SpaceX's Crew Dragon Spacecraft Exploded Because of 'Corrosive Bullets' and Leaky Valves

Pubblicato il 16 lug 2019
Yesterday SpaceX released a report on their analysis of the '4/20' Anomaly that destroyed their Crew Dragon spacecraft during a ground tests. It was the result of oxidizer leaking into the wrong part of the plumbing and then being propelled through the system as fast as a bullet.
www.spacex.com/news/2019/07/15/update-flight-abort-static-fire-anomaly-investigation
Raptor Engine images by Jack Beyer
www.jackbeyer.com/print

Commenti

  • So what are the benefits of using burst discs instead of check valves?

  • It would seem having a valve that opens rapidly instead of modulating was the main culprit but not even mentioned. Maybe the engineers there should have asked their local plumber about rapidly opening/closing shutoff valves and the pitfalls of check valves!

    • @Scott Manley G'day @Scott Manley Yes that's a good point, but we're not talking about the Janitor finishing his hamburger and then wandering out with a Stillson wrench to turn the valve on. A valve that "gradually" opens means milliseconds in order to control the pressure wave front. From your report, the NASA recommendation was for burst valves to be used, and at least that indicated they were considering the shock wave (indeed used that property to perform a function). It seemed by their design, the implications of the pressure wave weren't even considered by the designers, and the proof was in the pudding. ... In this case the pudding exploded. Not exactly a handy design feature for something intended to save the crew! How fast do you want your crew fried as both the "rescue system" and the rocket are exploding? After any accident there are always people like me who will come out with "Coulda ... Woulda .... Shoulda"; those of us in the aviation industry deal with it constantly. That may grate the current generation where everything is always "awesome", but I feel that's the price to pay when dealing with other people's lives/money. I sometimes get the impression the team there are making it up as they're going along. That's fine when building video game; how many there are putting their hand up to sit on top of the candle when they light it? Would you? Obviously all just my personal opinion.

    • @Pete F OK, so how slowly do you want an emergency escape system to operate when your rocket is exploding?

    • @Scott Manley Scott thanks for replying, but from what you mentioned it was the OPENING of the valve in the pressurising gas that caused the "obstruction" to be propelled along the fuel line, causing the damage. This is a very basic and known problem and a first year apprentice plumber would be aware of that issue (as would any home owner dealing with "water hammer"). Obviously I know nothing more about the system than what you described (very well incidentally), but it appeared there was no good reason to have the pressurising system's valve operate so quickly. Pressure shock waves in hydraulic systems is nothing new, neither is the fact that check valves leak. Fluid Dynamics is a first year engineering subject, at least down here in Oz. I thought this failure was a face-palm moment and I'm literally embarrassed for the engineers involved.

    • Do you have time for a slow opening valve if your rocket is exploding?

  • psi really? Use real units man.

  • There is a mystery on the 600 psi disappeared from the He tank. It has been tremendously compensated by the perfect quality of the graphics.

  • Its Tony Bruno

  • corporations can only afford to deliver profit..

  • I believe they haven't killed anyone on their tests yet. but some have

  • I would love to see a video what would happen if a rocket just ascended vertically. Straight up no trajectory

  • Can the pressure system not be brought up to operational pressure slowly so the N2O4 in the lines doesn’t arrive like a bullet?

    • Scott Manley ah, yes well there’s that point. I’d forgotten this was for an escape system. Can the plumbing not be prefilled with an inert buffer? Wouldn’t that prevent the bullet effect of leaked oxidizer without having the untestable burst disk be a part of the mechanism?

    • How long do you want to wait in an emergency abort?

  • I would think the psi in the Helium tank would be higher. ie: my air tank (scuba) holds 3k psi. (not to compare per use)

  • Nope...it exploded because of lack of foresight. Blame it on a valve like an oil rig? No purge valve, no trap in piping, no sensors behind or after valve to test pressures, detect oxidizers in helium? Plz Even a hobbiest at the drag strip installs a purge valve on their NOS system for safety.

  • What about doubling up the check values? Having 1 Check value, then another right after it? You could even add a sensor of some kind in-between the check values and before it. A sensor type that can emit a scanning signal of some kind that is non-reactive to the materials used. If a leak is occurring in one check value, hopefully the second check value can ensure that it wont get very far and the sensors can detect the problem long before device activation.

  • Thank you Scott Manley!!

  • So, Scott, why are the valves replaced by the burst disks, and not supplemented by them by, for example, placing the burst disks between the LF/O tanks and the check valves? These are both relatively low-cost parts, in the context of a rocket engine, but each protects from different kinds of failures, and importantly, these different kinds of failures occur at different times.

  • Great news! The silence was worrying me.

  • 1080p, full HD resolution in MS Paint...lol, you killed it Scott! The best!

  • im impressed with the super draco bell you managed to make in paint. lol

  • Hmmm... So titanium under the right circumstances can act as catalyst? Actually this accident can open new avenues.

    • Not a catalyst; a fuel.

  • The one way valves go back to default closed position to stop fluids going back down the pipe, they are usually spring loaded to auto shut when the pressure drops to below spring force etc, but of course can be motorised

  • Come on it was 4/20 they were all smoking that maryjane. This is just a coverup. But great breakdown, it really helps when everything is simplified down like this.

  • You Painted that diagram like a pro.

  • The sooner Spacex gets rid of the helium pressurization systems with Starship the better. It has caused so many headaches.

  • As usual, a quality video (even with MS Paint) full of great info, and explained in a way that everyone can understand. Thank you Scott.

  • Why not keep a bit of helium pressure in the lines at all times?

  • Lucid Explanation. Thank You.

  • by adding bust disk spacex will be making the thrusters a one time use right? one time per flight till the disk can be replaced?

    • No but it was originally designed to propulsive landing and what not just a escape system the system can leave orbit and reenter the atmosphere if the second staged malfunctioned in orbit the draco thruster can do alot more just be used for escape system

    • because you're planning to use the escape system multiple times in the same mission?

  • Most importantly, we now know that there is a spy inside spaceX

  • Well, seems like The old school engineering firms such as Boeing , Lockheed and Northrop know something after all.

  • As stressful a job as it would be to try and figure this out, it's gotta be incredibly rewarding to piece it together and find the answer. Never doubted the SuperDracos for a moment.

  • Of course they need this loss, to try to make the space agencies seem real and credible, but are not.

  • Wait a minute? Aren't the superdracos supposed to be in-flight restartable?

  • This has been documented for decades. It’s known as the “Taco Bell Effect.” When high pressure gas accelerates a soft blob of material directly through a one-way check-valve apparatus, causing an unexpected blowout. Every taco tuesday, someone experiences this effect.

  • 02:34 ,,something werent quite right'' you told at the start on the video they know exactly what went wrong... seems to me they actually dont now anything^^ in my opinion this ,,something werent quite right'' should be the main thing to look after or am i wrong??

  • so it is Space X causing the helium shortages in the US.

    • Nah, it’s economics. In the 90’s politicians started selling off the strategic helium reserve, that meant it wasn’t profitable to harvest helium. Now supply can’t match demand.

  • 3:55 I think you meant to say "Minecraft Fire" 😂

  • Ok my assessment of blame on the engines may have been wrong, but I'd like to know if this check valve was 3D printed. If it was I'd feel completely vindicated. Either way I'm still skeptical of 3D printing as a practical way to manufacture strong parts, but I do hope nothing explodes in the next launch abort test.

    • Check valves will not be 3d printed. 3d printing is used for short run items with complex geometry, check valves are used everywhere and have simple geometry compared to a rocket combustion chamber.

  • Wouldn't another method of preventing this problem be to use an automated gradual opening of the pressurant valve? Burst disks can't be a final solution if they want to have multiple restart capability... Furthermore, what if part of the burst disk breaks off and blocks a passage downstream? I know they aren't supposed to, but then check valves aren't supposed to leak either.

    • @Scott Manley Good point. I'm beginning to understand why my spaceX application was turned down xD

    • There’s no such thing as a gradual opening when your booster is already exploding

  • It's a good thing that this happened at this stage. Fluid dynamics under loads and interactions of different types, gets very complicated.

  • I hope NASA won't Need Another Seven Astronauts.

  • So they intend to blow up the space station next , spacex extreme fireworks company .

  • Excellent explanation. I come here for good info, and am never disappointed. Thank you!

  • this theory doesnt add up-

  • Didn't that really clever guy invent a testable check valve, ya know the one Elon Musk named his car company after...

  • "...you don't want your liquids flowing back into your gas section. At least it might cause burping..." If they flow all the way back to the Y-joint you're going to get a hell of a burp. They're hypergolic! Yikes!

  • This is so ironic, as Tesla invented a one way vale that cannot fail. Someone tell Elon!

    • "this cannot fail" is usually a phrase found in the category "famous last words"

  • Very simply explained and diagramed. Nicely done

  • Corrosive bullets. **sweats in Elite Dangerous experimental effects**

  • Would have to agree that Rupture Disks are way more reliable than Check Valves. Not reusable but could be easily replaced between tests. Failure for the win.

  • This happens when I eat Taco Bell.

  • Thanks for explanation. BTW Dragon2 doesn't have to worry about busy parking at station because it uses the old Shuttle PMA2 docking port which nobody uses. (but still has to take appointements to make sure they schedule crew rotations etc).

  • Scott, your quick and simple responses and explanations are amazing and still manage to be great quality. It’s a bliss to have them.

  • This was really interesting! Thanks for your explanation. The 1080p Microsoft paint diagram was brilliant

  • Man you do a great job Manley!

  • What is burst disc and how does it work.

  • I would guess frame shift drives are immune to this problems?

  • Great video man was able to comprehend all of it :D

  • No isolation valves behind the check valves? Odd....

  • When your blood flows backwards through the venal valves, trouble.

  • Nice job explaining this.

  • Nice ms Paint skills, lol. Thumbs up!

  • No one was hurt. This is why we test.

  • No one ever said reinventing the wheel would necessarily be easy.

  • Thats a rough couple months for Mr. Musk

  • Ok, so the Mars Observer Failure Investigation Board concludes that Nitrogen Tetroxide had leaked past the check valve and either reacted with the titanium or caused some other problem. Yet, Spacex writes ”It is worth noting that the reaction between titanium and NTO at high pressure was not expected. Titanium has been used safely over many decades and on many spacecraft from all around the world..." . Someone help me out with this contradiction...

  • Do not worry, Werhern Von Braun had same problems in 40`s. You get it... or contact me, I can send you to Mars and back.

  • Only in aerospace could moving vehicles between parking spaces once every 3 months be called "very busy"!

  • A question to Scott Manley: Did this video get demonetised by any chance?

  • you see what happen to star hopper big fire today

  • You could say that fuel block accelerated *applies sunglasses* Like a rocket!

  • Good to know, thank you.

  • .....until then, I'm Scott Manley......ok, so who will you be after then, then ?

  • All in a day's rocket science.

  • Hmm a bit confused. Why didn't it catch fire during the leak if the chemicals had mixed with metals they shouldn't have (if I followed your explanation correctly), and by removing the check-valve isn't that increasing risk further upstream in the helium lines & tank?

  • really psychadelic Scott but yeah a hotchpotch of meturalogical rifling of heliospheric magnetical sheer however fitting a burst disc inline will be maybe an interesting engineering topic. In the 23rd they might actually bend to the will of utilising a really fine electrochemical safety monitoring signal with alternative secure randomised sweep which would obviously sense any potential pockets then rather than wast a perfectly understandable solution to the expensive bumble of unmentionable pressurised catalytical converts to the sensor array to electrochemical plasma which perhaps engineers might even have service tools already of the coast of the outer planets. In 20-21st century paint it out of your memory alternatives it might be wise to be disgruntled and comment awe they could change the lines on my superwet bike man or something like that as if to be competative. Maybe the spokeswoman was press conference or something and well thanks anyway

  • Amazing can figure out why this happened

  • Actually lucky to catch in testing since this problem would happen rarely. But then using abort only if Falcon 9 rocket fails in flight which hopefully won’t happen in crewed flight?

  • SpaceX went back with footage and found the location where the "potato" quality video was made - and fired the person that did it. She is not allowed to speak about what happened due to legal issues.

  • ...in Microsoft Paint!

  • Nice quick death though.

  • Scott, Since Carbon fibre, titanium and aluminium alloys burn, why not use Basalt Fibre for most of its parts. It's light, corrosive resistant, and can withstand very low to high temperatures. Scott, it seems to be a very interesting material. Please, why don't you look into it further? Thanks