SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD (2)

[woods170]

Addressing the anomaly:

“We are close to figuring it out. It might have been formation of solid oxygen in the carbon over-wrap of one of the bottles in the upper stage tanks. If it was liquid it would have been squeezed out but under pressure it could have ignited with the carbon. This is the leading theory right now, but it is subject to confirmation. 

How is that not a flaw in the vehicle and a "business process error"?

I suggest you get up to speed on your knowledge of cryogenics and material interactions with liquids under cryogenic conditions. You will be surprised to learn what can happen. Some of those lessons were learned the hard way on NASA spacecraft going back as far as the late 1970's. One lesson learned is that it does not always require a hardware change to keep things from going wrong. Proper procedures can go a long way to keep trouble away.

[jpo234]

If you fill high pressure He into an empty bottle, the He will expand. An expanding gas gets colder.

The He is compressed. It will get hotter.

...and it's compressed a lot.

Cheers, Martin

Maybe they compress the helium, then cool the compressed gas to 60K or so, then use the cooled gas that to fill the bottles.  Without doing the calculation, I'd think they have to do something like this.  If they loaded room temperature or above helium, then they would have to wait until the LOX cooled the bottle (and its contents) to cram enough helium in.  That would be a slow process, I'd think, since the gas within the bottle is not circulating.

Exactly. And cold high pressure He expanding into a partially filled COPV will get colder (unless there are esoteric effects going on, like the Joule-Thompson effect).

[woods170]

  The other thing we discovered is that we can exactly replicate what happened on the launch pad if someone shoots the rocket.

"Exactly replicate"? That, to me, implies that they have shot a partialy fueled upper stage and saw it explode in the same fashion, which I doubt they have done so this:

We don’t think that is likely this time around, but we are definitely going to have to take precautions against that in the future. We looked at who would want to blow up a SpaceX rocket. That turned out to be a long list. I think it is unlikely this time, but it is something we need to recognize as a real possibility in the future.”

... is a joke right?

Let's assume that Elon actually said this. Then it's no joke. Well, at least not to him.

[Jim]

I suggest you get up to speed on your knowledge of cryogenics and material interactions with liquids under cryogenic conditions. You will be surprised to learn what can happen. Some of those lessons were learned the hard way on NASA spacecraft going back as far as the late 1970's. One lesson learned is that it does not always require a hardware change to keep things from going wrong. Proper procedures can go a long way to keep trouble away.

That is not a  "business process error", having LOX penetrate the overwrap is a design flaw.   There can be work done to the COPV's to prevent this.

[Fan Boi]

Planning for the potential of someone shooting at the rocket? Wow, going to switch to KOPV I suppose (Kevlar Overwrapped Pressure Vessel).
kidding of course...

[Norm38]

First, take shooting the rocket to the wacky thread please

As noted above, Helium doesn't adiabatically cool from expansion (at these temps, at least). Need another cooling mechanism.

Please see Page 8 of this NASA reference on Joule-Thompson Inversion Curves.
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720020315.pdf

If I read the Helium graph correctly, then to get a positive coefficient at 50K, requires ~15MN/m^2 of pressure, or ~150atm.  If flight COPV pressure is ~300atm, then there would be a fair amount of time where the filling of the COPV would have a negative coefficient, and the helium would cool as the tank filled?

Am I reading that right?  Below 150atm the COPVs cool as filled, and above 150atm they heat up?

[Jakusb]

How sure are we that this actually has been said by Elon? How credible is the source?
Sorry if this already has been established. If so feel free to remove this post.

[Norm38]

That is not a  "business process error", having LOX penetrate the overwrap is a design flaw.   There can be work done to the COPV's to prevent this.

Having just finished a DFMEA review, I agree.  Yes, risk can be managed by careful control of operating procedures to keep the LOX from locally freezing.  But risk is better mitigated by not allowing LOX to penetrate the COPV in the first place.

[woods170]

I suggest you get up to speed on your knowledge of cryogenics and material interactions with liquids under cryogenic conditions. You will be surprised to learn what can happen. Some of those lessons were learned the hard way on NASA spacecraft going back as far as the late 1970's. One lesson learned is that it does not always require a hardware change to keep things from going wrong. Proper procedures can go a long way to keep trouble away.

That is not a  "business process error", having LOX penetrate the overwrap is a design flaw.   There can be work done to the COPV's to prevent this.

Actually, that is very likely not a design flaw. It's practically impossible to prevent oxygen molecules from penetrating the overwrap, given a COPV submerged in LOX. This is something Air Liquide found out in the 1970's. It's the very reason why Ariane 5 has it's COPV's outside the tanks.
Having oxygen molecules nestle inside the wrap or even between the wrap and aluminium shell is not necessarily a problem, as long as those molecules are in a liquid state. If however SpaceX managed to get oxygen molecules in a frozen state in between the wrap and the shell, well then they found something they probably had not expected. It very well could have been the result of changed operational procedures.

[savuporo]

Having oxygen molecules nestle inside the wrap or even between the wrap and aluminium shell is not necessarily a problem, as long as those molecules are in a liquid state. If however SpaceX managed to get oxygen molecules in a frozen state in between the wrap and the shell, well then they found something they probably had not expected. It very well could have been the result of changed operational procedures.

Subcooled LOX at 66K cant be helping here

[woods170]

That is not a  "business process error", having LOX penetrate the overwrap is a design flaw.   There can be work done to the COPV's to prevent this.

Having just finished a DFMEA review, I agree.  Yes, risk can be managed by careful control of operating procedures to keep the LOX from locally freezing.  But risk is better mitigated by not allowing LOX to penetrate the COPV in the first place.

The only practical way to prevent LOX from penetrating the composite overwrap of a COPV is by not immersing the COPV in LOX. I guess none of you here have heard of thermal micro-cracking of composite materials? NASA, Ball Aerospace, ESA and Air Liquide have boatloads of knowledge on this thanks to work done for spacecraft such as IRAS, COBE, ISO and Herschell.

Going into speculation from this point forward but I have a feeling that on Amos-6 SpaceX has discovered an unexpected result of using sub-cooled LOX.

[Dante80]

Is there a possibility that the procedure was not changed for Amos-6, SpaceX never took into account the sub-cooled LOX freezing possibility and the rest of the FT LVs launched hadn't exploded by sheer luck? This would make the failure a design issue, not an operations one.

[jpo234]

How sure are we that this actually has been said by Elon? How credible is the source?
Sorry if this already has been established. If so feel free to remove this post.

The marksman comment in the thread came from a known SpaceX employee.

[envy887]

Is there a possibility that the procedure was not changed for Amos-6, SpaceX never took into account the sub-cooled LOX freezing possibility and the rest of the FT LVs launched hadn't exploded by sheer luck? This would make the failure a design issue, not an operations one.

Between testing, pre-launch operations and launches, they have done hundreds of subcooled tanking cycles on dozens of COPVs. If this was a random occurrence so rare that it didn't happen through all that, I'd be surprised if they were able to replicate it this quickly.

[HMXHMX]

That is not a  "business process error", having LOX penetrate the overwrap is a design flaw.   There can be work done to the COPV's to prevent this.

Having just finished a DFMEA review, I agree.  Yes, risk can be managed by careful control of operating procedures to keep the LOX from locally freezing.  But risk is better mitigated by not allowing LOX to penetrate the COPV in the first place.

The only practical way to prevent LOX from penetrating the composite overwrap of a COPV is by not immersing the COPV in LOX. I guess none of you here have heard of thermal micro-cracking of composite materials? NASA, Ball Aerospace, ESA and Air Liquide have boatloads of knowledge on this thanks to work done for spacecraft such as IRAS, COBE, ISO and Herschell.

Going into speculation from this point forward but I have a feeling that on Amos-6 SpaceX has discovered an unexpected result of using sub-cooled LOX.

Interestingly, the matrix material for these vessels is urethane, which should minimize or eliminate micro-cracking, as its CTE is well matched to the fiber (unlike epoxy).  I've tested a lot of carbon-epoxy + urethane-coated composites, but never got ignition at normal tank pressures (we are talking >1300 tests).  But to take your point, I've never used sub-cooled LOX.

[Dante80]

Is there a possibility that the procedure was not changed for Amos-6, SpaceX never took into account the sub-cooled LOX freezing possibility and the rest of the FT LVs launched hadn't exploded by sheer luck? This would make the failure a design issue, not an operations one.

Between testing, pre-launch operations and launches, they have done hundreds of subcooled tanking cycles on dozens of COPVs. If this was a random occurrence so rare that it didn't happen through all that, I'd be surprised if they were able to replicate it this quickly.

Agreed, but Musk says that the leading theory has not been confirmed yet. This would be compatible with what I said.

[envy887]

First, take shooting the rocket to the wacky thread please

As noted above, Helium doesn't adiabatically cool from expansion (at these temps, at least). Need another cooling mechanism.

Please see Page 8 of this NASA reference on Joule-Thompson Inversion Curves.
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720020315.pdf

If I read the Helium graph correctly, then to get a positive coefficient at 50K, requires ~15MN/m^2 of pressure, or ~150atm.  If flight COPV pressure is ~300atm, then there would be a fair amount of time where the filling of the COPV would have a negative coefficient, and the helium would cool as the tank filled?

Am I reading that right?  Below 150atm the COPVs cool as filled, and above 150atm they heat up?

I don't know, I've never filled a helium vessel. Can someone with experience relate on the temperature vs. pressure curve?

[dror]

Actually, that is very likely not a design flaw. It's practically impossible to prevent oxygen molecules from penetrating the overwrap, given a COPV submerged in LOX. This is something Air Liquide found out in the 1970's. It's the very reason why Ariane 5 has it's COPV's outside the tanks.
...

Hence, a design flaw (?!)

Going into speculation from this point forward but I have a feeling that on Amos-6 SpaceX has discovered an unexpected result of using sub-cooled LOX.

Isn't that a flaw in the design?

[rsdavis9]

Every design flaw that has a process to not aggravate the flaw becomes a feature with operating parameters to avoid.
So design flaw or feature.
I'm a software engineer we often time "label" some shortcoming as a feature so as not to have to address it.
Not everything in the world is worth fixing.
Sometimes it is just better to document the procedure and move on.

[Norm38]

^^^ Yeah, they invented a new type of Diesel, but it's got a hell of a knock.