SpaceX rocket explosion.

[john72]

See they are back to producing flopits again!

[janoc]

Brumby, very cool!

May I tweet that?

It does seem external to the vehicle!

That could very well be the spot where the various umbilicals carrying the fuel attach to the rocket. If there was a sudden leak, then all it takes is a spark from static electricity or because a piece of metal scraped some other metal and things go BOOM.

Musk said the failure occurred during a fuel transfer to the second stage

Wow, fueling up the second stage, for a static fire test of the first stage, sounds like being a bit OCD.

Or maybe the launch software only has one sequence, and that 2nd stage refueling is in there?

Well, I guess they need to tighten up that leaky pipe joint and start from the beginning again.

Could well be that in order to have the correct mass for the test fire the upper stage must be fueled too. Otherwise the rocket could be too light and the bolts/tethers would not be able to hold it down. But that is only my speculation.

[ChunkyPastaSauce]

Musk said the failure occurred during a fuel transfer to the second stage

Wow, fueling up the second stage, for a static fire test of the first stage, sounds like being a bit OCD.

Or maybe the launch software only has one sequence, and that 2nd stage refueling is in there?

Well, I guess they need to tighten up that leaky pipe joint and start from the beginning again.

Could well be that in order to have the correct mass for the test fire the upper stage must be fueled too. Otherwise the rocket could be too light and the bolts/tethers would not be able to hold it down. But that is only my speculation.

I'm sure there are many reasons.... one of them is to check for resonance issues... highly dependent on mass and position of masses...everything must be setup exactly how they would be during actual launch.... fuel, payload, and all.

[Alex Eisenhut]

Last seen at the scene:

[TerraHertz]

For better gloa... um, analysis, here's a slo-mo video.
http://9gag.com/gag/aGDGKE0

[EmmanuelFaure]

Jet fuel can't melt steel beams

[rob77]

Jet fuel can't melt steel beams

that puppy doesn't run on jet-fuel, it uses RP-1 for 1st stage - it's a slightly different stuff - but still can't melt steel beams !

[ajm8127]

That is one hell of an explosion. You can see the camera shake when the concussion wave of the main explosion hits it. Looks like the camera is about 2.5 miles (4 km) away.

The sound is obviously delayed quite a bit, but there seems to be a popping sounds about 4-5 seconds before the first sound of the explosion makes it to the microphone. It's at approximately 1:18 in the video.

[TerraHertz]

http://www.theregister.co.uk/2016/09/09/falcon_fireball_investigation_is_spacexs_biggest_challenge_yet_says_musk/

Elon Musk says SpaceX Falcon 9 fireball investigation is 'biggest challenge yet'
'Anomaly' wasn't rocket engine's fault, he claims
Musk tweet:

?@elonmusk
Important to note that this happened during a routine filling operation. Engines were not on and there was no apparent heat source.

Also: http://www.theregister.co.uk/2016/09/01/spacex_rocket_suffers_anomaly/

Edit to add:
Btw, I found http://keepvid.com/ works very well for downloading copies of youtube videos.
http://www.uslaunchreport.com/  are the people who recorded the hi-res video of the explosion, and that video is here:  youtube.com/  watch?v=_BgJEXQkjNQ

Using keepvid to download it, you can get the full 720p version. 49.5 MB. Then use VLC player or whatever to single-frame step through it. The explosion starts around 1:11.
Now the interesting thing is, that I'd seen some sites claiming there was an 'object' caught flying across the visual field for a few frames at the time of the explosion. I'd assumed it was either an insect near the camera, or arseholes tampering with the video and/or posting shopped stills.

Well guess what. In that original hi-res copy direct from www.uslaunchreport's youtube channel, there really is an 'object', and it's visible in 7 frames around the time of explosion start, traveling in a regular line, uniform and very fast speed from right to left in the frame. How interesting. Could still be a near insect I suppose, but that's an exceptional timing coincidence.

I've made a GIF animation of it, at http://everist.org/pics/misc/SpaceX_object.gif   900KB
(Animated GIFs don't seem to work via eevblog image attachments.)

Also note the object is traveling in a straight line before the segment when the explosion occurs, and a slightly different straight line in the rest of its path. The thought of recoil from something being fired from the object, altering the object's path, suggests itself.
And then there's this: http://www.whatdoesitmean.com/index2108.htm   (Yes, I know, one of the freakier sources. Still interesting background details.)

[Brumby]

Don't you just love a good UFO or conspiracy theory....

If you look more closely at the GIF frames, you will see that the flight path deviation of the object actually occurs after the explosion has begun.  Remember, the frames only capture a single point in time - and the exact instant the explosion commences occurs between two frames - in the interval before the flight path deviation.

As for what the object is, I have a very simple theory - which anybody who has watched a number of countdowns or outdoor video of the facilities will have noticed.... Birds.  I have often seen them circling in the sky - and it's not outside the realms of possibility that one flew across the scene going from point 'A' to point 'B'.

To prove my point, if you pick up the video at 1:40, you will see a large bird come in from the right, do a loop to the right of the engulfed tower and then continue off to the left.  It's speed is comparable to the 'object' under suspicion - and if you're not convinced it's a bird .... then tell me what this is:


Sorry ... but unless it was an eagle with an aerial bombing inclination, it has no relation to the explosion .... IMHO.

[TerraHertz]

Don't you just love a good UFO or conspiracy theory....

Ha ha, yes I do. Especially when the potentials include both of those (and birds, insects and tampering.) This is why I bothered to dl the original video and single step it. Because of some things I'd read. Expecting it to be rubbish.

If you look more closely at the GIF frames, you will see that the flight path deviation of the object actually occurs after the explosion has begun.  Remember, the frames only capture a single point in time - and the exact instant the explosion commences occurs between two frames - in the interval before the flight path deviation.

Oh really, do you think I don't know that? Obviously, the line of flight change, if it was sudden, may be anywhere in that segment marked 'Explosion'. I just drew the lines between the captured points so I could see any trend. I'm mostly surprised the camera used had such a high shutter speed. Framerate is normal, but the short shutter time to capture that movement without motion blur is impressive.

As for what the object is, I have a very simple theory - which anybody who has watched a number of countdowns or outdoor video of the facilities will have noticed.... Birds.  I have often seen them circling in the sky - and it's not outside the realms of possibility that one flew across the scene going from point 'A' to point 'B'.

To prove my point, if you pick up the video at 1:40, you will see a large bird come in from the right, do a loop to the right of the engulfed tower and then continue off to the left.  It's speed is comparable to the 'object' under suspicion - and if you're not convinced it's a bird .... then tell me what this is:

I'd noticed that bird. Demonstrating that the camera resolution is fine for resolving birds. Also there are a few things I think _are_ insects at other points in the video.

I'd expected the thing people were talking about would be a bird (distant) or insect (close to camera.) It could still be an insect. Bird is out, since the object shape is not at all birdlike, has no sign of flapping wings in any of the 7 frames. 'High speed vehicle' is also not ruled out, especially since there's no way to tell if it's closer or further than the rocket and towers.

Another thing I want to get around to, is timeshifting the audio to match the visuals. (ie remove sound travel delay.) To see if the 'pop sound' that's been mentioned as pre-explosion, coincides with the moment the 'object' passes. Supersonic shockwave? If it's coincident, the timing may give some indication of the object's range.

Speaking of sound, hats off to the camera man. I recall that youtube channel is operated by US army disabled vets (because they have the time and skills to do these long onsite video recordings, and need the money.) Whoever did that recording was very chill, and kept his mouth shut to not mess up the audio. You can hear a slight intake of breath when he sees the explosion start, but that's all.

Currently I'm looking for confirmation of that information about the Israeli company that owned the satellite, and the deal to sell it to a Chinese company, contingent on successful launch of the satellite. Because that really does make it politically contentious, ie motive.

Edit: Btw, anyone interested should dl & examine the original 720p video, not my anim-GIF. For one thing, the GIF has a quantized color table, so has lost some information.

[helius]

Another thing I want to get around to, is timeshifting the audio to match the visuals. (ie remove sound travel delay.) To see if the 'pop sound' that's been mentioned as pre-explosion, coincides with the moment the 'object' passes. Supersonic shockwave? If it's coincident, the timing may give some indication of the object's range.

Scott Manley has already done that, and analyzes the progress of the fire:

[TerraHertz]

Scott Manley has already done that, and analyzes the progress of the fire:

Nice video, but that's not quite what I meant. I want to do it precisely over just say 20 frames, with visual alignment of the sound envelope to the frame series. Use the explosion start (timing uncertainty being the inter-frame interval) for sync.
Then look at the sound envelope in the few frame intervals before the explosion. Might be subtle, or nothing.
I don't have the tools to do this yet, since I'm not normally interested in such things.

I'm keen to do it, because of another instance where an audio recording with a video DID contain something extremely significant. The video of Fukushima Unit 3 exploding. http://everist.org/Fukushima/20110314_Fuku3_explosion.flv
(And the significance: http://everist.org/Fukushima/20120430_Message_of_Fuku3.htm  )

In that one though, the key detail could be heard by ear (the 3 bangs), so I never did the resync. Still would like to though, so now I've two instances motivating me to find software tools/method to do the resync.

[Brumby]

Bird is out, since the object shape is not at all birdlike, has no sign of flapping wings in any of the 7 frames.

Bird is not out ... not in my book, anyway.

Gliding birds don't flap wings - and you won't see a recognisable bird shape from a side view without a lot better resolution.  I had to step through the frames to get the best shot of the bird that was banking.

A couple of other things fit with the object being a bird.  The first is the course.  It's the same as the one at 1:40, without having thrown in the loop.  The second is straight line velocity ... they are pretty much the same.


What would really be good is some closer imaging and/or some taken from another direction.

[ez24]

I think it was Dave flying by, saying "don't turn it on - take it apart"

[Towger]

I am sure there were high speed security cameras allover the site.  They are standard equipment (here anyway) on sites with large quantities of explosive material, such as explosive factories.  I don't see a NASA launchpad being any different.   Their purpose is to record any explosions fast enough to see clearly where it originates from.

[bitslice]

just in case this was missed:

"At the time of the loss, the launch vehicle was vertical and in the process of being fueled for the test. At this time, the data indicates the [explosion] originated around the upper stage liquid oxygen tank"

[Cyberdragon]

just in case this was missed:

"At the time of the loss, the launch vehicle was vertical and in the process of being fueled for the test. At this time, the data indicates the [explosion] originated around the upper stage liquid oxygen tank"

LO2 is non-flammable. There still had to be a fuel leak somewhere.

[Brumby]

just in case this was missed:

"At the time of the loss, the launch vehicle was vertical and in the process of being fueled for the test. At this time, the data indicates the [explosion] originated around the upper stage liquid oxygen tank"

LO2 is non-flammable. There still had to be a fuel leak somewhere.

True - but ANYTHING combustible in normal air will burn explosively in a pure O₂ environment.

Perhaps that provided the thin end of a very sharp wedge.


Besides, the statement cited refers to a location - it does not state that it was initially involved, even if there is that inference

[EEVblog]

[Tomorokoshi]

My pet theory:
There was a small leak in a fuel line. It vaporized or was atomized enough to fill a fair amount of the free space inside the craft. Because of the non-conductive fluid flow, a static charge was built up in either the fuel volume or on insulated surfaces. Eventually enough charge built up to cause an arc, igniting the fuel.

[TerraHertz]

Yes, that's a highly likely one. Though what is visible in the first frame of the explosion sure looks like it was centered in the space between the rocket and the tower.
I was wondering if somehow they may have managed to contaminate some fueling line with the other kind of fuel. Producing an initial explosion in one of the pipes between the tower and craft.

Another one - presumably while pumping liquid fuel and liquid O2 into the rocket tanks, there has to be some venting of gas from the empty tank space. Same as filling an LPG cylinder. Surely, they didn't just vent both O2 and fuel fumes into the air around the rocket? That's so stupid I can't imagine they could possibly do that. But how does it work? Are there bleed removal pipes to the tower?

Also, if the cause was some goof like forgetting to turn on a bleed extraction pump, would spaceX admit it? I wonder what the insurance and payload contracts say about screwups, as opposed to true accidents?

[ez24]

Helium - who would have guessed (and nobody did)

http://www.geekwire.com/2016/spacex-launch-pad-blast-helium/

And here I thought it was Dave flying over and yelling "take it apart".  Umm maybe Dave was taking helium, can you imagine what he would sound like ? 

[eevbstedt]

Thanks Dave!  Thunderfoot delivers the goods.

[HP-ILnerd]

In retrospect, a helium COPV fail would explain how this had such a percussive start.  I.e., it didn't appear to start with a little plume of fire somewhere which propagated, it just abruptly went FOOM.  A COPV bursting would have busted all sorts of stuff in one go, allowing the RP-1/LOX to mix quickly.

In happier (and exciting news) first pix of the SpaceX test-firing the Raptor rocket motor:
https://twitter.com/elonmusk/status/780275236922994688
and a pretty one with shock(Mach) diamonds:
https://twitter.com/elonmusk/status/780278836860628992