PARIS – Insufficient screening by SpaceX of commercially sourced hardware on the upper stage of a SpaceX Falcon 9 v1.1 rocket was the most likely cause of a June 2015 launch failure that resulted in the loss of a cargo vessel bound for the International Space Station (ISS), according to a NASA investigation of the mishap.
Both NASA and SpaceX conducted probes into the June 28, 2015 launch failure, though to date only a summary of the NASA-led investigation has been publicly released. The nine-page document, dated March 12, says both agency and company investigations shared conclusions as to the direct and immediate causes of the accident, which resulted following the rupture of the Falcon 9 upper stage liquid oxygen tank.
However, the summary asserts NASA’s determination of the initiating cause of the breach is at odds with the findings of the SpaceX-led inquiry, which attributed the root cause of the mishap to a material defect in an axial strut.
The strut, used to support a helium bottle inside the upper-stage liquid oxygen tank, broke under the stress of G-forces during Falcon 9 ascent. Freed, the helium bottle struck and breached the oxygen tank’s dome, destroying the stage and resulting in the loss of NASA’s seventh Commercial Resupply Services (CRS-7) mission.
According to the NASA summary, the strut broke because “SpaceX chose to use an industrial grade (as opposed to aerospace grade) 17-4 PH SS (precipitation-hardening stainless steel) cast part … in a critical load path under cryogenic conditions and strenuous flight environments.” The summary adds that SpaceX disregarded the supplier’s recommendation for a 4:1 factor of safety and used it without proper modeling or sufficient load testing under expected flight conditions.
In comments made less than a month after the mishap, SpaceX Chief Executive Elon Musk said the strut had flown hundreds of times on the Falcon 9 upper stage, and that the company had subjected the part to “a whole bunch of tests on the ground,” though it had missed what he said was a material flaw.
“The strut that we believe failed was designed and material certified to handle 10,000 lbs. of force, but actually failed at 2,000 lbs. of force, which is a five-fold difference,” Musk said during a July 20, 2015 teleconference. “In the future, obviously, we’re not going to use these particular struts, and we’re going to move to individually test each strut independent of any material certifications,” a change he said would result in some cost increase to the rocket, but likely not enough to affect the price offered to customers.
While the NASA investigation found that breakage of the axial strut was credible, it did not denote it as the most probable cause of the mishap, since the agency also viewed manufacturing damage, improper installation, collateral damage, or some other part of the axial strut breaking as equally credible causes of the liberated helium bottle.
Instead, NASA’s findings attributed the most probable cause of the mishap to SpaceX’s decision to use the commercial-grade part absent adequate modeling and loads testing, a “design error” directly related to the launch failure, the summary concludes.
Shortly after the launch failure, NASA was invited to participate in the SpaceX-led accident investigation, along with experts from the FAA and the U.S. Air Force. However, one month after the probe began, the agency tapped its Launch Services Program (LSP) to independently review the mishap: with a subsequent NASA launch on the Falcon 9 manifest – the Jason-3 ocean altimetry satellite – the agency sought to ensure SpaceX implemented any corrective actions resulting from the probe.
LSP briefed its results to senior NASA leadership on Dec. 18, 2015, attributing the mishap to a SpaceX design error and recommending the company pay more attention when using industrial-grade parts.
In addition, both the NASA and SpaceX-led investigations found room for improvement on the Falcon 9 upper stage, as well as in processes associated with its telemetry, insight that Musk said was welcome.
“We’re looking not just for the issues that caused this failure but any near misses that could potentially affect future flights,” he said in the July 2015 teleconference, noting that he expected the investigation and subsequent recommendations to result in a safer launch vehicle as the company prepares to carry astronauts to the ISS.
Among the additional findings, according to the summary, was the need for SpaceX to establish and maintain proper nitrogen purge rates for stage testing and launch base operations. The company was likewise directed to pay more attention to manufacturers recommendations when using other commercially-sourced hardware, notably the use of wire ropes to provide structural support the Falcon 9 liquid-oxygen transfer tube assembly.
The report also noted SpaceX’s use of non-deterministic network packets in their flight telemetry increases latency, resulting in substantial portions of lost anomaly data. It recommended SpaceX “re-think” a new telemetry architecture for the Falcon 9 v1.2, a subsequent iteration of the rocket, in an effort to speed the delivery of telemetry.
In its summary, LSP said that SpaceX had corrected or mitigated all of the investigation’s credible causes and technical findings ahead of the Jason-3 mission. That flight was the last of the Falcon 9 version 1.1 launch vehicle, and flew successfully on Jan. 17, 2016.
Amy Svitak is a Space Intel Report contributing editor. She can be reached at firstname.lastname@example.org