Elon Musk appeared on CNBC and offered a definitive explanation for his company's recent launch explosion:
SpaceX CEO Elon Musk says that his company has finally gotten to the bottom of the September 1st Falcon 9 explosion — claiming it was the "toughest puzzle" they've ever had to solve. And now that the problem is known, he expects SpaceX to return to flight in mid-December.
Speaking on CNBC yesterday, Musk said "it basically involves liquid helium, advanced carbon fiber composites, and solid oxygen. Oxygen so cold that it actually enters solid phase." So what does that mean exactly? Musk gave some hints a little while ago during a speech he gave to the National Reconnaissance Office. According to a transcript received by Space News, he argued that the supercooled liquid oxygen that SpaceX uses as propellant actually became so cold that it turned into a solid. And that's not supposed to happen.
This solid oxygen may have had a bad reaction with another piece of hardware — one of the vehicle's liquid helium pressure vessels. Three of these vessels sit inside the upper oxygen tank that holds the supercooled liquid oxygen propellant. They're responsible for filling and pressurizing the empty space that's left when the propellant leaves the tank. The vessels are also over wrapped with a carbon fiber composite material. The solid oxygen that formed could have ignited with the carbon, causing the explosion that destroyed the rocket.
Musk called the issue one that had "never been encountered before in the history of rocketry." One of SpaceX's customers, Inmarsat, may find an alternative for one of its upcoming satellite launches. SpaceX launches could resume mid-December.
For comparison's sake, at standard pressure:
- Liquid Helium: 4 K (about −269 °C or −452.2 °F).
- Solid Oxygen: 54.36 K (−218.79 °C, −361.82 °F).
- Solid Nitrogen: 63.15 K (−210.00 °C, −346.00 °F)
- Liquid Nitrogen: 77.355 K (−195.795 °C, −320.431 °F)
- Liquid Oxygen: 90.19 K (−182.96 °C; −297.33 °F)
- Dry Ice (Solid CO2): 194.65 K (−78.5 °C; −109.3 °F).
Related Stories
An editorial by Jason Rhian discusses NASA's handling of the Orb-3 (Orbital Sciences) and CRS-7 (SpaceX) accidents. Both were Commercial Resupply Service missions to the International Space Station. SpaceX intends to fly NASA astronauts using Falcon rockets within the next couple of years:
A recent post appearing on the blog Parabolic Arc noted NASA will not be releasing a public report on the findings of the SpaceX Falcon 9 CRS-7 explosion that resulted in the loss of the launch vehicle, the Dragon spacecraft, and the roughly $118 million in supplies and hardware the spacecraft was carrying. The post also notes that the Orb-3 accident was handled differently by NASA, but were the two accidents so distinct as to warrant two totally dissimilar approaches?
The premise of the Parabolic Arc report was somewhat inaccurate. NASA didn't refuse to issue a public report; the truth is, no public report was ever produced. NASA officials noted on Wednesday, July 19, that, as the agency was not required to create such a report, one was not generated.
When asked about the discrepancy between the two incidents, NASA officials noted that the Orb-3 failure had occurred on a NASA launch pad (at the agency's Wallops Flight Facility Mid-Atlantic Regional Spaceport's Pad-0A – which is managed by Virginia Space, not NASA). Whereas the Falcon 9 CRS-7 mission had launched from SpaceX's own pad (SLC-40, which is not their pad it was leased to them by the U.S. Air Force) on a commercial flight licensed by the Federal Aviation Administration (FAA). Therefore, NASA was not required to produce a report on the CRS-7 accident. However, Orb-3 was also licensed by the FAA, making this distinction tenuous.
The problem submitted by SpaceX as the root cause of the CRS-7 accident was a failed strut in the rocket's second stage. SpaceX stated that it had fixed the problem and, for all intents and purposes, the matter was dropped.
Fast forward 14 months and another Falcon 9, with the $185 million Amos-6 spacecraft, exploded while just sitting on the pad, taking the rocket, its payload, and some of the ground support facilities at Canaveral's Space Launch Complex 40 with it. Since the Amos-6 accident, SpaceX has moved its operations to Kennedy Space Center's historic Launch Complex 39A, under the 20-year lease with NASA that SpaceX entered into in April of 2014.
With limited information made available to the public, conspiracy theories, including those involving it being struck by a drone and snipers hired by SpaceX's competition, sprung up in articles and on comment boards on sites such as NASASpaceFlight.com and elsewhere regarding the cause of the Amos-6 explosion. This demonstrated the need for a transparent accounting of accidents involving public-private efforts such as NASA's Commercial Resupply Services contract.
Extra: Meanwhile, NASA has growing confidence in the test flight schedule for Boeing and SpaceX's crewed flights: http://spacenews.com/nasa-and-companies-express-growing-confidence-in-commercial-crew-schedules/
Related: NASA Advisory Committee Skeptical of SpaceX Manned Refueling Plan
SpaceX Identifies Cause of September Explosion
After Months of Delay Following Explosion, SpaceX Finally Launches More Satellites
Problems With SpaceX Falcon 9 Design Could Delay Manned Missions
Elon Musk Accuses Tesla Employee of Being a Union Agitator
SpaceX Technician says Concerns about Test Results Got Him Fired
Two SoylentNews readers sent us this story:
SpaceX concludes AMOS-6 explosion investigation
SpaceX has just released the concluding update to their investigation into the explosion that abruptly terminated the AMOS-6 while the rocket was still being fueled. It confirms the failure of a composite overwrapped pressure vessel inside the second stage LOX tank, and identifies several credible causes. SpaceX believes it now understands the problem well enough to avoid it going forward, and is hoping to return to flight with the Iridum NEXT launch on Jan 8.
SpaceX to Hopefully Resume Launches This Sunday
SpaceX has concluded its investigation into the September 1st accident and will attempt to return to launching satellites starting on January 8th:
An accident investigation team "concluded that one of the three composite overwrapped pressure vessels inside the second stage liquid oxygen tank failed," SpaceX said Monday in a statement on its website. The September failure was likely because of an oxygen buildup or a void in the buckle in the liner of the vessel, the company said.
At this time however SpaceX has not gotten the FAA's approval to resume operations.
The U.S. Federal Aviation Administration, which regulates commercial space ventures, still is reviewing the mishap.
"The FAA has not yet issued a license to SpaceX for a launch in January," the agency said by e-mail Tuesday.
Also at Ars Technica and USA Today.
Previously: Spacecom Seeks $50 Million or a Free Flight After SpaceX Rocket Explosion
NASA Advisory Committee Skeptical of SpaceX Manned Refueling Plan
SpaceX Identifies Cause of September Explosion
SpaceX Delays Launches to January
SpaceX's controversial rocket fueling procedure appears 'viable,' says NASA safety advisory panel
A NASA safety advisory group weighed in Thursday on SpaceX's highly scrutinized proposal to load rocket propellants while astronauts are aboard, saying it appears to be a "viable option."
Several members of the Aerospace Safety Advisory Panel said that as long as potential hazards can be controlled, loading crew before fueling is finished could be acceptable.
"My sense is that, assuming there are adequate, verifiable controls identified and implemented for the credible hazard causes, and those which could potentially result in an emergency situation ... it appears load-and-go is a viable option for the program to consider," panel member Capt. Brent Jett Jr. (Ret.) said during Thursday's meeting.
SpaceX and Boeing Co. each have NASA contracts to develop separate crew capsules to transport astronauts to the International Space Station. Both SpaceX and Boeing are scheduled to conduct uncrewed flight tests of their vehicles in August, with crewed flight tests set for several months later.
A Falcon 9 blew up during propellant loading in 2016.
Previously: NASA Advisory Committee Skeptical of SpaceX Manned Refueling Plan
Related: SpaceX Identifies Cause of September Explosion
Problems With SpaceX Falcon 9 Design Could Delay Manned Missions
(Score: 0) by Anonymous Coward on Monday November 07 2016, @07:00AM
I keep being amazed how they figure out these things (it does sound fairly believable to my amateur mind) even after all the evidence has gone up in smoke. Literally, in this case.
So what is the solution? I didn't see that mentioned, even though I RTFA. Are they going to launch quicker to prevent the O2 from getting too cold? Are they backing off on the new supercooled design? Are thry tacking on additional insulation blankets? Or something even more drastic?
Enterprising young rocketry buffs want to know! :-D
(Score: 2) by ledow on Monday November 07 2016, @08:37AM
Because when you just make stuff up, you don't need a solution.
"Never happened before in the history of rocketry" for something as "simple" (dangerous, yes, but deliberately simple in principle BECAUSE it's so dangerous) as fuelling with the standard rocket fuel? That's quite worrying.
We know oxygen is a factor. What we don't know is WHY IT IGNITED, and that's much more worrying. Carbon fibre, to my knowledge, doesn't just spontaneously combust even under such extreme circumstances. And why would solid oxygen explode when liquid oxygen wouldn't?
The explanation seems little more than guesswork precisely because of the lack of footage, data or anything else.
Wasn't Musk at one point insinuating that someone could have shot the rocket from an nearby rival's rooftop? This sounds just as likely (i.e. not).
(Score: 3, Interesting) by khallow on Monday November 07 2016, @10:33AM
as fuelling with the standard rocket fuel?
It's not standard what they are doing. They are cooling LOX to near its freezing point so it's much easier to freeze than normal LOX would be. But in turn, they get higher density. The higher the density of propellant, the better thrust per weight they get and the more mass they can put in orbit. The problem appears to be that some LOX leaked out and froze due to the nearby helium tank.
The explanation seems little more than guesswork precisely because of the lack of footage, data or anything else.
They have a bunch of telemetry, data on the state of the rocket from a variety of sensors throughout the rocket. It's not perfect, but it does tell them a fair bit about what happens to the rocket.
We know oxygen is a factor. What we don't know is WHY IT IGNITED, and that's much more worrying.
It ignited because it's frozen oxygen mixed in with a burnable carbon fiber composite. Why is easy to understand.
Wasn't Musk at one point insinuating that someone could have shot the rocket from an nearby rival's rooftop? This sounds just as likely (i.e. not).
I understand they looked through a lot of video from people watching the launch to rule out such sabotage.
(Score: 5, Informative) by VLM on Monday November 07 2016, @02:03PM
What we don't know is WHY IT IGNITED
Some decades ago there was an attempt to use charcoal and liquid O2 as a mining explosive instead of ANFO. It went pretty much nowhere because it was just too unpredictable and unpredictability is a killer with explosives. Yeah yeah all the parts are non-toxic, the smoke is non-toxic, if there's a misfire you wait for the O2 to evaporate and its safe as a bag of charcoal (which, basically, it was).
https://en.wikipedia.org/wiki/Oxyliquit [wikipedia.org]
Pure O2 is quite a PITA to work with. If there's a finger print in an oxygen regulator, sometimes the finger grease spontaneously ignites and blows the regulator apart, which usually does nothing good to the contents of the O2 tank or the people standing nearby.
Something I always found creepy about industrial pressure vessels is if the compressor is far enough away you can hear all kinds of creaking and wriggling as a tank pressurizes and probably they drew a little helium off to actuate a valve or move a servo and the tank wiggled a microscopic fraction of an inch leading to kaboom.
Things like space shuttle tanks and Saturn-V tanks "should have" blown up, but being made of bulk solid aluminum they can't burn quite as well as carbon fiber, which is basically charcoal fiber soaked in plastic, which should burn like hell.
I could see some new general engineering rules coming out of this like no using carbon fiber in contact with liq O2. I'd be surprised if this is the first time someone blew up a carbon fiber tank with liq O2.
(Score: 2) by ledow on Monday November 07 2016, @08:18PM
Good links.
"I'd be surprised if this is the first time someone blew up a carbon fiber tank with liq O2."
In which case, why would you try it?
(Score: 2) by VLM on Monday November 07 2016, @10:30PM
I should have searched NTRS before even posting.
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20010020209.pdf [nasa.gov]
To summarize our tax dollars at work, it takes some effort to blow up liq O2 and carbon fiber, but it most certainly can be done and the paper was full of wiggle words about if theres no source of ignition energy (like a helium tank banging the hell around inside it) then it should probably behave most of the time.
Also they didn't test solid O2 and generally nice dense solid O2 was more of a headache for the explosives.
So... Liq O2 by itself doesn't spontaneously cook off, but solid O2 with a tank in the middle being Fed with is less healthy.
Waaaay back when it blew up there was some talk from spacex about it all being an operational procedure thing, so yeah, just don't make an O2 slushie and let the liq helium sit there for hours beforehand.
I donno if you can man-rate a rocket with slush O2 and a CF helium tank inside it... even if operationally you're really careful about not slushing the tank more or less.
(Score: 0) by Anonymous Coward on Monday November 07 2016, @04:32PM
I keep being amazed how they figure out these things
Experiments. You figure these out by postulating a possible problem and then duplicating the problem in the lab. You know, the same as in almost any other discipline, like computer programming.
Are they going to launch quicker to prevent the O2 from getting too cold?
It probably got cold due to boiling during decompression.
(Score: 3, Funny) by DECbot on Monday November 07 2016, @06:23PM
FUCK BETA!
cats~$ sudo chown -R us /home/base
(Score: 0) by Anonymous Coward on Monday November 07 2016, @03:51PM
It sounds like they don't have a clear evidence path to a cause.
Something about the filling procedure causing a problem with the high pressure Helium tanks inside the really cold O2 tanks seems their leading suspect.
In addition, the article says it is possible that this area was actually the cause of the previous failure attributed to struts in the tank.
To gather more information, it might be nice to have some sort of secondary, sealed layer outside the fiber so that they can monitor for small Helium leaks before they get out of hand?
(Metal, wrap, leak path/space, membrane, more wrap?)
(Score: 2) by butthurt on Monday November 07 2016, @11:58PM
[...] the supercooled liquid oxygen that SpaceX uses as propellant actually became so cold that it turned into a solid. And that's not supposed to happen.
What "supercooled" means is that the temperature of the liquid is below the material's freezing point. Nucleation and freezing can occur around an impurity--or, if the temperature is low enough, even in an extremely pure liquid.
https://en.wikipedia.org/wiki/Supercooled [wikipedia.org]
https://en.wikipedia.org/wiki/Nucleation [wikipedia.org]