NASA Will Not Release Public Report on SpaceX Falcon 9 Dragon Failure

Dragon capsule separated from Falcon 9 launch vehicle.

NASA will not publicly release the results of its own investigation into the catastrophic failure of a SpaceX Falcon 9 rocket that launched a Dragon resupply ship into the Atlantic Ocean in June 2015.

After saying it would release a summary of the agency’s investigation, NASA passed the buck to the FAA on an accident that destroyed $118 million worth of cargo the space agency was sending to the International Space Station (ISS).

“Since it was an FAA licensed flight, NASA is not required to complete a formal final report or public summary, and has deferred any additional products related to the matter at this time,” the agency’s Public Affairs Office (PAO) said in an email.

When Parabolic Arc checked with NASA last fall, the agency said officials were still working on a final report on the accident more than a year after it occurred. In December, PAO said it expected the report and a public summary to be ready in summer 2017.

Last week, the space agency reversed itself when Parabolic Arc checked on the agency’s progress.

Summer was the best information I had in December. I do not have an estimate for when a formal report or public summary will be finalized.

As you know, SpaceX CRS-7 was an FAA licensed flight and the FAA was responsible for overseeing the mishap investigation. NASA had full access to the data, was a part of the SpaceX investigation, and then conducted its own independent review of the incident. The Launch Services Program team briefed its results to NASA leadership in December 2015 (ahead of the Jason-3 mission in January 2016).

Since it was an FAA licensed flight, NASA is not required to complete a formal final report or public summary, and has deferred any additional products related to the matter at this time. The data is important for historical purposes, but the mishap involved a version of the Falcon 9 rocket, the version 1.1, that is now no longer in use.

NASA’s reasoning is curious given how the space agency responded to a similar accident that occurred eight months earlier.

Two Accidents, One Report

A massive explosion occurred right after the Antares rocket hit the ground.

On Oct. 28, 2014, an Orbital Sciences Antares rocket lifted off from Wallops Island in Virgin carrying $51 million worth of NASA supplies aboard a Cygnus cargo ship bound for the space station.

Cygnus never got there. Antares had barely cleared the tower when the booster exploded in the most spectacular American launch failure of the 21st century. The cargo was lost and the launch pad suffered severe damaged.

An aerial view of the Wallops Island launch facilities taken by the Wallops Incident Response Team Oct. 29 following the failed launch attempt of Orbital Science Corp.’s Antares rocket Oct. 28. (Credit: NASA/Terry Zaperach)

The Cygnus launch was conducted under the same NASA Commercial Resupply Services (CRS) program as SpaceX’s Dragon flights. It was also a commercial launch licensed by the FAA.

Here’s another similarity: the explosion involved a version of the Antares rocket that would no longer be used. The explosion was traced to a failure of a turbo pump on an aging AJ26 first stage engine that was originally built for the Soviet Union’s lunar program more than 40 years earlier.

Within two months of the accident, Orbital announced it would replace the AJ26 engines with newly manufactured RD-181 engines. The change of engines required substantial modifications to Antares, which would not fly again for almost two years after the crash.

So, many of the factors NASA has cited for why it won’t release a public report on the Falcon 9 accident were also present for the Antares accident. Yet,  the space agency still released a detailed executive summary of the mishap board’s findings one year after the accident.

[See NASA Independent Review Team Orb–3 Accident Investigation Report Executive Summary, Oct. 9, 2015]

NASA’s only public description of what its investigation found on the Falcon 9 accident is in an audit conducted by agency’s Office of Inspector General (IG).

[ See NASA’s Response to SpaceX’s June 2015 Launch Failure: Impacts on Commercial Resupply of the International Space Station, June 28, 2016]

The report focused primarily on the impact of the loss of Dragon on NASA’s resupply program and the agency’s response to the failure. The audit devoted exactly one page to the cause of the accident. It also raised more questions than it answered.

Two Accidents, Two Different Approaches

The bottom of the Antares explodes right after liftoff.

Despite suffering similar catastrophic failures in the ISS resupply program, NASA approached the investigations differently.

For the Antares failure, a contractor-led mishap board was formed composed of four Orbital employees, two NASA employees and one third-party expert. The investigation was done under the terms of the agency’s CRS contract with Orbital.

In the case of the Falcon 9 failure, two investigations were conducted. A mishap board was formed led by a SpaceX official that included 10 company employees and only a single FAA voting member. NASA officials served as non-voting observers along with representatives from the FAA, National Transportation Safety Board and the U.S. Air Force.

NASA’s Launch Services Program (LSP) conducted its own investigation. This was done because LSP had a contract with SpaceX for the future launch of Jason-3, an ocean-observation satellite that is a collaborative effort of NASA, NOAA and two European space agencies, CNES and Eumetsat.

“Before using a particular launch vehicle for a NASA mission, LSP certifies the vehicle for flight through insight and approval processes,” according to the IG audit. “The LSP investigation confirmed SpaceX’s implementation of corrective actions before approving the January 2016 Jason-3 launch.”

Although NASA had approved Antares to launch Cygnus supply ships to the space station, LSP had not certified the booster to carry other agency payloads into space. As a result, LSP was not tapped to lead an investigation into Orbital’s failure.

Two Investigations, Two Conclusions

Falcon 9 breaks up during a launch failure in June 2015

The SpaceX-led mishap board concluded that an outside supplier was responsible for the accident that brought down the Falcon 9 booster.

The investigation found that the probable cause of the accident was a defective strut attached to one of the helium tanks inside the second stage liquid oxygen (LOX) tank. The strut broke in flight, causing a rupture in the helium system that over pressurized the LOX tank and caused it to burst.

Because the rocket crashed into the ocean, SpaceX was not able to recover any hardware to confirm the suspect strut was indeed defective. SpaceX said it made the determination after studying data from the accident. The company also found material defects in other struts it tested after the accident.

Elon Musk (Credit: SpaceX)

Only the 11 SpaceX employees on the mishap board signed the final accident report that was submitted to the FAA and NASA for review. The lone FAA vote representative did not sign the report for reasons that are unclear.

SpaceX CEO Elon Musk announced the company’s preliminary conclusion about the single defective strut about a month after the crash. However, there were immediate doubts about the conclusion. Several sources familiar with the accident told Parabolic Arc they believed the failure was actually more complex than SpaceX claimed.

The IG’s audit indicates that the LSP’s conclusions fell along those lines.

LSP did not identify a single probable cause for the launch failure, instead listing several “credible causes.” In addition to the material defects in the strut assembly SpaceX found during its testing, LSP pointed to manufacturing damage or improper installation of the assembly into the rocket as possible initiators of the failure. LSP also highlighted improper material selection and such practices as individuals standing on flight hardware during the assembly process, as possible contributing factors.

These three sentences are as detailed as the audit gets about NASA’s conclusions on what might have caused the failure. Still, there are some interesting nuggets in there.

For one, it makes clear that NASA doesn’t entirely buy the “defective strut snapped causing the rocket to explode, the supplier is to blame” conclusion reached by SpaceX.

The investigation acknowledged other struts in that same batch had material defects when SpaceX tested them after the accident. But, there are a number of other credible causes for the failure (manufacturing damage, improper installation and materials selection) that pointed to problems at SpaceX.

The space agency was concerned enough for NASA Administrator Charles Bolden and Associate Administrator for Human Spaceflight William Gerstenmaier to write a letter to SpaceX in February 2016 “expressing concerns about the company’s systems engineering and management practices, hardware installation and repair methods, and telemetry systems based on LSP’s review of the failure.”

The mention of the letter, which is a pretty big deal given what it said and who wrote it, is confined to a footnote in the IG audit. The letter did appear to have borne some fruit.

“SpaceX has taken action to correct the deficiencies that led to the failed strut assembly and to address NASA’s concerns by conducting inspections, replacing suspect parts, and conducting additional testing. The company also reviewed the certifications of all spaceflight hardware and altered its quality control processes to better align with NASA technical standards,” the audit stated.

Unanswered Questions

The IG’s audit leaves a number of questions unanswered that a detailed public summary of the LSP’s investigation would be able to address.

What type of manufacturing damage might have occurred? What improper materials might SpaceX have been selected? And what issues did NASA found with the telemetry system?

Here’s another issue: we know from SpaceX’s own public statements that the year before the accident, the company was having trouble with helium leaks in the Falcon 9. The problems corresponded to a switch from tanks provided by an outside vendor to ones produced in house at SpaceX.

The leaks were merely a major nuisance in 2014, limiting SpaceX to six launches instead of the 11 or 12 it planned to conduct. But, what if the leaks were merely a prelude to the loss of the Falcon 9 in June 2015 for reasons other than a lone defective strut?

The IG report also noted issues with the process used to get Falcon 9 back into the sky. On Dec. 18, 2015, the FAA Office of Commercial Space Transportation (FAA AST) issued a launch license for Falcon 9 to return to flight after reviewing SpaceX’s mishap report.

“Although FAA officials had access to LSP meetings during the SPX-7 investigation, the FAA did not receive LSP’s final report with findings and recommendations for corrective actions before SpaceX obtained FAA approval to return to flight in December 2015,” the IG audit stated.

“While the FAA was not required to review LSP’s findings before issuing a license, this uncoordinated approach increased the risk the FAA approved a launch without fully understanding the LSP investigation’s findings and recommended corrective actions,” the audit added.

The same day FAA approved SpaceX’s launch license, LSP officials briefed NASA’s top management on the results of its investigation and the company’s actions to address the cause.

Three days after the FAA issued its launch license and LSP officials gave their briefing,  the Falcon 9 made a spectacular return to flight by launching a satellite and landing the first stage back at Cape Canaveral.  This was the first flight of the upgraded Falcon 9 v.1.2.

The following month, NASA’s faith in SpaceX was rewarded with the successful launch of Jason-3 on the last Falcon 9 v1.1 booster. Seven more successful flights followed with the upgraded version of the launch vehicle.

Credit: USLaunchReport.com

Then came Sept. 1. A Falcon 9 exploded on the launch pad while being fueled for a pre-flight engine test, destroying Spacecom’s Amos 6 communications satellite. The failure once again originated in the the second stage as it did with the Falcon 9 that occurred 14 months earlier.

SpaceX ruled out any problem with the revamped strut assembly. Instead, it said that super-cold liquid oxygen from the LOX tank had built up in buckles in a helium tank’s outer carbon-composite liner. The LOX turned solid, resulting in friction that caused the helium tank to explode.

The on-pad explosion raises some additional questions. In identifying and correcting the problem with the strut, did SpaceX actually address the underlying root cause of the June 2015 failure? Are there any common links between the helium leaks in 2014, the in-flight failure and the launch pad explosion? Or are they simply separate, unrelated problems with a troubled upper stage?

A Lack of Transparency 

The IG audit faults NASA for the ways in which the agency has responded to the two failures in the Commercial Resupply Services program. The space agency lacks a standardized procedure for mishap investigations, did not find elements of a full “root cause” determination, and allowed both SpaceX and Orbital to conduct their own investigations without screening board members for conflicts of interest, the audit found.

“We acknowledge SpaceX’s investigation was transparent and the observers from FAA, ISS, LSP, NTSB, and USAF had access to the investigation’s data and analysis. However, an investigation led by the employee responsible for the SPX-7 launch and run by the contractor responsible for the failure raises questions about inherent conflicts of interest,” the report stated.

“While NASA, Orbital, and SpaceX, have similar incentives to safely and quickly return to flight, the structure of the contractor-led investigations may not result in a full review of all programmatic and organizational contributing factors and consequently these factors may not be fully addressed to prevent future failures,” the report added. [Emphasis mine]

Another issue is NASA’s inconsistency in the release of information about these two failures. The space agency published a detailed summary of the Antares failure. But, the technical explanation of NASA’s findings about the Falcon 9 failure has been reduced to three sentences in an IG audit.

That doesn’t seem right. The public is paying billions of dollars to these companies to haul supplies up to the space station. When one of them failures, it has a right to know what happened regardless of which company it is that had the bad day.

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

  • But Doug, these are PURELY PRIVATE companies getting hundreds of millions of taxpayer dollars! We shouldn’t expect our PUBLIC tax dollars to entitle us to PRIVATE information!

    Ummmm, FOIA time?

  • Rebel44

    I would love to see as much info as possible to be released about both investigations, but some of the arguments in this article are pretty poor
    “questions about inherent conflicts of interest” – they are company employees, what the hell is wrong with that? Should they have paid some contractors to do it (who would also had a potential conflict of interest since they would also be paid by SpaceX) who wouldnt know Falcon 9 as well as those employees, resulting in significant delay in investigation?

    I also dont like your excessive suspicion of relationship between 2 SpaceX accidents – in post-AMOS-6 tests COPV failure has been replicated using same fueling procedure in destructive tests – this obviously wasnt present in CRS-7 which used Falcon 9 1.1 that didnt use densified propellant.

    Lets be clear, reason for investigation is to find and fix the issue that caused, or could have caused, the failure – not to spend 2 years to create a small mountain of paperwork

  • redneck

    The comparative results of the two investigations might be revealed by looking at the number of Antares launches since the accident relative to the number of Falcon launches. Orbital ATK has a much longer history of working with NASA than SpaceX and the results show.

    It goes back to process oriented vs performance oriented.

  • DP Huntsman

    I also had the same thought; i.e., FOIA time. There appears to be no credible public interest in not releasing the report; quite the opposite, in fact; as well as then appearing consistent in how investigations regarding contract failures are handled.

    Additionally…. SpaceX should be asked, directly, whether a NASA is refusing to release this public interest-related report, at SpaceX’s request.

  • Hari

    Proprietary supercedes the public’s right to know.. SpaceX has a successful product, in the Falcon 9, but it also has a lot of rivals who would dearly like to see them fail.

  • Terry Rawnsley

    Proprietary does not always supercede the public’s right to know. Hence tobacco and asbestos settlements and scores of others. In this case, NASA, as the customer and representative of the taxpayers, already knows and can release the information (or not) subject to FOIA requirements. If you want to keep your business secrets, don’t screw up or hurt people. In this case SpaceX screwed up. They’re just lucky nobody got hurt.

  • Hari

    God-forbid an astronaut perishes during an F9 launch. Tragic as it would be they would have done so with the full knowledge of potentially dying or suffering injury.
    SpaceX has already stated that no human will launch on an F9 until the rocket has flown upto 50 times. The accidents in 2015 and 2016 occurred in the pre-crewed phase of the F9 program with both causes identified and remedied.
    Why should taxpayers need further insight when NASA and other SpaceX customers (including the U.S. Air Force) seem happy with things?

  • Mr Snarky Answer

    “SpaceX ruled out any problem with the revamped strut assembly. Instead, it said that super-cold liquid oxygen from the LOX tank had built up in buckles in a helium tank’s outer carbon-composite liner. The LOX turned solid, resulting in friction that caused the helium tank to explode.”

    No this is totally incorrect, the buckles are in the inner aluminum liner that allow LOX/SOX to pool between liner and outer carbon fiber overwrap. Think about a balloon that has lost some helium over time and how grooves/ridges can form radially on the balloon where extra material is not stretched out do to loss of pressure. Similar here but do to thermal expansion coefficient and temperature variations across the tank.

    “The on-pad explosion raises some additional questions. In identifying and correcting the problem with the strut, did SpaceX actually address the underlying root cause of the June 2015 failure? Are there any common links between the helium leaks in 2014, the in-flight failure and the launch pad explosion? Or are they simply separate, unrelated problems with a troubled upper stage?”

    This is FAKENEWS. No one is asking that question except in Doug’s World these days. The conditions which led to the COPV failure for Amos 6 involved stratification of the helium exceeding the design limit of the tank. Those conditions are not possible in previous v1.1 configuration without sub-cooled LOX. Furthermore, the upper stage involved in Amos 6 was block III and currently flying is block IV with a plumbing/design change to eliminate the stratification issue, hence they’ve moved back to fast loading of LOX. F9 block V will come a tank design change to mitigate the liner buckles as stated by SpaceX.

  • Terry Rawnsley

    They would also have the right to know everything about the vehicle that they were flying on. Assumption of risk does not apply when all known risks are not disclosed. In other words, you cannot hide behind trade secrets and claim that victims assumed the risk. This is all academic anyway as SpaceX has cooperated with both NASA and the USAF. Anybody that interested can file a FOIA request. It may not be granted but the courts are here to weigh the equities between proprietary interests and public interests and decide accordingly.

  • Andrew Tubbiolo

    The formula is easy remember folks. Risk is public, profit is private. Once you understand that concept you’ll be much happier. Now just focus on the glories of Space X’s recent launch rate and movie on.

    Don’t kid yourself a lot of people actually think this way, the private vested interests certainly do, and Space X is not nearly as bad as the others.

  • Andrew Tubbiolo

    You are correct in pointing out the hypocrisy in the argument, however that does not make your contention correct. Only that it’s easier for the Russians to get away with it than Space X.

  • Hari

    “Anybody that interested can file a FOIA request.” Indeed, and this could include the competitors of SpaceX – anywhere in the world.
    It would also be reassuring to we who are interested whether NASA astronauts currently awaiting launch on Soyuz were informed what precisely the root cause of the carrier rocket or upper stage failure was? Or doesn’t Roscosmos have to abide by American standards of spaceflight safety?

  • Douglas Messier

    NASA didn’t cite proprietary information as a reason for not releasing a public report.

  • Mr Snarky Answer

    NASA HSF is one giant hypocrisy. One minute they are contemplating flying crew on first SLS, as they did on shuttle, all while fly crew on Soyuz. Meanwhile the Russian space program is melting before our eyes, and yet make CCP (Boeing and SpaceX) jump through useless hoop of crazy Loss of Crew criteria.

  • Mr Snarky Answer

    And yet any detailed report would very likely contain proprietary information, such as material selection and manufacturing/installation steps. So perhaps NASA feels it is obvious?

  • roflplatypus

    >most spectacular American launch failure

    Personally, my favorite is https://en.wikipedia.org/wiki/GPS_IIR-1. Nothing like melting a bunch of cars with flaming alumin(i)um.

  • Andrew Tubbiolo

    All large organizations and political movements depend on hypocrisy. There’s no escaping it. I think the best actions you can make against it would be along the lines of two arguments. 1) Making an argument that some types of hypocrisy are tasteful. 2) That they hypocrisy of your enemies is distasteful.

  • Terry Rawnsley

    Why Hari, yes, competitors can file a FOIA request. That request would probably be denied and the courts would uphold the decision because there is no compelling public interest in releasing the information to them. My point, for the last time, is that proprietary information does not AUTOMATICALLY “supercede the public’s right to know.” As for Soyuz, I’m quite certain that NASA possesses all the design specs for the vehicle and have seen all pertinent accident investigation reports. If they believe that the vehicle is safe, they can assign astronauts to the crew. The astronauts themselves are not slaves and I’m fairly certain they can decline the assignment if they aren’t confident in the vehicle. It may cost them their jobs but at least they are aware of the known risks they are being asked to accept.

  • Hari

    In that case you should ask Elon Musk or Gwynne Shotwell to cease and desist from using the proprietary excuse whenever the F9 aborts on the pad or has to be returned to the hangar. If F9 is a clear and present danger to the public, as cigarettes clearly are, then I agree with you. But it isn’t, therefore SpaceX is entitled to some confidentiality.

  • Terry Stetler

    Doug,

    They don’t have to comment since confidentiality is part of the CRS contract, which is a public record. If NASA did comment it would be a breech of contract.

    “5.2. Mishap Investigation and Corrective Action for Mishaps Occurring Post Launch and Prior to Integrated Operations.

    a) An initial investigation by the Contractor is required for all mishaps which have been reported to NASA. NASA reserves discretionary authority to investigate mishaps which involve NASA personnel or resources regardless of location. The Contractor has the discretion to perform any collateral investigations. However, investigations implemented by NASA will take priority with regard to access to evidence, data, and witnesses. ***The proceedings of NASA investigations will remain confidential.*** The Contractor will have an opportunity to comment on the investigation report in accordance with NASA protocols”

  • Greg Brance

    Under the CRS contracts NASA does not own the rockets, and the spacecraft itself. NASA is purchasing a service from a Commercial Company. The flights themselves are licensed by the FAA. The only agency obligated to release a public report is the FAA. However the FAA is not going to do that. There is no stipulation in the CRS contract in the CRS contracts that gives NASA the right to release a public report on a mishap.

    Relevant clause in the CRS contract. Notice the sentence that says the proceedings of NASA investigations will remain confidential. Just because the US taxpayer is paying billions of dollars for the services doesn’t mean the public has a right to know.

    5.2. Mishap Investigation and Corrective Action for Mishaps Occurring Post Launch and Prior to Integrated Operations.

    a) An initial investigation by the Contractor is required for all mishaps which have been reported to NASA. NASA reserves discretionary authority to investigate mishaps which involve NASA personnel or resources regardless of location. The Contractor has the discretion to perform any collateral investigations. However, investigations implemented by NASA will take priority with regard to access to evidence, data, and witnesses. The proceedings of NASA investigations will remain confidential. The Contractor will have an opportunity to comment on the investigation report in accordance with NASA protocols.

  • Terry Stetler

    “I also had the same thought; i.e., FOIA time. There appears to be no credible public interest in not releasing the report; ”

    The public interest is in not violating contract law. The CRS contact has a non-disclosure clause. I posted the language upthread.

  • Douglas Messier

    This is the whole problem. The IG reports says there are various ways to conduct mishap investigations. So you have a contractor led one with substantial NASA participation on the board that results in a detailed public summary. Then you’ve got a SpaceX led investiation with 11 company employees and a lone FAA rep that blames the contractor and a NASA investigation that is kept confidential.

    Two similar accidents in the same N ASA program and two different results in terms of public info.

    So why did N ASA tell me they would release a summary then decide they weren’t going to do it months later?

  • Douglas Messier

    FAKE NEWS! From FAKE NAME GUY!

  • Vladislaw

    “Since it was an FAA licensed flight, NASA is not required to complete a formal final report or public summary,”

    seems pretty cut and dried… NOT REQUIRED

  • Vladislaw

    Because they realized they were “not required” to do so …

  • Mr Snarky Answer

    Not really a problem for most people.

  • Douglas Messier

    I was seriously misled by PAO in multiple communications BEFORE the email last week that NASA would release a public summary. So, not so cut and dried.

  • Douglas Messier

    MORE FAKE NEWS BY FAKE NAME GUY

  • Douglas Messier

    FYI: Here’s the response I received from NASA PAO on Sept. 16, 2016 when I inquired about the status of the report. This story remained consist in period emails into December. It was not until last week that NASA announced it had no obligation and no plans to release the results publicly.

    Hi Doug,

    Thanks for reaching out.

    Let me check on whether the report from the investigation has been completed. We expect to have a summary with publicly releasable information when it is complete, but the report itself won’t be released since it contains information restricted by U.S. International Traffic in Arms Regulations and company-sensitive proprietary information.

    Cheryl

  • Mr Snarky Answer

    I figured there was >95% chance you would be triggered by FAKENEWS

  • Vladislaw

    Was there a name on those communications? Can you share the comms? Post them?

  • Douglas Messier

    See above for the first response from September 2016. There were several other similar emails going through December. At that time, I was told an internal report and public summary would likely be ready by Summer 2017.

  • Jacob Samorodin

    Are you a legal eagle?…Even if you are going on legal precedent, which I doubt, it;s SCOTUS that would ultimately decide the issue if anyone wants to bring this before the courts.

  • Douglas Messier

    Ah, so you ARE a troll. That gives me a lot of options here.

  • Mr Snarky Answer

    One can not be a troll but still keep it a little spicy for fun.

  • publiusr

    No no–only SLS bashing–only that is allowed on space forums–I thought you got the memo.