2014 incident may provide clue to cause of SpaceX Falcon 9 failure
A SpaceX Falcon 9 v1.1 encountered an anomaly about two minutes into the flight of the CRS-7 mission - resulting in the complete loss of the booster and Dragon spacecraft. Photo Credit: Jared Haworth / SpaceFlight Insider
One week ago today, Space Exploration Technologies’ (SpaceX ) Falcon 9 v1.1 booster suffered a catastrophic failure, approximately 2 minutes and 19 seconds into flight, resulting in the loss of the rocket, the Dragon capsule sitting atop the rocket, and critical cargo intended for resupply and expansion of the International Space Station (ISS).
Shortly after the failure, SpaceX’s CEO, Elon Musk, tweeted that “[t]here was an overpressure event in the upper stage liquid oxygen tank. Data suggests [a] counterintuitive cause.”
The aftermath of August 21, 2014, Cimarron Pressure vessel failure. Photo Credit: C. Vincent / WAFF
NASASpaceflight.com made the observation that “the note from Mr. Musk provides a potential pointer towards the helium pressurization system’s bottles in the Second Stage. [The] Composite Overwrapped Pressure Vessels (COPVs) […] have been a topic of engineering discussion for SpaceX in the past, with ‘bad trends’ in a number of helium bottles causing a manifest debate relating to the CRS-6/SpX-6 and TurkmenistanSat missions.”
On August 21, 2014, Cimarron Composites, a Huntsville, AL, company “which develops and produces high performance composite tanks and pressure vessels,” and who lists SpaceX as a client on its website, suffered a failure of what appeared to be a pressure vessel similar to those used by SpaceX.
As reported by WAFF 48: “Pieces of metal were scattered several hundred yards from the site of the explosion. The percussive effects of the explosion knocked light fixtures inside the facility off the walls. […] ‘The ground shook… ceiling tiles were damaged. It was a big shock that went through everything.’ […] ‘[T]he building across the street […] had a piece of metal that actually went through it.'”
Damage to building caused by August 21, 2014, Cimarron pressure vessel failure. Photo Credit: C. Vincent / WAFF
In addition, an online search in the days following the August 21, 2014, incident, and now no longer available except in a Google search results summary, revealed that Cimarron and / or its owner / president, Tom Delay “developed a 300 liter type 3 pressure vessel that contains 5,000 psi helium for the Space-X rockets.”
Shortly after the 2014 incident, SpaceFlight Insider contacted Cimarron, requesting additional information, including an inquiry as to whether the test / failure was, in any way, related to SpaceX. To date, no response has been received.
To be clear, it’s not believed that Cimarron produced / manufactured any COPVs flown on the SpaceX CRS-7 booster, as it’s understood that SpaceX has now moved COPV production “in-house.” However, it does appear that Cimarron has relatively recently produced, for SpaceX, COPVs similar to those flown on CRS-7, and a failure, similar to the one that apparently took place last year in Huntsville, could certainly produce damage sufficient to take down a launch vehicle.
The CRS-7 mission launched from Cape Canaveral Air Force Station’s Space Launch Complex 40 in Florida at 10:21 a.m. EDT (14:21 GMT) with more than 4,000 lbs (1,814 kg) of cargo, crew supplies, and experiments bound for the Space Station. This was the first major failure of a Falcon 9 rocket since the booster first took to the skies in 2010.
Scott Johnson
Scott earned both a Bachelor's Degree in public administration, and a law degree, from Samford University in Birmingham, Alabama. He currently practices law in the Birmingham suburb of Homewood. Scott first remembers visiting Marshall Space Flight Center in 1978 to get an up-close look at the first orbiter, Enterprise, which had been transported to Huntsville for dynamic testing. More recently, in 2006, he participated in an effort at the United States Space and Rocket Center (USSRC) to restore the long-neglected Skylab 1-G Trainer. This led to a volunteer position, with the USSRC curator, where he worked for several years maintaining exhibits and archival material, including flown space hardware. Scott attended the STS - 110, 116 and 135 shuttle launches, along with Ares I-X, Atlas V MSL and Delta IV NROL-15 launches. More recently, he covered the Atlas V SBIRS GEO-2 and MAVEN launches, along with the Antares ORB-1, SpaceX CRS-3, and Orion EFT-1 launches.
Just to be clear, it appears the falcon 9 was not party to the failure, merely a nearby victim. The falcon 1 (the second stage) that sits atop the falcon 9 appears to be the failure source. The possibility of a sabotage or a cargo systems failure, while unlikely, have still not been fully investigated or ruled out either. There is no justification in blaming this on the falcon 9 booster.
Hi Dan,
Taking your post point-by-point:
1. No, sorry, judging from your comments things aren’t “clear.”
2. The Falcon 9 was the only likely “party” to this failure. As SpaceX’s CEO and Founder, Elon Musk, himself has Tweeted: “There was an overpressure event in the upper stage liquid oxygen tank. Data suggests counterintuitive cause.” In short, an oxygen tank in the Falcon 9’s second stage became over pressurized and exploded.
2. The Falcon 1, was a completely different rocket, which completed its last flight in 2009. I’m uncertain why you’re stating something so obviously false as if it were fact – but it’s not helpful. If you’re that unaware of the basic facts, you might want to consider doing some research.
3. Your comments about sabotage or a cargo systems failure, suggests you’re willfully ignoring basic facts.
4. There is every justification to blame the Falcon 9 as its second stage appears to be the cause of the failure.
While I appreciate you want something else to be at fault, unless something dramatically changes, that doesn’t appear to be the case.
When it comes to obtaining your information – I strongly advise you avoid blogs and Facebook fan pages as they’re not a source of viable information. I’m also not suggesting you only get your information here. There are a number of other wonderful websites that could help clear things up for you: NASASpaceFlight.com, SpaceFlightNow.com, SpaceFlight101.com, Space.com and others.
Sincerely and with kind regards, Jason Rhian – Editor, SpaceFlight Insider
The falcon 9 is a 9 rocket booster, the second stage can be most anything, but in this case is for all intents and purposes the same as the second stage of the Falcon 1. The falcon 9 booster stage does not seem to have been involved and in fact seems to have worked well after the failure of the single rocket second stage. There are two most likely causes for overpressure in the LOX. Either a vent freeze up, or something else that caused the tank to heat up beyond the capability of the vent to prevent an overpressure situation. The Helium control valve might cause it if it were to open on its own but the helium tank exploding would cause an underpressure LOX situation and only after the second stage engine tried to fire. In any case the falcon 9 aka booster was not involved. The only reason sabotage comes into remote consideration is that the cluster of failures in four separate rocket systems in such a short period of time makes a person wonder. Exactly who would benefit from such an act and how many millions or billions of dollars are at stake. Given a shot at that kind of money, what limit is there on the willful bad acts by humans.
Hi Dan,
1. No, the second stage can’t “be anything.”
2. No, the F9 v1.1 upper stage isn’t essentially the same as the second stage of a Falcon 1.
3. It’s already been said that the oxygen tank on the Falcon 9’s second stage is believed to be the culprit.
4. Yes the Falcon 9 rocket was involved. You can repeat inaccurate statements all you like – it doesn’t make them true.
5. There’s no evidence the issues encountered by the Falcon 9, Soyuz-U and Antares were caused by sabotage.
6. Virgin Galactic isn’t involved in either CRS or CCP – not sure why you’re mentioning them.
7. I’m guessing the competition you mentioned is CRS. Given NASA has already selected the companies for the first phase of CRS and CCtCap – I don’t understand why you’re implying otherwise.
You should review the 4 links I provided you with in my prior message.
Sincerely, Jason Rhian – Editor, SpaceFlight Insider
Dan, the Falcon 9 refers to the whole rocket. The piece on top of the “Falcon 9” you mentioned is actually the second stage of the Falcon 9. The Falcon 1 is a completely separate rocket with a single Merlin first stage and a single kestrel second stage. The Falcon 1 is retired and no longer flies. The Falcon 9 suffered the failure Sunday. The possibility of sabatage is minimal. Dragon appears to be fine, so cargo failure may have occured in the trunk of the vehicle in the IDA. Helium systems have given SpaceX trouble before so either it or the IDA are possible failure vectors. Sabatage isn’t one.
I think sabotage is unlikely as well, but given the billions of dollars to be gained by the winner of this competition it can’t be ruled out until the fault is absolutely known for all four flights. Virgin, progress, Spacex and Orbital. A fault tree investigation is unlikely to be able to prove it if it is sabotage. My personal favorite is a frozen vent tube. We should know more by the end of the week according to the latest tweet from Mr. Musk.
Charlie, It is helpful to have you clear that up about the second stage being part of the Falcon 9 booster. Does the F9 v1.1 reflect the whole booster? When communicating about the second stage, I am not sure if this is a way of identifying the two stages together or just a way of identifying the matched boosters first and second stages together and then being more specific by specifying the second stage in isolation to the First. I have not seen much on the second stage in terms of model numbers or versions. I thought the v1 was the earlier engine configuration in three rows and the F9 v1.1 was the eight engine arrangement equidistant concentric with one center line. IN other words square vs. circular. This is not to say there were not other changes too, just the obvious. Obvious also was the failure upstream of the First stage. A failure in the Trunk could have caused a secondary effect on the Second stage LOX, but I am sure the engineers are looking at that too. It seems clear that the first stage operated for some time after the explosion and is not the cause. If one characterizes the Booster as the two stages together, they could say that the Falcon 9 failed, but engineers like to be more specific and clearer than that. I think we will find an eventual cause as tank failure due to high pressure likely caused by valve failure, over heating, or exceeding tank design limitations.
The entire rocket is referred to as the Falcon 9 v 1.1, so to be specific it was the Falcon 9 second stage that failed. Changes from v1.0 includes the rearranging of the engines, upgrading the engines to produce more thrust, and lengthening the fuel and oxidizer tanks on both stages now that the engines can handle the extra weight.
I think what Dan is trying to say is the first stage of the Falcon 9 was most likely not the culprit. The second stage design will have to be fixed for all future launches.
I suggest you stop using the word “sabotage.” In the United States, you’re innocent until proven guilty. No official announcement had even suggested sabotage. Forget about it.
It could be anything. No armchair rocket scientist is going to have anything credible at their disposal without an inside scoop on the workings of SpaceX, NASA, or any I’d their contractors. Making stuff up and perpetuating rumors only hurts your own credibility. Don’t pretend to have a clue. There are plenty of Reddit threads for speculations and conspiracy theories.
Buzz Aldrin said the the Atlas John Glen sat on was a very dangerous launch vehicle. But over the years it developed into one of the safest.
Soon SpaceX will have the ability to reuse its first stage. Being able to bring it home study it, harden it creates the possibility and likelihood to make rockets safer than ever.
Jason Rhian, I love the fact that you regularly respond to comments, right or wrong. That’s very rare these days and I commend you for it. FYI the Soyuz/Progress M-27M that didn’t make it to orbit on April 28,2015 was the Soyuz-1-2a model. The past successful launch was the older Soyuz U.
Correction Soyuz-2-1a not 1-2a
CBS reported today that the Dragon capsule survived the breakup of the booster and continued to transmit telemetry until it impacted the Atlantic. That’s important to note, considering that SpaceX will soon be in the business of delivering not just cargo but also crew. Had this been a crewed Dragon equipped with parachutes, it seems reasonable to think that the crew would’ve had a decent chance of survival, even in a booster failure as catastrophic as this. While SpaceX clearly has work to do to improve the reliability of the Falcon 9 booster, the fact that Dragon survived demonstrates the fundamental advantages of a capsule configuration over winged craft such as shuttle or Dreamchaser.
In the interest of full disclosure, I am not an expert on rockets, but a follower. That said, I do know a little bit professionally about FMEA (Failure Mode Engineering Analysis):
1. The investigation team will make a full list of possibles and assign likely and unlikely causes of the accident. I saw the comment about sabotage. This could be a cause, but I figure unlikely.
2. All parts recovered involving the failed part will go through the “best practice” metallurgical and engineering analysis for it’s industry and science, to prove out the possible causes and assemble the data pack.
3. The documentation produced will rival Encyclopaedia Britannica. (Probably understating it)
4. Ultimately within SpaceX and NASA and any other relevant interested will assign the primary agreed cause of the accident. Upgrades will be applied and, once approved, SpaceX should be back on schedule with CRS-8. I have no idea when. My GUESS is, December. Based on nothing but seems reasonable.
Progress resupply re-started faster I think, but I think this is more the Russian “just do it” management style. Please refute me if I am off.
The video clearly shows the failure occurred in the upper part of the second stage. What ever caused the failure, it ruptured the upper oxygen tank. Although not certain, I believe the oxygen tank sits on top the kerosene tank and the cloud that formed around the failure appears to be caused by the cooling affect of an oxygen tank rupture. Although a rupture of the dragon capsule might also rupture the oxygen tank, I would view this as unlikely but not ruled out. I would zero in on the oxygen tank as the failure point. I know there is talk of the composite helium tanks, but I don’t view this as likely. Yes they have been a major concern at NASA for shuttle and other programs, but a failure during first stage without a triggering event(drop in atmospheric pressure would affect the oxygen tank much more structurally than the helium tank because of the pressure level in the two tanks) is highly unlikely but not impossible. I don’t have access to the data, but I would look for a ramp up in pressure in the tank. If I see any, I would look to component failures in the pressurization system. The most important thing to look for is an event that occurs close to or at the same time as the failure no matter how obscured. If none is found, I would look at the plot of external pressure verses mission duration. It is possible that their was a structural flaw in the oxygen tank that gave way when enough delta pressure from inside to outside reach a level that caused a rupture. I would also look at rupture models to simulate the failure. There appears to be little or not sideways torque on the lower stage after the failure. It keeps flying. What kind of explosion would not affect the lower stage in anyway?
Any thoughts on whether it might be hydrocarbon (or other flammable) contamination in the helium system? If a few drops of oil sprayed into the liquid oxygen tank, they could ignite and cause an overpressure event. I’ve read that the failure happened after they started the LOX chill-down, which means it might be the first time helium was released into the LOX tank after the flight started. Vibration could shake loose debris or liquids.
If they designed their system properly, all components including helium components would be cleaned based on oxygen compatibility. They should have no hydrocarbons because they should have been restricted the use from cleaning the components or properly dried them out. Since you mentioned LOX chill-down, I would assume this is the chill down of the engine lines from the tanks exit valve down and into the engine chamber. If it is tie to this event, then the ignition may have started in these lines. Even though you don’t have hydrocarbons, a poorly manufactured oxygen valve could ignite. Metal rubbing on metal could ignite with the oxygen. However, I am sure that SpaceX’s designs should have precluded or minimized this from happening. You mentioned cool down but nothing about tank ullage pressure ramp up. I am not sure if they have one, but we did ramp up the oxygen pressure in the external tank prior to launch of the shuttle.
Accidents happen. I agree they _should_ be oxygen-compatible everywhere. But that depends on more than design.
The Falcon is flight-pressure stabilized, so I was thinking they would pressurize the second stage on the ground, not in flight.