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

[HMXHMX]

- COPV's will be redesigned to be LOX hardened (option least adviced given that composites and LOX don't like each other very much)

There are other composites than just carbon based ones.
No idea whether bringing back old glass fiber composite technology back is easier than the other options but it's not like it has never been done before...

Specific tensile strength difference.  I.e.  a CF overwrap compared to a GF overwrap is significantly lighter for an equivalent burst performance.

Sometimes you just have to accept the mass hit and move on.

[SWGlassPit]

I'm only talking about a difference of tens of milliseconds (likely only two video frames).  The disturbance created by the breach is still going to have to propagate to whatever failure path created that cutoff.  It would only be perfectly simultaneous if the initial location of the breach coincided perfectly with either the boxes or the data/power lines.  Otherwise, it would take a small -- but finite -- amount of time for the failure to propagate far enough to kill the data stream.

Helium lines would run parallel to data and power harness in the cable tray on the outside of the vehicle.

Yes, but a breach anywhere else in the helium system, particularly in the interior of the lox tank, doesn't mean the cable trays are instantly demolished. At this time scale, very little is actually simultaneous. Even a COPV burst takes five to ten milliseconds from initial fiber failure to loss of tank integrity.

[Echium]

I should add also that the specific modulus for carbon fiber is way higher than for glass, so for an equivalent weight cylinder the fatigue performance is also much improved (in principle and in particular for metallic, Type 3, lined vessels).  However, I note that fatigue performance is not generally a key requirement for launch vehicle COPV space application. Maybe just three cycles (of which two are test) and then the articles life is over.  Typically Cryogenic COPV's are 150% proof tested for burst with (widely available) liquid Nitrogen, I wonder if this was the first time that this second stage Helium COPV had seen 4 deg K (ie Helium)?  It's a tough area for the Composite resin matrix at these extreme temperatures.  So brittle.

[Herb Schaltegger]

I should add also that the specific modulus for carbon fiber is way higher than for glass, so for an equivalent weight cylinder the fatigue performance is also much improved (in principle and in particular for metallic, Type 3, lined vessels).  However, I note that fatigue performance is not generally a key requirement for launch vehicle COPV space application. Maybe just three cycles (of which two are test) and then the articles life is over.

Except for SpaceX's application of them in Stage 1, where they may be expected to undergo dozens of cycles. One wonders if they are using the same technological tweaks - whatever those may be - on the Stage 2 COPVs for the same of commonality.

[mvpel]

The data obtained from the Phase I Study demonstrates conclusively that any composite material utilized in COPVs in a cryogenic/impact environment must be designed and qualified with data from representative environments. Typically aerospace designs are extremely weight critical and historically tend to ignore the degradation due to cryogenic environments that was demonstrated during Phase I contract. Based on the data collected during the Phase I STTR, it can be stated conclusively that the failure to recognize the degradation of mechanical properties at cryogenic temperature will almost certainly lead to reduced margins of safety if not worse consequences.

THE CRYOGENIC BONDING EVALUATION AT THE METALLIC-COMPOSITE INTERFACE OF A COMPOSITE OVERWRAPPED PRESSURE VESSEL WITH ADDITIONAL IMPACT INVESTIGATION - Eric A. Clark - Utah State University, 2009

[Echium]

I should add also that the specific modulus for carbon fiber is way higher than for glass, so for an equivalent weight cylinder the fatigue performance is also much improved (in principle and in particular for metallic, Type 3, lined vessels).  However, I note that fatigue performance is not generally a key requirement for launch vehicle COPV space application. Maybe just three cycles (of which two are test) and then the articles life is over.

Except for SpaceX's application of them in Stage 1, where they may be expected to undergo dozens of cycles. One wonders if they are using the same technological tweaks - whatever those may be - on the Stage 2 COPVs for the same of commonality.

Perhaps when they started winding / producing their own cylinders 'in-house', they chose to kick off with the smaller, second stage cylinders (easier to start learning to wind with) and have continued to purchase the larger first stage ones from the original external contractor? Perhaps a logical strategy to choose.  Urethane matrix for Stage 1  &  Epoxy for Stage 2 COPV's?

[john smith 19]

What are the odds of them pointing a tiger team at developing an autogenous pressurization system for the F9/FH LOX tanks? Just to retire the cLOX v He issue once and for all. Implement as "v1.2.1"

Maybe for Raptor based second stage...but unless you can figure out how to autogenous pressurize RP-1 you are going to need the He anyway.

If they need more reliable copv's it might be worth it to switch to a titanium liner for them in the short term. It would be expensive (which I'm guessing is a big part of the choice to use aluminum as the liner material), but they would get the first stage bottles back and it could hold them over until a raptor based autogenously pressed upper stage could be flown.

Not submerged in lox.  Ti is not compatible. Stainless would serve and is what the vendor uses for similar COPVs  sold to other customers.

I'm pretty sure the Saturn LOX tanks were Ti in LOX because their was a cost  reduction project to switch to SS "blown up" using LN2 (Autofrettage?) to put the walls in permanent compression. IIRC they were 1/18 the cost at the same weight.

It would probably depend upon whether Space X can conclusively determine the exact cause of the failure. If they can, then probably the COPVs can stay where they are. If, on the other hand, it’s a bit nebulous as to what happened or multiple issues are identified, then maybe a larger redesign is warranted — or needed to keep key customers happy — that includes relocating the COPVs. Time will tell.

I dunno. The engineer in me just doesn't like the idea of putting tanks pressurized to hundreds or thousands of pounds inside a tank that's built for 30.

Yeah, it was a cool idea but so were a lot of my FUBARs, too. Sometimes you just have to backtrack.

Historically when you put high pressure gas in a tank it gets hot. The colder  you can cool that PV the faster you can pump gas in and the more you can pump in. "But PV store huge energy, a rupture will rupture the LOX tank." you say. Indeed they do, which is why USAF safety regs measure them in terms of lbs of TNT equivalent.

But so what? PV ruptures --> mission failure. LOX tank ruptures --> mission failure.
While it's not failing you get more pressurized gas for less tank weight loaded faster.

I'll note that AFAIK no PV has failed in flight. They are usually reckoned one of the more reliable parts of a design.

But HMX points out if one did fail (especially COPV) it will release a lot of energy, quite likely some of the debris fast enough to cause a spark inside the LOX tank and pepper the walls with fragments, possibly moving fast enough to penetrate them.

What are the odds of them pointing a tiger team at developing an autogenous pressurization system for the F9/FH LOX tanks? Just to retire the cLOX v He issue once and for all. Implement as "v1.2.1"

Maybe for Raptor based second stage...but unless you can figure out how to autogenous pressurize RP-1 you are going to need the He anyway.

IIRC both stages of the Titan 2 did this. The trick is to burn the propellant either fuel rich or just on stociometric and use the reaction products to pressurize the tank. However unless you strip the water vapour you get ice in the tank. Worked fine for a 1 shot system, probably not what you want for a reusable system.

[Jim]

I think too much is focused on just the COPVs.   There are other components

[Jim]

IIRC both stages of the Titan 2 did this. The trick is to burn the propellant either fuel rich or just on stociometric and use the reaction products to pressurize the tank. However unless you strip the water vapour you get ice in the tank. Worked fine for a 1 shot system, probably not what you want for a reusable system.

That was only the fuel side, the ox side just heated the N2O4.

[Jim]

What are the odds of them pointing a tiger team at developing an autogenous pressurization system for the F9/FH LOX tanks? Just to retire the cLOX v He issue once and for all. Implement as "v1.2.1"

Would require an engine redesign and reduction of performance

The RP-1 COPV's are in the RP-1 tank

[Proponent]

Why bother putting helium tanks inside the RP-1 tank?  I thought the point of putting helium tanks inside the lox tank was to take advantage of the increased tensile strength afforded by the very low temperature of lox.  That wouldn't apply (much) to the RP-1 tank, even if it is chilled a bit.

[Jim]

Save space

[Kryten]

I'm pretty sure the Saturn LOX tanks were Ti in LOX because their was a cost  reduction project to switch to SS "blown up" using LN2 (Autofrettage?) to put the walls in permanent compression. IIRC they were 1/18 the cost at the same weight.

The Saturn LOX tanks were aluminium;

During the countdown, pressurization was supplied by a ground source, but during flight, a helium pressurant was supplied from elongated bottles stored, not on the fuel tank, but submerged in the liquid oxygen (LOX) tank. In this medium, the liquid helium in the bottles was in a much more compatible environment, because the cold temperature of the liquid helium containers could have frozen the RP-1 fuel. There were additional advantages to their location in the colder LOX tank. Immersed in liquid oxygen, the cryogenic effect on the aluminum bottles allowed them to be charged to higher pressures.

http://history.nasa.gov/SP-4206/ch7.htm

[jbbo78]

There has been mention of incompatibility between the composite material and subcooled LOX, embrittlement and/or delamination of the overwrap in subcooled LOX, thermal contraction differences between the composite and the metal liner in subcooled LOX, and even the struts having different strength characteristics in subcooled LOX.  These ideas all center around SpaceX making a quick decision to subcool the LOX without considering known (if not obvious) failure modes for COPVs.

IANARS and I don't play one on TV, and yet all of these things popped into my head at some point in the last month.  I would be amazed if the good folks at SX hadn't considered and accounted for all of these things in some way.  Even if they somehow had not, I doubt that the investigation would be so "complex" and leave them scratching their heads if the cause was any of the above especially with all the debris they have on hand to analyze.  I expect the cause will be something far less obvious or predictable.

There's lots of plumbing related to the high pressure He both inside outside and all of it is at the same or probably higher pressure than the COPVs.  I'm sure there are all kinds of failure modes throughout the system that don't involve a COPVs directly failing but would end up being be just as catastrophic.  Given the over-engineering of parts in rocketry I expect that the failure will be found to be due to an unexpectedly high pressure wave somewhere or possibly even a much-weaker-than-expected part (similar to the previous strut manufacturing issue).  I just don't see this being because of subcooled LOX directly causing something to fail due to the temperature effects that they somehow forgot to consider.

[Kaputnik]

Save space

Seems a bit inefficient to save space by displacing propellant, no?

[Proponent]

That was my first thought on reading Jim's answer.  But suppose you do mount the tanks outside the RP-1 tank.  Now then the fuel tanks can be a little bit smaller, but the body of the rocket will have be to longer.  And since the He tanks won't fit snugly together, there will be some wasted space, meaning that the increase in the volume of the rocket body will be greater than the decrease in volume of the RP-1 tank.   And this little bit of extra rocket body, unlike the the fuel tank, is not pressure supported, it will likely be heavier than the extra bit of RP-1 tank.

I'm assuming, of course, that for some reason you can't fit the He tanks into some space that's wasted anyway, like the interstage.

In principle another advantage of putting the He tanks within the prop tanks is that the pressurization of the prop tanks reduces the necessary strength of the He tanks.  In this case, though, with the He tanks being at 5500 psi and the prop tanks at just 40 or 50 psi, that benefit is going to be pretty small and maybe not with the additional operation complexity it imposes.

[cscott]

A large breach in the cryogenic helium system, sufficient to rupture the second stage in less than 1/10th of a second means one thing, and one thing only IMO: a COPV blew spectacularly.

Given that we know that SpaceX has had prior problems with LOX immersed COPV's (other than the CRS-7 incident), I think it is a safe bet that either one of the below design changes will be implemented:

- COPV's will be moved out of the LOX tank, possibly into an interstage-like extension of the upper stage
- COPV's will be redesigned to be LOX hardened (option least adviced given that composites and LOX don't like each other very much)
- COPV's will be replaced by different tankage. For instance all-metal tanks.

Anyone of those changes is IMO going to eat into F9's performance.

I'll take that bet.  I'm equally confident they will fix what they've got, not throw it out and try something completely different and lower-performance.

[Herb Schaltegger]

Save space

Seems a bit inefficient to save space by displacing propellant, no?

Think of the alternatives, though. The interstage is full (MVac nozzle, stage latching and separation mechanisms, grid fin hydraulics and avionics for Stage 1). Below the RP1 tank is the octoweb, engines and gimbaling systems, and that area is exposed to very large thermal loads from launch, reentry and landing.

EDIT: The above is stage 1, obviously. For stage 2, there is no interstage space to speak of, and below the tank is the MVac; concerns there are probably thermal.

[NaN]

(trimmed)

IANARS and I don't play one on TV, and yet all of these things popped into my head at some point in the last month.  I would be amazed if the good folks at SX hadn't considered and accounted for all of these things in some way.  Even if they somehow had not, I doubt that the investigation would be so "complex" and leave them scratching their heads if the cause was any of the above especially with all the debris they have on hand to analyze.  I expect the cause will be something far less obvious or predictable.

Welcome to the forum! You've identified one of the basic problems with much of the speculation which goes on in these threads. SpaceX has far, far more context than must of us do, both into the systems (rocket, ground, testing, etc) and the data for this specific event. Theories that appear excitingly dangerous to some - to pick at random, the idea of the entire stack swaying in the wind incurring a resonance that causes structural failure - is extraordinarily obvious to the people who actually built the thing. To pick something less obvious like the COPVs, we have little idea of the analyzing and testing regimen SpaceX has unit-tested them with. They would obviously not just plunk the things in subcooled LOX and hope for the best.  Perhaps the COPV are, by this point, far more bulletproof and well validated than people are crediting. How would we know?

There is always the chance someone will get lucky and correctly identify the root failure, but more likely that it will be something new and unexpected. That's ok, the discussion is quite interesting on its own merits, but people just should not get attached to any one theory. IMO.

[Space OurSoul]

What about moving the common bulkhead up by the volume of the COPVs in the LOX tank and putting all the COPVs in the RP1?