fueled, while the Draco and SuperDraco use hypergolics.

  4. Sorry that handle is already taken. Silver badge

    "A static firing test at the Cape Canaveral Air Force Station in Florida ended abruptly"

    Technically, the test didn't even begin...

  5. Your alien overlord - fear me

    "make it more dense to fit into the tanks" - this is the same principle, I believe, that the US Army employs on it's Abrahms crews !!!!

    1. 's water music

      tankers

      but how do they chill the crews? Mary Jane? Trance music? Netflix &?

  6. Boris the Cockroach Silver badge
    Mushroom

    Oh well

    My guess was the flight destruct package had gone off.... maybe that explains why I'm not a rocket scientist (and my Kerbals suspected that already)

    According to the press, the helium tank inside the O2 tank ruptured releasing its 5000 psi of helium in one go..... and as we've already seen... those O2 tanks dont exactly like that.

    Still at least they've got a root cause, which makes solving the problem a lot easier

    1. danR2

      Re: Oh well

      "At this stage of the investigation, preliminary review of the data and debris suggests that a large breach in the cryogenic helium system of the second stage liquid oxygen tank took place. [Updated 09/24: At this time, the cause of the potential breach remains unknown.] "    --SpaceX site.

      The update implies the team is still fumbling in the dark: 'preliminary', 'suggests', ... 'cause... remains unknown.'

  7. ravenviz Silver badge
    Trollface

    Re: move some missions to the Kennedy Space Center

    Are they going to blow that up as well?

    1. Anonymous Coward
      Anonymous Coward

      Re: move some missions to the Kennedy Space Center

      Elon Musk is likely to be a P.U.T.I.N. operation to disrupt amerika in any case, so yes.

  8. John Smith 19 Gold badge

    A few notes

    There are extensive threads on this here including a thread devoted to the wild and wacky.

    F9 telemetry is by FO Ethernet connections and there are 3 000 channels of it. I've no idea what sampling schedule they use on each channel. In principle some could just be switch opens/closures and a time stamp. It's a common rule of telemetry engineering to choose sample rates that are adequate for the task being sampled but with Gigabit Ethernet and an FO line they may be more generous. The issue is of course lining those samples up exactly with when they were taken.

    Unauthorised activation of the Flight Termination System was ruled out some time ago.

    High pressure gas bottles store a lot of energy (one of their uses is to provide the muscle to open and close fluid valves on rockets). Range Safety assess them in lbs of TNT equivalent, so pretty serious. However AFAIK no tank has ever failed and they are designed to leak (bleeding off the energy) before burst (which is short and very explosive). However F9 COPV's (gas tanks) are Aluminium lined carbon fibre overwrapped, where most are steel lined. This has suggests thermal expansion/contraction mismatch coupled with mfg issues.

    That said a chain is only as strong as its weakest link. It's more likely some of the couplings or pipework ruptured. Presumably shrapnel from this punched into the fuel tank and had enough thermal or kinetic energy to ignite the now mixing LOX/RP1

    BTW SX expect to open their refurbed pad 39 in November so I expect they will begin launches from their ASAP after that from that pad. How long this pad will need to get back to being launch ready is anybodies guess. Given that SX are quite into "continuous improvement" I'd guess they'll also want to make a few tweaks in the rebuild as well.

    1. cray74

      Re: A few notes

      Thank you, sir. That's one of two reasons why I love The Register's comments: the education you can get from them.

    2. danR2

      Re: A few notes

      "Range Safety assess them in lbs of TNT equivalent, so pretty serious."

      I find it strange that from Reddit's, at any rate, initiation of discussion until the recent announcement, no discussion of helium pressurant issues or possibilities was ever broached.

      1. Yet Another Anonymous coward Silver badge

        Re: A few notes

        4litres (= a milk jug) of LHe expands to 24 cubic meters (=an apartment) of room temperature/pressure gas. So blocked He pipework can get very exciting very quickly.

        It's also why we aren't allowed to travel in elevators with the LHe dewers and why that glass internal wall along the side of the NMR lab is a REALLY stupid idea.

        1. Poncey McPonceface
          Flame

          Re: A few notes

          Sir,

          What is a dewer? And what is the NMR lab? Enlightenment appreciated.

          1. John Smith 19 Gold badge

            What is a dewer? And what is the NMR lab?

            I think a dewer is a glass of a particular single malt Whisky. :)

            A Dewar is the usual scientists name for a vacuum insulated flask, named after their inventor.

            What most people would call a Thermos (which was a company that made them).

            I'd guess he means a Nuclear Magnetic Resonance lab that does NMR to study chemical compounds, rather than imaging.

            1. Fruit and Nutcase Silver badge

              Re: What is a dewer? And what is the NMR lab?

              @Poncey McPonceface

              Liquid Helium is used to cool the super conducting magnets used in MRI scanners, and in the event of an uncontrolled vent of gaseous Helium into the room, the risk of shattering a glass wall along the room containing the instrument is what I think Yet Another Anonymous coward meant.

              see "Cryogens" here...

              http://www.psqh.com/sepoct06/mrisuites.html

              see also

              https://en.wikipedia.org/wiki/Nuclear_magnetic_resonance

              https://en.wikipedia.org/wiki/Magnetic_resonance_imaging

              1. This post has been deleted by its author

            2. Kernel

              Re: What is a dewer? And what is the NMR lab?

              "I think a dewer is a glass of a particular single malt Whisky. :)"

              No, that would also be a Dewar - you need to spend more time drinking whisky. :)

        2. anothercynic Silver badge

          Re: A few notes

          So the question is... who designed your NMR Lab? That *is* insane.

        3. Alan Brown Silver badge

          Re: A few notes

          "that glass internal wall"

          Are you sure it isn't transparent aluminum?

          (more seriously, 4-6 laminated 6mm glass will stop most things thrown at it and an internal partition wall would be just as susceptable to being blown out for the event envisaged (the glass has much higher mass)

          Bomb/blast resistant glass walls have been a thing for quite a while. I encountered my first one in the mid 1980s. (http://www.wrightstyle.co.uk/curtain-wall-facades/blast-resistant-curtain-wall-facades/) Some bright spark decided to try and shoot his way through one where I worked and only scratched the top layer.

          1. phuzz Silver badge
            Boffin

            Re: A few notes

            The important question is, how strong is the rest of the room compared to the glass wall? Ideally there'll be a deliberately weakened wall panel in case of sudden over pressure.

            The danger with an MRI/NMR machine is that the liquid helium is used to cool a superconductor. If any part of the superconductor gets above a critical temperature, it will stop superconducting, thus becoming a normal conductor that has to carry a lot of current, which leads it to heat up, fast. This single non-superconducting part is now getting really hot, and will heat up the surrounding area above it's critical temperature, causing those bits to suddenly heat up, and within a couple of seconds your nice cold superconducting magnet has just reached room temperature, and is busy converting it's magnetic field into heat.

            This means that the liquid helium coolant stops being liquid, and starts taking up a lot more room. Also, the liquid nitrogen that is used to keep the helium cold, also starts to become a gas.

            Tl/dr It's called a magnet "quench", and it looks like someone opened the taps on a whole load of pressurised gas (there's some pretty cool videos out there)

        4. taxythingy

          Re: A few notes

          I'm guessing the glass wall isn't as stupid as you think.

          First, the helium will boil off quickly if dumped on the ground, but that's nowhere near as rapid an expansion as an explosion. Most over-pressure will have a decent chance to vent through the air ducting system.

          Second, the helium that evaporates is coming off at -260ish degrees celsius, so it isn't going to get to anywhere close to ATP volume initially. With the venting, that should prevent a pressure blow-out of the glass. (Atmospheric Temperature & Pressure, if you asked. Ideal gas law and some heat capacity calculations should suffice for modelling scenarios)

          Thirdly, the glass wall will likely contain much of that helium in the NMR room for sufficient time, protecting the atmosphere of the adjoining space so that people can evacuate. The system only has to prevent a dilution of >10% for time required to evacuate that space.

          NB: That doesn't mean RUD of your cooling system is a 'good thing'. It's not.

        5. Vic

          Re: A few notes

          4litres (= a milk jug) of LHe expands to 24 cubic meters (=an apartment) of room temperature/pressure gas

          Errr - one of us has forgotten O-level chemistry. 24l at room temp/pressure is one mol - which is 4g of He. That's around 32ml...

          Vic.

          1. Yet Another Anonymous coward Silver badge

            Re: A few notes

            and 1 cubic meter = 1000 litres and 4kg = 1000 x 4cc

            1. Vic

              Re: A few notes

              1 cubic meter = 1000 litres

              Yes.

              4kg = 1000 x 4cc

              No.

              Liquid helium has a rather different density to liquid water...

              Vic.

  9. Anonymous Coward
    Anonymous Coward

    @ Mark Zuckerberg

    HAHAHAHAHAHAHHA....

    1. You aint sin me, roit
      Joke

      Re: @ Mark Zuckerberg

      Shouldn't that be HeHeHeHeHeHeHe?

      1. John Brown (no body) Silver badge

        Re: @ Mark Zuckerberg

        For the full schoolboy humour effect, it's actually....

        Ha ha ha,

        he he he,

        Zuckerburg shat on the lavotreeeee!

  10. Tom Paine

    Tuesday's Mars annoucement

    Anyone bored enough to have read this far down the comments may like to know Elon Musk's scheduled to unveil the architecture for his bat-shit crazy* Mars Colonial Transporter fantasy.

    Excellent summary of the, uh, challenges:

    http://arstechnica.co.uk/science/2016/09/between-a-rocket-and-a-hard-place-elon-musk-to-give-the-speech-of-his-life/

    * credibility of project may vary according to how much Star Trek you watch.

  11. TRT Silver badge

    Fireball...

    I wish I was a spaceman. The fastest guy alive. I'd fly you round the universe.

  12. Anonymous Coward
    Anonymous Coward

    Facebook's Satellite

    "Facebook’s $200m Amos-6 satellite."

    Wait, WHAT?

    Somehow this passed me by - http://www.bbc.co.uk/news/technology-34451081

    "Over the last year Facebook has been exploring ways to use aircraft and satellites to beam internet access down into communities from the sky," Mr Zuckerberg wrote.

    let me fix that:

    "Over the last year Facebook has been exploring ways to use aircraft and satellites to beam advertising down into communities from the sky," Mr Zuckerberg wrote.

    1. Chris 239

      Re: Facebook's Satellite

      Initialy I skimmed your post and thought you'd repeated the same text twice then I read it more carefully and saw I was right.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon