SpaceX Eyes Falcon 9 Return to Flight on Jan. 8


SpaceX Accident Investigation Update
January 2, 2017, 9:00am EST

Over the past four months, officials at the Federal Aviation Administration (FAA), the U.S. Air Force (USAF), the National Aeronautics and Space Administration (NASA), the National Transportation Safety Board (NTSB), along with several industry experts, have collaborated with SpaceX on a rigorous investigation to determine the cause of the anomaly that occurred September 1 at Space Launch Complex 40 (SLC-40) at Cape Canaveral Air Force Station in Florida. This investigation team was established according to SpaceX’s accident investigation plan as approved by the FAA. As the primary federal licensing body, the FAA provided oversight and coordination for the investigation.

Investigators scoured more than 3,000 channels of video and telemetry data covering a very brief timeline of events – there were just 93 milliseconds from the first sign of anomalous data to the loss of the second stage, followed by loss of the vehicle. Because the failure occurred on the ground, investigators were also able to review umbilical data, ground-based video, and physical debris. To validate investigation analysis and findings, SpaceX conducted a wide range of tests at its facilities in Hawthorne, California and McGregor, Texas.

The accident investigation team worked systematically through an extensive fault tree analysis and concluded that one of the three composite overwrapped pressure vessels (COPVs) inside the second stage liquid oxygen (LOX) tank failed. Specifically, the investigation team concluded the failure was likely due to the accumulation of oxygen between the COPV liner and overwrap in a void or a buckle in the liner, leading to ignition and the subsequent failure of the COPV.

Each stage of Falcon 9 uses COPVs to store cold helium which is used to maintain tank pressure, and each COPV consists of an aluminum inner liner with a carbon overwrap. The recovered COPVs showed buckles in their liners. Although buckles were not shown to burst a COPV on their own, investigators concluded that super chilled LOX can pool in these buckles under the overwrap. When pressurized, oxygen pooled in this buckle can become trapped; in turn, breaking fibers or friction can ignite the oxygen in the overwrap, causing the COPV to fail. In addition, investigators determined that the loading temperature of the helium was cold enough to create solid oxygen (SOX), which exacerbates the possibility of oxygen becoming trapped as well as the likelihood of friction ignition.

The investigation team identified several credible causes for the COPV failure, all of which involve accumulation of super chilled LOX or SOX in buckles under the overwrap. The corrective actions address all credible causes and focus on changes which avoid the conditions that led to these credible causes. In the short term, this entails changing the COPV configuration to allow warmer temperature helium to be loaded, as well as returning helium loading operations to a prior flight proven configuration based on operations used in over 700 successful COPV loads. In the long term, SpaceX will implement design changes to the COPVs to prevent buckles altogether, which will allow for faster loading operations.

SpaceX is targeting return to flight from Vandenberg’s Space Launch Complex 4E (SLC-4E) with the Iridium NEXT launch on January 8. SpaceX greatly appreciates the support of our customers and partners throughout this process, and we look forward to fulfilling our manifest in 2017 and beyond.

Editor’s Note: So, this is a two-step solution. In the interim, they’re still flying with helium tanks that could have these buckles (voids) in them. But, they believe that warmer helium and changes in the helium loading procedures will prevent another explosion. The long term solution is to redesign the COPV’s to prevent the buckles.

In essence, SpaceX doesn’t want the Falcon 9 grounded while they redesign and test the COPV’s.  The question is how long is that going to take? How many flights are there going to be with COPV’s inside the LOX tank that may have the same buckles in them?

The other question is this: SpaceX seems to be saying they were using a different helium loading procedure when the rocket was destroyed than they had used previously. Was it just for this one occasion? Had they used it on previous flights? It would astonish me if they were testing some new procedure for the first time with a $195 million satellite atop of the booster.

I’d also note that I don’t see any reference to probable cause. Instead they’re talking about “credible” causes. I also don’t see any reference to the FAA signing off on the launch next week.





  • Saturn13

    Going back to the way they did 700 successful COPV loads is a good idea. I know water turning into ice expands. So LOX frozen does the same thing and trapped broke filaments causing enough friction heat to ignite the carbon, maybe. Or t

    he broke stran

  • Eric Rusler

    Elon did talk about the aluminum liner buckling so… there’s a few more kinks to work out of those tanks. I thought the previous mfr for those tanks did not use aluminum liners.

  • patb2009

    Water turning into ice and expanding is a pretty unique property of water.
    There are about 6 other materials also known to expand when cooled ( Silicon, Germanium, Gallium, Bismuth, Antimony, Acetic acid).

    Now what probably happened was the Aluminum liner was cooling as the tanks were loaded with LHe, then LOx was added to the tanks andd began
    permeating through little gaps in the composite and condensing and
    freezing, then the LHe was pressurized, the liner pushed out hard, and tripped the event…

    COPV under pressure and LOX may be incompatible.

  • SOX is actually more dense than LOX.

    They’re not loading liquid helium. (Helium is gaseous down to about 4°K at one atmosphere.) It’s gaseous, but fairly cold. Gaseous helium is kinda strange in the range near the subcooled LOX temperature (about 66°K). Above its inversion temperature of 45°K, gaseous helium actually gets hotter as it expands through a resistance valve. Below that temperature, it cools (which is what almost every other gas does at temperatures that are much, much higher).

    I have no clue what the full set of thermodynamics are for the helium loading. Presumably, the COPV starts out at 1 atmosphere and high pressure expands into it. Later, it will have to be pumped in to be pressurized up to about 300 atmospheres.

    You could see how fairly minor changes in the He temp could radically change the temperature of the tank in the early loading stages. Later, I’d expect almost everything to raise the temperature of the tank, but the buckling phenomenon is going to occur when you set up some kind of temperature gradient between the LOX and He sides of the liner. That only requires a fairly small transient to occur.

  • From the countdown timeline, it looks like the LOX is loaded first, then the gaseous (not liquid) He is loaded, but then there’s an item labelled “Stage 2 LOX Flow Adjustment for Helium Cryo Load”. Take a look between T-19:30 and T-13:00.

    My money’s on the early part of the helium load requiring that high-pressure He expand under resistance into the COPV when it’s nearly at atmospheric pressure. He’s inversion point if 45°K, so the old load procedure may have actually warmed the tank at the beginning, while fairly small reductions in the He initial temperature (below the inversion point) could then cause the tank to get quite a bit colder.

    Buckling is going to require a pretty hefty gradient across the liner. Something was making the He side of the liner a lot colder.

  • Eric R.
  • I think that, through most of the helium loading, the aluminum liner will expand more via hoop stress from the high-pressure helium than it will thermally contract.

    My guess is that all of the unpleasantness occurred very early in the helium fill cycle, before pressure was high enough to stabilize the liner and when Joule-Thomson effects caused the temperature of the gas to drop a lot. Get the inlet He above 45°K and the Joule-Thomson effects will warm the He, not cool it.