SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD (1)

[CameronD]

Whatever happened, I don't think it was the LOX tank overpressurizing in stage 2...

Agreed ..but I figure there aren't that many things up there that can cause an explosion like this.

We can plainly see the RP-1 tank rupture, but if RP-1 loading is complete and the tank is full it isn't going to explode on it's own even with a decent-sized spark nearby.  Can't be very many scenarios to check off of the list..


EDIT:  How about thermal failure?.. maybe a valve chilling down too fast or something (like part of a tank shell) not properly insulated that freezes, cracks and sparks??  LOX can do nasty things to gear that's not ready for it.

[punder]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

[Kabloona]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

Don't know, but Helodriver got a closeup photo of the returned stage outside Hawthorne showing the pressure gauges for the tanks. IIRC, one was around 10 psi and the other was less than 5 psi, so the difference there was in the single digits.

[CameronD]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

That wouldn't cause a bang in and of itself - this stuff is well below ignition point.  There'd first be a spray of stuff outward like we saw with the S2 launch failure.  This time the 'bang' seems to happen first.

..and it must have been at or near the bottom of the RP-1 tank for it to lose it's contents downwards like that.

[Prober]

time for a lessons learned?

Maybe dropping the trunk is a good idea now?

What lesson? The trunk is required for passive aerodynamic stability after SD's stop firing, while the stack is coasting up to apogee. Without the trunk, the capsule will tumble immediately after SD's stop firing, with unpleasant consequences for passengers.

know that, several fixes for that but you need to drop some weight in my view or?


btw: doesn't SpaceX use the helium for the release of Dragon from the Falcon 9?   

[intrepidpursuit]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

That wouldn't cause a bang in and of itself - there'd first be a spray of stuff outward like we saw with the S2 launch failure.  This time the 'bang' seems to happen first.

In CRS-7 it appeared that the LOX tank ruptured at the top rather than at the common bulkhead. If the Common bulkhead ruptures we won't see it right away but allows the fuel to mix to setup a scenario for a kaboom. Then there just has to be an ignition source.

[John-H]

From the location of the umbilical, can I assume that the fuel goes directly to the tank, while the LOX pipe runs through  or around the fuel tank. If there is a leak in this pipe,  an explosive mixture could build up and then go off suddenly, causing a flash and then a tank breech.

John 

[jongoff]

In some ways I hope for SpaceX's case that is was some sort of subtle vehicle failure--one they can catch and design-out, not a GSE failure. The reason I say this is that the last time a US rocket blew up on the pad without an engine failure of some sort was over half a century ago. There've been well over 1000, and possibly over 2000 liquid fueled rocket launches since then, without any of them blowing up on the pad due to GSE issues. So having a pad systems failure actually makes SpaceX look a lot less professional than if it was a subtle design flaw.

On the what actually happened, it still really looks like the failure started inside the stage, not an external explosion that happened to rupture the tanks. That's not objective fact, and I may be misreading it, but that's what it looked like from the video. I just don't see some sort of "both umbilicals leaked in just the right way to also catch a spark" sort of scenario as being realistic. It's wild speculation, but I still think something to do with the common bulkhead did it.

People keep pointing out that it didn't look like the CRS-7 overpressurization, but that was a much slower event caused by a tube breaking, which meant you would've had choked flow out of a small diameter line. If you had a more rapid overpressurization event, it might look totally differently. A COPV failing more dramatically for instance might happen much. much faster, especially if it ruptured the common bulkhead. If say a dome came off of a COPV, it would probably be going fast enough that the whole bulkhead would be ruptured in less than one frame of the video. And the energy from that sort of a failure would not only mix the propellants, but could also quite possibly ignite the mixed propellants.

My Rambling $.02

~Jon

[CameronD]

From the location of the umbilical, can I assume that the fuel goes directly to the tank, while the LOX pipe runs through  or around the fuel tank. If there is a leak in this pipe,  an explosive mixture could build up and then go off suddenly, causing a flash and then a tank breech.

No, the LOX is colder than the RP-1 and at a higher pressure.  If there's a leak in the pipe LOX might enter the RP-1 tank (which is already full) and get released from a vent someplace which might then go 'bang' somewhere else.

[punder]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

That wouldn't cause a bang in and of itself - there'd first be a spray of stuff outward like we saw with the S2 launch failure.  This time the 'bang' seems to happen first.

In CRS-7 it appeared that the LOX tank ruptured at the top rather than at the common bulkhead. If the Common bulkhead ruptures we won't see it right away but allows the fuel to mix to setup a scenario for a kaboom. Then there just has to be an ignition source.

Electric signal and power wires could be sheared by a bulkhead that fails catastrophically under pressure? The liquid forms a short across suddenly exposed lines of different potential.

[OneSpeed]

If we have a "graphics wiz" that can superimpose a Falcon cutaway over a video still, that would be great!

I'm hardly a 'graphics wiz', but it looks like the epicentre is adjacent the common bulkhead.

[Craftyatom]

Is the decision to have the payload attached our not attached during a static fire at the discretion of the customer?

I was looking back at static fire video and JCSAT-14,16 no payload attached and the same with EUTELSAT-ABS.

However Thaicom-8 payload was attached.  Also for CRS-8, CRS-9 Dragon was attached.

SpaceX really dodged a bullet that this didn't happen on CRS-9.

I believe earlier info says it's the customer's decision.  CRS flights are attached because it's SpaceX's ship, and they decide to do it that way, for both mating and testing reasons (helps build confidence in attachments, resonance modes, electrical systems, etc).

Customers are generally given the choice between having their payload on during the static fire, which speeds up the timeline a bit and gives better test data, or not having it attached, which means the payload won't be damaged in the event of an anomaly.

I personally hope SpaceX and its customers continue to mate prior to static firing.  The chance of the type of anomaly seen today occurring is low, even before taking into account the fact that SpaceX will be taking a long, hard look at every inch of these systems in the next 6 months.  Choosing to give up test data because of such a slim chance seems a waste to me, but then again, it depends on whether your "launch insurance" is also "integration and testing insurance".

[intrepidpursuit]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

That wouldn't cause a bang in and of itself - there'd first be a spray of stuff outward like we saw with the S2 launch failure.  This time the 'bang' seems to happen first.

In CRS-7 it appeared that the LOX tank ruptured at the top rather than at the common bulkhead. If the Common bulkhead ruptures we won't see it right away but allows the fuel to mix to setup a scenario for a kaboom. Then there just has to be an ignition source.

Electric signal and power wires could be sheared by a bulkhead that fails catastrophically under pressure? The liquid forms a short across suddenly exposed lines of different potential.

There is very little wiring inside the tanks and nothing would run through the common bulkhead. But I agree that once you have a massive failure like that there are lots of moving parts (that shouldn't be) and the potential of something producing a spark is much greater.

[Kabloona]

In some ways I hope for SpaceX's case that is was some sort of subtle vehicle failure--one they can catch and design-out, not a GSE failure. The reason I say this is that the last time a US rocket blew up on the pad without an engine failure of some sort was over half a century ago. There've been well over 1000, and possibly over 2000 liquid fueled rocket launches since then, without any of them blowing up on the pad due to GSE issues. So having a pad systems failure actually makes SpaceX look a lot less professional than if it was a subtle design flaw.

On the what actually happened, it still really looks like the failure started inside the stage, not an external explosion that happened to rupture the tanks. That's not objective fact, and I may be misreading it, but that's what it looked like from the video. I just don't see some sort of "both umbilicals leaked in just the right way to also catch a spark" sort of scenario as being realistic. It's wild speculation, but I still think something to do with the common bulkhead did it.

People keep pointing out that it didn't look like the CRS-7 overpressurization, but that was a much slower event caused by a tube breaking, which meant you would've had choked flow out of a small diameter line. If you had a more rapid overpressurization event, it might look totally differently. A COPV failing more dramatically for instance might happen much. much faster, especially if it ruptured the common bulkhead. If say a dome came off of a COPV, it would probably be going fast enough that the whole bulkhead would be ruptured in less than one frame of the video. And the energy from that sort of a failure would not only mix the propellants, but could also quite possibly ignite the mixed propellants.

My Rambling $.02

~Jon

All good thoughts. Only problem is, repercussions from a COPV failure (ie major redesign) are going to be a lot more painful for them than a simple GSE issue. So I'm going to take the other side and hope it was in fact a GSE issue.

[punder]

The event is so sudden, instantaneous really, that it makes me think the most likely cause is failure of the common bulkhead with instant mixing of the LOX and kero. What is the pressure differential between the tanks?

That wouldn't cause a bang in and of itself - there'd first be a spray of stuff outward like we saw with the S2 launch failure.  This time the 'bang' seems to happen first.

In CRS-7 it appeared that the LOX tank ruptured at the top rather than at the common bulkhead. If the Common bulkhead ruptures we won't see it right away but allows the fuel to mix to setup a scenario for a kaboom. Then there just has to be an ignition source.

Electric signal and power wires could be sheared by a bulkhead that fails catastrophically under pressure? The liquid forms a short across suddenly exposed lines of different potential.

There is very little wiring inside the tanks and nothing would run through the common bulkhead. But I agree that once you have a massive failure like that there are lots of moving parts (that shouldn't be) and the potential of something producing a spark is much greater.

Conduits outside the tanks and maybe electrical interfaces with the t/e. Conductive liquids, flying metal and sheared wires...

[CameronD]

Conduits outside the tanks and maybe electrical interfaces with the t/e. Conductive liquids, flying metal and sheared wires...

The liquids in this case are non-conductive and hundreds of degrees F below ignition point.  No, I'm sticking with Jon's idea that a COPV failure or some similar plumbing failure inside the stage itself triggered the bulge pointed out several pages upthread.

[intrepidpursuit]

Conduits outside the tanks and maybe electrical interfaces with the t/e. Conductive liquids, flying metal and sheared wires...

The liquids in this case are non-conductive and hundreds of degrees F below ignition point.  No, I'm sticking with Jon's idea that a COPV failure or some similar plumbing failure inside the stage itself triggered the bulge pointed out several pages upthread.

O2 overpressure alone doesn't make a kaboom, just a pop. There has to be a mixing of fuels an an ignition source in order to cause an explosive boom like we saw. That is what we are trying to explain. Elon already confirmed the problem was in the S2 O2 tank, that is not in question.

[nisse]

Is it really a fireball we see at first? Could a stuck lox-venting-valve cause an overpressure explosion? I mean they are near oxygen freezing temperatures which could form solid oxygen that plugs some venting valve. Once you have a overpressure explosion the rest you see happens.

[RotoSequence]

Elon already confirmed the problem was in the S2 O2 tank, that is not in question.

Elon confirmed that the problem "originated around (the) upper stage oxygen tank," not necessarily that the fault was in or of the O2 tank itself.

[northenarc]

 My take is its failure of a COPV, or associated helium plumbing, and sudden pressure introduction blew off/out one of the umbilical lines or associated internal plumbing which started a rapid external fire that propagated back inwards and further over pressured the LOX tank until the 'horizontal failure' seen on the video.