Investigating the SpaceX Rocket Explosion of September 1, 2016

[Jonathan Scott]

I've viewed their earlier stuff and I'm very sceptical. It is certainly true that it seems a bit risky having a lot of pure oxygen around flammable materials, but the time factors don't seem consistent with that.

The amount of material which catches fire in the first frame and the implied speed of the flame suggests a significant amount of LOX plus some fuel was already mixed. I still think it's more likely that some major internal failure within the helium high pressure system caused an external rupture around the common bulkhead area and that the ejected LOX and fuel then caught light.

 

[jim hardy]

I've viewed their earlier stuff and I'm very sceptical.

seconded.

I'm betting on a weld failure. Metallurgy around welds in aluminum is troublesome. Still haven't heard what alloy COPV is made from.
https://app.aws.org/wj/supplement/WJ_1987_03_s73.pdf
Is it true they used a colder LOX on this rocket, 66K ? About 20 degrees colder ?

 

[Dotini]

Seedy conspiracy theories are tossed into the case by the Washington Post:
https://www.washingtonpost.com/amph...b60514-874c-11e6-a3ef-f35afb41797f_story.html

At a conference in Mexico earlier this week, Musk said that finding out what went wrong is the company’s “absolute top priority,” but he said what caused the explosion is still unknown.

“We’ve eliminated all of the obvious possibilities for what occurred there,” he said. “So what remains are the less probable answers.”

 

[jim hardy]

Two questions are nagging at my alleged vrain oops my alleged brain

First one stems from high school chemistry

Inside that Helium tank...
pv = nrt

Pcold X Vcold = nr Tcold
Phot X Vhot = nr X Thot

Pcold = Phot X Vhot/Vcold X Tcold/Thot

if that aluminum data is right, it shrinks 20% from room temp to ~ -200c
then vcold/vhot = around 0.8³ = 0.512, call it 1/2

Pcold = Phot X 2 X Tcold/Tot , since Tcold is maybe 70K and Thot maybe 300K, Pcold = Phot X 2 X 70/300 = 0.45 Phot
the tank should lose pressure as it cools because the gas contracts more than the aluminum does.

Okay that was one question i had. How does tank respond to cooling, does its stress increase or decrease and it looks like it decreases overall.Next question is one of heat transfer ,
Does the helium inside the tank cool quickly ?
or is there a time when the aluminum skin is stretched tighter around the helium while cooling progresses inward through the gas as if through layers of an onion ?
That takes time and pressure won't fall so quickly as if cooling of the gas were immediate..
Seems the tank's skin would be sensitive to rate of cooling because the tension in it depends on its own shrinkage versus that of the gas it surrounds.
Skin gets cold first.

Surely they've calculated that out.
I assume LOX tank fill process include measurement of helium tank pressure ?

just rambling, sometimes such musings help one along in troubleshooting.

 

[nsaspook]

Seedy conspiracy theories are tossed into the case by the Washington Post:
https://www.washingtonpost.com/amph...b60514-874c-11e6-a3ef-f35afb41797f_story.html

At a conference in Mexico earlier this week, Musk said that finding out what went wrong is the company’s “absolute top priority,” but he said what caused the explosion is still unknown.

“We’ve eliminated all of the obvious possibilities for what occurred there,” he said. “So what remains are the less probable answers.”

I liked the UFO one better than the 'Grassy Knoll' gunman.

 

[Dotini]

I liked the UFO one better than the 'Grassy Knoll' gunman.

I wasn't aware of any alien conspiracy - what has Mr Musk done to offend them? Remember, those are birds and bugs flying over the rocket. But Mr Musk has real rivals, critics and potentially enemies in a wide variety of industries, government agencies and companies right here on Earth. If in fact there was a conspiracy against the AMOS-6 mission, then not only Musk, but Zuckerberg and the state of Israel were also victims. And now they're your mortal enemies if you are the guy with the laser or rifle on top of the ULA building. For now, better to blame the shoddy, LOX-infused strongback pipe insulation and a stray spark, or some other one-in-million metal failure, despite the amusement value inherent in conspiracies.

 

[Dotini]

More video and analysis of the strongback piping arrangement from TechX.

 

[Jonathan Scott]

More video and analysis from TechX.

I'll admit they do have some plausibility, if their technical facts are correct. It appears (correct me if I'm wrong) that they are saying that during the LOX loading, the outsides of the insulated pipes could get cold enough for some of the venting cold oxygen to recondense to liquid and get absorbed into the insulation. Presumably this absorbed liquid would then decrease the effectiveness of the insulation and cause the outside to get even colder, amplifying the effect and eventually creating a significant amount of a known explosive mixture which could be set off very easily.

I previously found their explanation implausible because of the size of the initial explosion, but if it is possible that a large volume of the insulating material had absorbed liquid oxygen before the explosion then that could fit the effects.

I'd agree that the first flash and the fact that it did not expand significantly after that could be explained as an external explosion primarily involving a solid form of fuel (insulation with absorbed liquid oxygen), in which case the next step would be caused by the shockwave from the initial explosion damaging the second stage and breaking the helium pressure system, leading to total destruction.

 

[Jonathan Scott]

This was the earlier TechX calculation of the potential explosive yield from polyurethane foam insulation soaked in condensed LOX:

 

[mfb]

We believe that the composite over wrapped pressure vessel [the helium bottle], known as a COPv, let go in the tank. What caused it, the exact reason it let go, we’re still investigating. I don’t believe it was a ground system cause, but we’re still looking at the data.
[...]
The more than likely — the overwhelmingly likely — explanation is that we did something to that rocket. And we’re going to find it and we’re going to fix it.

Source

If it is some problem of the procedure, it can be very easy to fix - load something slower/faster, adjust some lines of code to keep something at a different temperature or something like that.

 

[Jonathan Scott]

From the look of the initial explosion, I still think that it happened after LOX and fuel were ejected by a COPV failure. That's what I've been saying since I first saw the video, and that would be consistent with what SpaceX is saying.

"TechX" (whoever that is) keeps posting these videos blaming an external explosion in the LOX pipe insulation. Although their latest version of that is more plausible than the initial suggestions, I personally think that would give a different appearance to the initial explosion. I don't know whether SpaceX uses polyurethane foam insulation as suggested nor whether it has taken care to seal the outside of the insulation sufficiently to prevent any significant absorption of LOX (given that low temperatures can degrade the foam structure), so I can't judge the plausibility of various technical aspects.

Does anyone know who this "TechX" is and what they are trying to achieve? Do they by any chance sell non-organic insulated pipework for cryogenic products?

 

[mfb]

Looks like TechX makes physics simulation software.

 

[Jonathan Scott]

Looks like TechX makes physics simulation software.

Is that definitely the same TechX as the Youtube stuff?

 

[mfb]

Hmm, it matched the topic, but then it would be odd to use WolframAlpha.

 

[jim hardy]

Have i already asked this ? When does the helium bottle get filled ?
Is it really at only 4 degrees K ? source http://newatlas.com/spacex-falcon-9-explosion-helium/45594/

 

[mfb]

As far as I remember the temperature was higher. The bottles are under high pressure and you want evaporation while releasing helium - subcooling the helium does not make sense.

 

[Jonathan Scott]

There's a new update on SpaceX news: http://www.spacex.com/news/2016/09/01/anomaly-updates

Here's the most interesting bit:

The investigation team has made significant progress on the fault tree. Previously, we announced the investigation was focusing on a breach in the cryogenic helium system of the second stage liquid oxygen tank. The root cause of the breach has not yet been confirmed, but attention has continued to narrow to one of the three composite overwrapped pressure vessels (COPVs) inside the LOX tank. Through extensive testing in Texas, SpaceX has shown that it can re-create a COPV failure entirely through helium loading conditions. These conditions are mainly affected by the temperature and pressure of the helium being loaded.

 

[jim hardy]

SpaceX has shown that it can re-create a COPV failure entirely through helium loading conditions. These conditions are mainly affected by the temperature and pressure of the helium being loaded.

Oh really ?

Hmmmm Helium loading not Lox loading like i'd thought. So much for post #144 https://www.physicsforums.com/threads/spacex-needs-us.884983/page-8#post-5581600

When do they fill the helium tanks? Are they immersed in LOX when filled ? Or does LOX loading come after helium ?

 

[Jonathan Scott]

When do they fill the helium tanks? Are they immersed in LOX when filled ? Or does LOX loading come after helium ?

I think helium loading starts before LOX loading is complete. The COPVs are immersed in LOX. There was more on the Nasa forum and elsewhere about it. The timeline of a previous launch has been published, but in this case they were using a modified timeline with very late loading of super-cold LOX and helium in order to get the maximum amount in and minimize the warm-up before launch. It is thought that this modified timeline could be a major factor in the anomaly.
The extra-cold LOX could perhaps have meant that LOX in the carbon fibre layers surrounding the COPV liners was cold enough to solidify when the liner was being filled with helium. There are theories that the embedded solid oxygen could have damaged the carbon during thermal contraction or even locally somehow got under such extreme pressure that it reacted with the carbon and possibly even ignited it locally. (I don't understand the mechanism of how that could happen).

 

[jim hardy]

The extra-cold LOX could perhaps have meant that LOX in the carbon fibre layers surrounding the COPV liners was cold enough to solidify when the liner was being filled with helium.

Thanks ! I hadn't thought of LOX freezing solid.

Oxygen freezes around 54K ? Helium liquefies around 3K ? Sure, it'd freeze on that COPV just like the frost on a good Gin&Tonic.

http://www.tvu.com/PShearStrSO2web.html
Experiments have been performed to test the shear strength of solid oxygen. Shear strength measurements were made in a cryostat by pulling a rod out of solidified liquid oxygen. The approximate shear strength of solid oxygen was measured as a function of temperature, increasing from 0.31 MPa at 45 K, to 4.46 MPa at 18 K. Solid oxygen was found to undergo plastic deformation at high temperatures, becoming increasingly strong and brittle as its temperature is decreased. Data and simple experiments confirmed a similarity of engineering material properties between solid oxygen and room temperature plastics.

I wonder if it expands or contracts on freezing ...

Sounds like they're getting close...
It's the small things of the Earth that confound the mighty.

 

[Jonathan Scott]

Oxygen freezes around 54K ? Helium liquefies around 3K ? Sure, it'd freeze on that COPV just like the frost on a good Gin&Tonic.

I don't think they use liquid helium, just helium gas cooled to get more in for the same pressure.

 

[mfb]

They don't fill in liquid helium, but it is possible that they overestimated the temperature somewhere.

 

[jim hardy]

They don't fill in liquid helium, but it is possible that they overestimated the temperature somewhere.

Ahhh thank you. As I've said often, I'm a plodder.

So if liquid oxygen at 1 atmosphere is 90 K
and they subcool it to 66K to get more density
and it surrounds the helium tank
and the helium approximates ideal gas , no phase change,

as LOX warms from 66 toward 90 K
helium tank pressure follows , from whatever pressure it had at 66K toward 90/66 = 1.36 X that pressure .

And it would be easy enough to get too much helium in there while it's cold. Especially if a temperature measurement is slow so that you think it's warming more slowly that actual and overestimate your 'headroom'.

Got it i think,
could be something as simple as sensor response time.

it'll be interesting to find what they meant by '...helium loading conditions' .

Two sayings from power plant: "Temperature taketh time" "Haste maketh waste"
thanks. old jim

 

[nikkkom]

Helium loading conditions might refer to helium ending up hotter than expected/designed for. When hot COPVs get splashed with cold LOX, composite overwrap could develop delaminations, which reduce strength and allow LOX ingress.

 

[liometopum]

Elon Musk says SpaceX finally knows what caused the latest rocket failure
http://www.theverge.com/2016/11/5/13533900/elon-musk-spacex-falcon-9-failure-cause-solved

Elon Musk Offers Icy Explanation For Spectacular SpaceX Falcon 9 Explosion
http://hothardware.com/news/musk-offers-explanation-for-falcon-9-explosion#GiSdfoWEJeV3dk7W.99

Elon Musk Says SpaceX Rocket Launches Might Resume Next Month
http://www.nytimes.com/2016/11/05/science/elon-musk-spacex-rocket-launches.htmlCould financial pressure have been a factor, as in this press release?:
Elon Musk’s SpaceX May Lose Inmarsat Launch Order
http://www.wsj.com/articles/elon-musks-spacex-may-lose-inmarsat-launch-order-1478165008

 

[jim hardy]

hmmm. they're dancing around mention of liquid helium...

from second link

According to Musk, the problem “basically involves liquid helium, advanced carbon fiber composites, and solid oxygen. Oxygen so cold that it actually enters solid phase.”

 

[Jonathan Scott]

hmmm. they're dancing around mention of liquid helium...

from second link

Yes, I think that under the pressure involved the helium is liquid when at full pressure, but I'm fairly sure it's nowhere near cold enough to be liquid at atmospheric pressure (that would be around 4K, but oxygen solidifies at around 54K). It does however seem that the helium was being loaded at below the freezing point of oxygen, because they think the failure involved liquid oxygen getting inside the composite wrap layers (presumably because of thermal contraction effects) and forming solid oxygen which can react with the carbon. They don't seem to give much away about the actual details.

 

[.Scott]

Additional details are provided in the NASA Space Flight forum (http://forum.nasaspaceflight.com/index.php?topic=41252.msg1606656#msg1606656).
LOX that manages to get between the metal helium tank and the composite overwrap flows back out as the tank is pressurized. But solid oxygen will not. This results in pure solid oxygen being forced against the carbon in the composite overwrap - apparently a condition that supports spontaneous combustion.

 

[Jonathan Scott]

Yes, I think that under the pressure involved the helium is liquid when at full pressure, but I'm fairly sure it's nowhere near cold enough to be liquid at atmospheric pressure (that would be around 4K, but oxygen solidifies at around 54K).

After catching up on the NASA Space Flight forum, I see that this has recently been discussed in there. As the critical temperature of helium is around 5K and they don't think SpaceX would be going anything like that cold, they say that the highly pressurized helium gas technically forms a supercritical fluid rather than a liquid, but Elon Musk has been referring to it as "liquid helium" anyway.

 

[jim hardy]

Thanks !

I guess if it's in there as a compressed liquid and gets warmed up, its pressure will follow saturation curve ?
i looked for a curve for helium but the ones i found didn't go to temperature so high as that of LOX at 1 atm .
Helium atoms being not sticky seems they'd act like an ideal gas very difficult to liquefy, that's why it condenses at only 5 degrees absolute? Somewhat higher temperature at pressure but still pretty doggone cold ?

Elon Musk has been referring to it as "liquid helium" anyway.

Maybe "Liquefied" would be a better term?

I just can't get past the thought of overpressurizing that helium tank by warming it with LOX.

A straightforward explanation will come out sooner or later .

old jim

 

[.Scott]

I guess if it's in there as a compressed liquid and gets warmed up, its pressure will follow saturation curve ?
i looked for a curve for helium but the ones i found didn't go to temperature so high as that of LOX at 1 atm .
Helium atoms being not sticky seems they'd act like an ideal gas very difficult to liquefy, that's why it condenses at only 5 degrees absolute? Somewhat higher temperature at pressure but still pretty doggone cold ?


Maybe "Liquefied" would be a better term?

I just can't get past the thought of overpressurizing that helium tank by warming it with LOX.

A straightforward explanation will come out sooner or later .

old jim

Here's a chart posted in that other forum:

LOX at 1 atmosphere will freeze at 54.36K; higher at higher pressures.

 

[Greg Bernhardt]

Elon Musk’s SpaceX says it has found cause of rocket explosion and plans to fly again Sunday
https://www.washingtonpost.com/news...ocket-explosion-and-plans-to-fly-again-sunday

 

[mfb]

http://www.spacex.com/news/2016/09/01/anomaly-updates.

Liquid or solid oxygen that went under parts of the helium containers (COPV) and lead to an ignition of the carbon fibers.

They'll go back to an older fueling procedure to avoid that, as long-term plan they want to redesign the COPV.

It will be a busy year for SpaceX, and we'll see how many things of their ToDo-list they can get:
In addition to returning to flights, SpaceX wants to launch one of the landed boosters soon (~February-April?), and the maiden flight of Falcon Heavy is planned for 2017 as well. It has 3 cores, one of them could also be a landed booster.
They want to make one final upgrade to the booster (the last update was the "full thrust" version) to get a bit more performance out of it, and to make re-use easier. The fairing that protects the payload is currently thrown away - a few million dollars per launch because it needs huge expensive oven to cure. They plan to recover it with parachutes.
Dragon V2, the manned version of the Dragon capsule, could have its maiden flight in 2017, and a manned flight in 2018.

 

[mheslep]

A novel aspect of the SpaceX launch procedures included fueling close to launch, made necessary (in part?) because of the super cooled liquid design SpaceX chose. Manned flights in the past were fueled prior to crew load. The accident, occurring during fueling, drew further attention to SpaceX procedures. I'm curious how SpaceX intends to address the fueling sequence as they proceed into manned flight.

October:

...Earlier this month, Gwynne Shotwell, the company’s president and chief operating officer, said investigators believed the cause likely was an operational issue, versus a design or manufacturing problem. One of the biggest questions, according to industry officials, is how the helium tank interacts with the surrounding supercooled liquid oxygen. The process is unfamiliar to most of the industry because such a supercooled oxidizer isn’t typically used on big rockets...

http://www.wsj.com/articles/spacex-probe-into-blast-focuses-on-fueling-issues-1477042203

From the WaPo article today:

The company said that in the short term, it plans to change the way it loads fuel. Eventually, it plans to change the design of the pressure vessels to prevent buckling.

Change fuel loading how? Fuel early? Time would be a factor with supercooled O2.

From the SpaceX link today:

...The corrective actions address all credible causes and focus on changes which avoid the conditions that led to these credible causes. In the short term, this entails changing the COPV configuration to allow warmer temperature helium to be loaded, as well as returning helium loading operations to a prior flight proven configuration based on operations used in over 700 successful COPV loads. In the long term, SpaceX will implement design changes to the COPVs to prevent buckles altogether, which will allow for faster loading operations.

The new COPV 'configuration' is unspecified, as are the 'prior' 'helium loading operations'.

 

[Q_Goest]

I don't know where this guy got his info but it appears to jive with what I've seen published from SpaceX, plus the explanation and conclusion reached seems believable.