I've just read the phys.org new item which summarizes the conclusions by SpaceX about the cause of the September 1st 2016 anomaly.

I'm citing quotes in the news item which come from a SpaceX announcement which is further quoting investigators, so I am not sure I'm getting the correct picture yet:

"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."

"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."

Question: Does the term "buckles" refer to existing static indentations of some kind in the as-fabricated aluminum liner, or to spaces that are formed by the aluminum liner actually buckling during the loading process?

Update: The 2018 Florida Today article After 2016 launch pad explosion, SpaceX updating Falcon 9 for astronauts suggests that it is the tank that was buckled (lower your volume to avoid advertisements):

Changes in helium loading operations prevented a repeat of the problem believed to have caused the explosion: buckles in helium tanks that trapped slushy liquid oxygen in gaps between the tanks and their composite wraps. Friction "or other mechanisms" ignited oxygen in the upper stage.

While Falcon launches resumed fairly quickly, a redesigned helium bottle has not yet completed testing nor won NASA’s approval for crew launches.

enter image description here

The next launch date has now been announced as January 8, 2017, just a few days away. Does the plan to avoid this happening in the immediate future consist primarily of ways to change the helium loading process to keep it from getting so cold that liquid oxygen in these trapped spaces simply can't freeze into solid oxygen (SOX), or are they also going to try to prevent these pocket of LOX from forming between the Aluminum liner and the carbon overwrap in the first place?

Aluminum in contact with pure oxygen, carbon in contact with pure oxygen, huge stresses, huge vibrations, only a bit of friction necessary to start a reaction... no physical barrier?

He built gradually to a crying jag, during which he claimed to be deeply touched by the idea of an inhabited planet with an atmosphere that was eager to combine violently with almost everything the inhabitants held dear. He was speaking of Earth and the element oxygen.

“When you think about it boys”, he said brokenly, “that’s what holds us together more than anything else, except maybe gravity. We few, we happy few, we band of brothers - joined in the serious business of keeping our food, shelter, clothing and loved ones from combining with oxygen.”

--Kurt Vonnegut, God Bless You, Mr. Rosewater (1965)

  • 1
    $\begingroup$ I'd like to know more about these "buckles" too. If they protrude into the tank, how the heck did that happen? I could see "bulges" protruding out of the tank, but "buckles"? $\endgroup$ – Organic Marble Jan 3 '17 at 17:24
  • 1
    $\begingroup$ Differential contraction of the Al and CF? $\endgroup$ – Hobbes Jan 3 '17 at 18:07
  • 2
    $\begingroup$ Not an expert, but I interpreted the "buckles" as being voids left after wrapping, possibly because the wrapping itself buckled a bit. I've seen similar when wrapping a sheet of something around a cylinder (although, I'm talking arts and craft here, not rocket science!) $\endgroup$ – Steve Jan 3 '17 at 21:25

In this context, Buckles are voids between carbon composite overwrap and aluminum liner.


Here is a quote from the NASA ASAP:

The SpaceX investigation did not find a single most probable cause of the initiating event, instead identifying several credible causes involving the COPV helium tanks. All credible causes were similar in that they involved LOX trapped between the overwrap and the liner with subsequent ignition through friction or other mechanisms.

The evidence recovered from the mishap showed indications of buckles in the COPV liner where LOX was likely trapped. Acting from the report findings, SpaceX was able to recreate a buckle event during a COPV test. Additional testing allowed SpaceX to identify specific conditions which would cause a buckle and trap oxygen in the gap between the liner and overwrap.

Using this data, SpaceX modified its helium loading configuration, process, and controls to ensure that the COPVs would not be exposed to these identified conditions and, accepting any residual risk, successfully resumed commercial launches with the existing COPV design. However, to further improve safety, SpaceX and NASA agreed that a redesign of the COPV was necessary to reduce the risk for missions with crew onboard. Using what they learned from the mishap investigation, SpaceX redesigned the COPV and NASA started a rigorous test program to characterize the behavior of the new COPV in the cryogenic oxygen environment.

ASAP Report

I found one source that says:

“Each COPV consists of an aluminum inner liner with a carbon overwrap.”

Given that the tensile strength of CF is much higher than AL, it's likely the AL yielded to some small extent.

COPV construction

  • $\begingroup$ "Question: Does the term "buckles" refer to existing static indentations of some kind in the as-fabricated aluminum liner, or to spaces that are formed by the aluminum liner actually buckling during the loading process?" It looks like your source avoids defining "a void". $\endgroup$ – uhoh Apr 16 '18 at 7:17
  • $\begingroup$ Assuming the two components were in contact prior to the formation of the buckle, which one moved to form the void? $\endgroup$ – Organic Marble Apr 16 '18 at 17:50
  • 2
    $\begingroup$ Almost undoubtedly it would be the liner that would buckle. COPVs are typically built so that the overwrap is permanently under tension. Using a process called autofrettage (en.wikipedia.org/wiki/Autofrettage), after overwrapping is complete, the vessel is pressurized past the yield point of the liner. The end result of this is that when the pressure vessel is unloaded, the liner is actually under compression. This reduces the tensile stress on the liner when the vessel is loaded up. $\endgroup$ – Tristan Jun 13 '18 at 15:12

My reading was that the buckles were in the carbon fibre overwrap, but on the underside, thus the "buckles under the overwrap".

  • $\begingroup$ hmm... the overwrap is something that can buckle? I'm thinking a buckle like a metal shell would do. Isn't the carbon fibre overwrap cured with epoxy or something hard and brittle? $\endgroup$ – uhoh Feb 3 '17 at 16:07
  • 1
    $\begingroup$ Ah, buckle means bend/flex.CF is stiff, but under extreme temp stuff, things bend. $\endgroup$ – geoffc Feb 3 '17 at 17:19
  • $\begingroup$ I noticed the phrase "buckles in helium tanks that trapped slushy liquid oxygen in gaps between the tanks and their composite wraps" in this Florida Today article (lower your volume to avoid annoying furniture advertisements). Do you still think "buckles" refers to the overwrap, and not to the tank itself? $\endgroup$ – uhoh Jan 13 '18 at 1:55
  • 1
    $\begingroup$ My reading of that same material led me to believe different contraction rates between the different materials caused the wraps to loosen at some stages of filling. If you wrap string around a balloon and the balloon shrinks, the string can flop loose, leaving a gap where they were once snug. That's what I understood the buckles to be. $\endgroup$ – Saiboogu Apr 16 '18 at 20:25

I've found some buckles!

From Scott Manley's video The Dumbest Mistakes In Space Exploration

tl;dr: The metal tank buckles inward

Yes it was sort-of funny when SpaceX had a rocket explode on the pad, but the reason for that was down to some pretty complicated material science and physics.

The slides show a buckled wall of a gas-filled metal tank (with carbon overwrap):

If Exterior pressure is great wall buckles in slightly, creating a void.

Because helium is colder the liquid oxygen can freeze into solid oxygen.

That says clearly that yes, the metal tank buckles inward.

As helium is loaded, the rising pressure pushes out SOX, which can not flow out, adding extra stress to small regions on the composite cover.

enter image description here

enter image description here

enter image description here

  • 1
    $\begingroup$ Good find! Makes sense at last. $\endgroup$ – Organic Marble Mar 8 at 13:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.