I’ve heard chatter that aside from the investigation into the RUD SpaceX is grounded until they can repair/rebuild LC 40. However SpaceX also has LC 39A. Work on the pad was completed earlier this year and it’s ready for Falcon 9 and Falcon Heavy launches. Is there a reason why they would not be able to use 39A once they are ready to resume launch operations?
So, Stranger - you got anything for us other than a hunch? 
Thanks as always for your informative and well-supported posts.
A lot of early reports conflated “fueling in preparation for static fire test” with “performing a static fire test”, so that’s where that impression came from. In practice, they would have had to fuel the vehicle full up even if they didn’t do any engine test. So there really won’t be any rational criticism of the test firing regime – launch prep never got that far, and the anomaly occurred during a necessary and inevitable process.
As long as the press gets it right that the static fire test wasn’t a factor at all. It never got a chance to become a factor.
Thanks for the correction.
At least all the nasty hydrazine got burnt up good, right??
I wonder what caused the large explosion around 4:00, I wouldn’t expect there would be much left of the rocket by then.
I certainly hope so. Hydrazine and its associated compounds, and nitrogen tetraoxide are very toxic. Even if they have fully combusted, the site it going to require substantial remediation as it is adjacent to Merritt Island National Wildlife Refuge.
My guess would be some kind of ground support equipment fuel reservoir (generator or chiller for the LOX).
I’m not going to speculate as to the cause (and I don’t have anything more than the same video to go from) but this is certainly a setback for SpaceX and the commercial space industry overall. However, I’ll also say that when you are handling this volume of combustable fluids (and especially cryogenic oxidizers, which pose particular hazards) and large structures, this kind of failure is always a risk that requires constant vigilance. Every single operator of large liquid propellant launch vehicles has experienced these kinds of failures.
The key to recovering from them and regaining confidence is a thorough root cause investigation and corrective action to reduce or eliminate design flaw or process errors which caused it. In the case of the Space Transportation System (‘Shuttle’) the process of making design and process improvements on such a flawed vehicle eventually ground the program to a near halt. SpaceX has shown more flexibility in terms of making rapid design changes, and more importantly, they instrument the hell out of everything, pasting on GoPro cameras like they are BeDazzling[SUP]TM[/SUP] the vehicle and support equipment.
So, hopefully SpaceX is able to quickly get to root cause and then move on to correct the issue. My concern is that this may impact their intention to use densified propellants, which is key to their “F9 Upgrade” configuration and achieving higher payload masses and orbital energies, which itself was key to getting to a critical portion of the large satellite market that F9v1.1 wasn’t able to service. I suspect that SpaceX is cutting their cost margins a lot closer than Musk and Shotwell would like to admit (and I’m still skeptical about the extent of cost savings on reuse of Stage 1), and while SpaceX has accomplished a number of impressive technical milestones at the end of the day they still need to turn a profit and demonstrate high reliability to their high payload value customers, especially for the EELV contract.
Stranger
To be clear, I wasn’t referring to anything past the first frame–that is, the first ~17 ms. Shortly after that, the tank has clearly ruptured and you get the “standard” deflagration as the LOX and RP1 in the upper stage mix. You can see the initial flare in the second image here.
However, upon reflection I do think you’re right. I had thought there was fire the entire length of the illuminated area, but there’s no reason to believe this–instead, I think a large portion of the “fireball” is really just clouds of water vapor illuminated by the, uh, energetic event. The actual amount of fire involved at that point might not be much.
Still no real news. Probably for the best that they aren’t relaying every stage of their internal investigation, even if it is frustrating to interested bystanders.
One interesting item of note is that there is a large panel (on the order of 1 m[SUP]2[/SUP] area) of some kind that is blown free of the initial explosion and only later starts to combust (presumably being saturated with liquid oxygen but ejected in advance of the initiation of combustion). This would seem to indicate some kind of rupture of a pressurized vessel preceding he conflagration. I have some private speculation as to what could have contributed to this but we’ll see what the investigation concludes.
One immediate impact from this has been a 41% decline in market valuation Spacecom, the operator of the AMOS-6 satellite. This will most likely result in a cancellation of the planned acquisition of the company by Xiunwei Technology Group, and a series of lawsuits. It is a good week to be a lawyer, and not anyone else involved in the launch.
Stranger
Its a lousy few weeks to be Elon Musk thats for sure.
I just cannot make up my mind about the guy. I love his visions and crazy ideas. But, some nagging part of me feels that he is a future perp walk candidate.
It’s not much, but Musk has tweeted a few more scraps:
Still working on the Falcon fireball investigation. Turning out to be the most difficult and complex failure we have ever had in 14 years.
There is in fact a (relatively) quiet pop sound a few seconds before the fireball, but doesn’t seem to be associated with any visible change.
As far as I know, SpaceX considers static fires to be essentially a launch rehearsal, and so would have all of their telemetry and other stuff going as they would with a real launch. So they should have plenty of data.
A bit of an update, and some good stuff coming up.
SpaceX released an update on their anomaly investigation, which narrows the problem to their cryogenic helium system. Although they weren’t more specific, it’s almost certainly a breach in the COPV, or “carbon overwrapped pressure vessel”, which is the tank that stores the pressurized helium, and is located inside the liquid oxygen tanks. COPVs are quite strong but store an immense amount of energy and a failure could easily cause the explosion.
In other news, SpaceX has released some sweet pictures of their new Raptor engine in action. Not clear yet how much of the engine is complete yet–it seems likely that it’s pressure-fed at this point, with the turbopumps to come later, but in any case a test firing is good to see. They claim 300 bar chamber pressure–that’s huge! 50% more than the Shuttle main engines, and 20% more than anything the Russians have in production.
And in other other news, SpaceX will livecast their Mars Architecture talk from the International Astronautical Congress at 1:30 CST tomorrow. Should be interesting! Hopefully will get to learn a lot more about Raptor, MCT, and the BFR.
This can’t be right!
SpaceX couldn’t have developed a methane engine!
According to ULA (Remember all those old rockets? ULA still makes 'em!), Blue Origin was to create the first methane engine - it even has a name! It’s called the “BE-4” and is the engine which will allow ULA to stop using those dirty Russian engines!
It says so right here: https://www.blueorigin.com/be4
SpaceX’s firing is just a bad dream - the Country will soon awaken!
My pre-coffee sarcasm detectors aren’t fully functioning yet. So I’m going to take usedtobe’s post at face value and blather a bit about the BE-4 and the Raptor.
First of all, that Raptor test firing was not a full-scale production engine. It’s a smaller scale prototype, likely developed with Air Force funding to develop a higher-performance upper stage engine. Conceivably, if SpaceX designed a new upper stage that used it, they could get a significant performance and capability boost for payloads going to high orbit. In any case, it’ll take a lot more work to develop the full-scale engine.
Second, while there haven’t been as much publicity, it seems that BE-4 development is nearly complete. Last I heard, ULA was waiting for a final full-scale engine test in the next few months, before committing to the BE-4 for the Vulcan rocket. Blue Origin also plans to use the BE-4 to power New Glenn, a new super-heavy lift vehicle with estimated payload capacity comparable to the Block 1 SLS. They’ve even started building the factory that will build this rocket.
Without getting into fanboy pissing contests about BO vs SpaceX, I’m just going to say that it’s amazing that there are two privately funded, ambitious, and credible new approaches for space exploration.
Some good stuff in the IAC preso. I’ll write up some stuff tonight unless someone beats me to it.
Some really impressive pics of a tank prototype. In fact this is more impressive to me than the Raptor shots; it’s hard to tell from those if the engine is 5% complete or 50% complete, but this is a seriously huge tank, and has apparently already survived their deep cryo testing.
300 tons to LEO fully reusable. Smart architecture, too. They recognize that the Mars component has a limited number of reuses just by virtue of the trip taking so long. The booster stage and refueling ships can be reused far more frequently and so the costs amortize to almost nothing.
They plan on landing the booster stage right back on the pad. Based on their proven accuracy for the barge landing, this seems… almost reasonable.
If you’re writing a substantive post, it’s probably worth starting a new thread. Someone else will at any rate, and it’d be kinda lame to refer them to the end of a thread where they’re not interested in the first couple hundred posts…
Yeah, this probably warrants a new thread. Ton of cool tidbits in the preso. Picked up a cold so it depends on how I’m feeling tonight, though…
I would caution against drawing the conclusion that the COPVs, which are just one type of component in a complex pressurized system for tank ullage. If the COPV was designed and qualified per the indusry standard (ANSI/AIAA-S-081A), the tank should fail in a non-catastrophic leak before burst (LBB) mode in the overwrapped section of the liner, and the neck portion should have sufficient margins that the overwrapped section fails first. Although an LBB could result in LOX tank overpressure and rupture if the leak were large enough to increase tank pressure beyond the capability of the relief valve, it should be gradual enough to be seen in telemetry (and in fact SpaceX has had a problem with COPVs leaking previously on the pad, although never catastrophically) and the kinetic energy imparted shouldn’t be enough of a mechanical shock by itself to rupture the LOX tank. Rupture of the pressurized lines (the plumbing that connects the tanks to a regulator or servovalve), or catastrophic failure of the regulator or valve could potentially create a tank rupture with sufficient mechanical shock to initiate combustion or detonation, and may not be readily observable in telemetry. Having to operate in a deep cryogenic environment complicates analysis of the system because it not only has to withstand flight acceleration and dynamic loads but also both thermal fatigue and coefficient of thermal expansion stresses. The COPV itself, it properly designed and autofrettaged, should be robust against thermal stresses as the overwrap keeps the liner in compression at all times, but evaluating thermal stresses on the overall plumbing and mounting structure can be quite challenging.
If the root cause or most likely cause is determined to be a COPV, that is potentially a significant problem, indicating either a problem with the design margins, manufacturing methods, and inspection processes. Since the Falcon 9 uses the same COPV in dozens of locations on the Stage 1 and Stage 2 vehicles, they would potentially have to change the design on hundreds of tanks in inventory or installed on in-process stages. COPVs are process intensive to build, requiring 100% inspection of welds on the liner/neck, carefully controlled winding of the overwrap, controlled autofrettage cycling to deform the liner to ensure compressive prestress even in the worst case conditions, and the post-autofrettage inspection to assure that uniform liner deformation and no signs of incipient cracking.
Stranger
Thanks for the correction, Stranger. My impression from elsewhere is that COPVs tend to only fail catastrophically, but I see that’s not true. Question: is there a significant difference here between winding techniques? Specifically, is a “braided” winding (using interleaved fibers as compared to large sections of linearly oriented ones) more resistant to catastrophic failure?
Autofrettage is a fun word. I went down the Wikipedia rabbit hole for some time when running across that word a few days ago.
Slight self-correction: the C in COPV is for “composite”, not “carbon”, though in this case we are talking about carbon fiber.
Wovern fabric–termed ‘broadcloth’–is used for large structural layups and conformal structures, or as the surface layers on a honeycomb panel, where it is laid down in mats, trimmed to size, and typically vacuum bagged or pressed in a mold to consolidate the the fibers and minimize voids or porosity in the matrix that can result in disbonds. Because the mats have non-joined edges (i.e. you can’t readily ‘weld’ the fibers together), they aren’t especially good at containing internal pressure without significant reinforcement.
Pressure vessels and overwraps are made by continuous winding of a roving (set of parallel tows) at a controlled helical angle. The structure may have local broadcloth patches (often termed “doilies”) for reinforcement around attachment points or apertures, and may have circumferential reinforcements in a cylinder section, but by controlling the spacing and wind angle can get a very optimized design. The art of continuous fiber winding is pretty mature (there are actually 6 and 7 axis CNC machines that can now wind significantly more complex shapes than cylinders and ellipsoids); the challenge of building COPVs is actually in the impermeable metallic liner, which has to be designed and pre-stressed to ensure that it shares the bulk of the pressure load evenly with the overwrap so that it doesn’t develop weak spots that can result in through cracks, or experience tensile stresses that can create low cycle fatigue damage.
COPVs are supposed to fail in leak before burst mode, which doesn’t mean that they always do, especially if difficult to inspect flaws or post-proof damage degrades the overwrap or the liner separates from the overwrap. Despite their price tag (a COPV the size of a basketball runs in the tens of thousands of dollars, and the ones used on the Shuttle program were hundreds of thousands of dollars apiece, plus the cost of regular inspections) there are only a handful of companies that make aerospace-grade COPVs because of the difficultly in producing COPVs reliably. On the aforementioned Shuttle, COPVs were a reliability driver insofar as a failure could both lease a critical system without fluid or fuel, and could pose a contamination hazard.
But there are other things that could have caused an overpressure failure in this case, so I would again caution against drawing conclusions without further evidence.
Stranger
Going from science to gossip, the Washington Post has a story today stoking the idea that ULA sabotaged the test:
And by “stoking,” I mean throwing the idea out there and then walking back from it. Like Trump talking about how nice it is that he didn’t say something mean.