Author Topic: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD (1)  (Read 1190893 times)

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 561
  • Liked: 414
  • Likes Given: 61
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #600 on: 09/01/2016 08:57 PM »
Accidentally (or deliberately) activated FTS would have activated the FTS for the entire vehicle, not just the second stage.  This failure was too localized to be FTS.
That exactly looks like what happened. The whole stack just exploded really fast.

It really didn't.  There was a localized explosion in the second stage with a relatively slow progressive failure down the stack.  If it was FTS, the whole vehicle would have ceased to exist in the span of two or three frames.

FTS isn't designed to obliterate the vehicle as much as it is to release the fuel to get rid of its mass, momentum, and explosive potential. When we've seen FTS before it causes a slow conflagration fireball just like we see here after that initial pop at the top of S2. It looked very much like the fireball we saw on GH2, although with a much better vantage point to see details.

For those saying FTS wouldn't cause a fireball, well, what other ignition source was there? We didn't see a green flash.

FTS is designed to cease acceleration and disperse the propellants.  This is often accomplished by a linear shaped charge running the length of the vehicle.  I suspect the Falcon is the same way.  You might not see an immediate fireball, but you would see the whole vehicle unzip basically at once if it were FTS, not the several seconds it took to completely come apart.

Offline DarkenedOne

  • Full Member
  • ****
  • Posts: 954
  • Liked: 58
  • Likes Given: 8
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #601 on: 09/01/2016 08:59 PM »
The vehicle is sitting in a cloud of concentrated, evaporating oxygen and is holding thousands of kilograms more on board. A small leak and a spark is all that is needed to cause an event big enough to rupture the tanks leading the chain reaction. Oxygen is abundant in this scenario - we don't need an explosive to turn the situation bad.

Oxygen alone will not cause an explosion.  It does not react with itself.  You need a fuel source. 

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 561
  • Liked: 414
  • Likes Given: 61
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #602 on: 09/01/2016 09:00 PM »
The vehicle is sitting in a cloud of concentrated, evaporating oxygen and is holding thousands of kilograms more on board. A small leak and a spark is all that is needed to cause an event big enough to rupture the tanks leading the chain reaction. Oxygen is abundant in this scenario - we don't need an explosive to turn the situation bad.

Oxygen alone will not cause an explosion.  It does not react with itself.  You need a fuel source. 
In a sufficiently high concentration of oxygen, many traditionally nonflammable things become quite flammable.

Offline Kabloona

  • Senior Member
  • *****
  • Posts: 4334
  • Velocitas Eradico
  • Fortress of Solitude
  • Liked: 2563
  • Likes Given: 531
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #603 on: 09/01/2016 09:02 PM »
I heard from one of the reddit that some believe a hydrazine leak from the satellite may of caused the initial explosion.  Given the location of the explosion being close the top of the oxygen tank, and the initial size explosion it seems like this is certainly possible.  I know from chemistry that hydrazine is particularly reactive, volatile, and very dangerous.  I am not an expert on launch systems and satellites, but what is the likelihood that this was the case.

Probably near zero. The payload was well above the plane where the first flash occurred, encapsulated in a fairing that remained intact long after the initial explosion. Also there's no way for hydrazine to get *into* the LOX tank, which appeared to rupture, suggesting the event may have started inside the LOX tank.
« Last Edit: 09/01/2016 09:04 PM by Kabloona »

Offline vandersons

  • Member
  • Posts: 84
  • Ireland
  • Liked: 63
  • Likes Given: 122
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #604 on: 09/01/2016 09:04 PM »
Have watched the explosion frame by frame now for at least a couple dozen times. The initial bright flash is very baffling. It comes from the area where the LOX appears to be venting so there is lots of O2 there but on the other hand there is also a lot of wind blowing the vapour away so not much could have been just sitting there waiting on a spark to ignite stuff.

With that in mind, what is in that area of the bright flash that could potentially be oxidised in such a flash? From reading up thread the RP1 and LOX filling lines are further down near the interstage, the flash occurred much higher up (that would exclude RP1 leaking). Are any of the fittings near the flash area in any way flammable enough to produce a flash like that given enough O2 was present?

Could a static spark be strong enough to make a hole in the LOX tank and igniting the cork and/or aluminium therefore producing the first bright flash that with additional LOX gushing out of the tank gets bigger very quickly until it ruptures both tanks on S2?

Surprisingly S1 holds out for quite a while. Only until after it has been completely engulfed by the falling flaming fuel and LOX mix it gives up an produces the second big explosion (the third being the payload hitting the ground and blowing its hypergolic's tanks).

The COPV idea kind of doesn't quite add up in my mind. If only the LOX tank would go pop from an overpressure event then the first thing we should be seeing in the sequence is a white cloudy mass of LOX blowing out in various directions but no flames until the cloud hits something combustible and hot enough to start the explosion. It should look similar to how the S2 deflagrated on CRS7 - first a puff of white cloud, then mixing with some RP1, then hitting the exhaust flames and going up in a big fireball. In this case the sequence is very different - very bright flash, fireball starts spreading, some white cloud starts appearing leading the fireball, fuel-air explosion happens that triggers the S1 to blow up as well.

Offline RoboGoofers

  • Full Member
  • ***
  • Posts: 333
  • NJ
  • Liked: 168
  • Likes Given: 244
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #605 on: 09/01/2016 09:07 PM »
The vehicle is sitting in a cloud of concentrated, evaporating oxygen and is holding thousands of kilograms more on board. A small leak and a spark is all that is needed to cause an event big enough to rupture the tanks leading the chain reaction. Oxygen is abundant in this scenario - we don't need an explosive to turn the situation bad.

someone posted this upthread, but lot of fuel and LOx mixing without an explosion: Atlas 190D



Quote
13,000 gallons of liquid oxygen and a full load of fuel sloshed over the stand and the nearby terrain.

here's a reference:
https://books.google.com/books?id=OVNuxBlXFHYC&lpg=PA32&ots=RWbZyy0by-&dq=1963%2C%20Atlas%20190D&pg=PA32#v=onepage&q=1963,%20Atlas%20190D&f=false

Offline iamlucky13

  • Full Member
  • ****
  • Posts: 1660
  • Liked: 102
  • Likes Given: 93
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #606 on: 09/01/2016 09:07 PM »
Wonder how the Dragon2 Trunk/Capsule would have dealt with this 2nd Stage RUD in a Pad Abort scenario?

Assuming the problem was recognized as severe by the computer and the LAS activated almost instantly, a second later the capsule would be 50+ feet away, and by the next second, roughly 300 feet away, etc.

If you pay attention in the video, after the explosion starts and the main fireball erupts and fades away, the fairing, halves still latched together, is visible falling starting about 7 seconds after the start. Given the smaller cross section of the Dragon 2, it's heat shielding, and potentially also stronger overall construction of the capsule, I think the crew has good odds in this kind of scenario. This is, after all, one of the design scenarios for the LAS.

Offline mlindner

  • Software Engineer
  • Senior Member
  • *****
  • Posts: 2090
  • Space Capitalist
  • Silicon Valley, CA -- previously in Ann Arbor, MI
  • Liked: 748
  • Likes Given: 289
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #607 on: 09/01/2016 09:07 PM »
Check the audio, there's a small pop about 5 seconds before the sound from the actual explosions arrive.

I think that's just someone messing around a car or something near the camera.

Interestingly enough, the small first bang heard is the original S2 explosion. The big bang 3 seconds later is the fuel-air type explosion when that fireball hit the ground and mixed up the RP-1 and LOX.

I hear three distinct "initial" sounds:

1:16--a very faint "plonk" like someone hitting a PVC pipe with a hammer in the distance...COPV rupture?
1:18--a faint pop/bang... S2 tank rupture?
1:23--very loud boom...the LOX/RP-1 explosion?

That first sound at 1:16 is unusual and seems to me like it may be the initiating event. A COPV or high pressure line/fitting letting go followed by S2 tank rupture 1-2 seconds later seems consistent.

Overanalyzation of the events.

1:16--unrelated
1:18--unrelated
1:23--corresponding to visual event at 1:11, there's an approximately 12 second delay

Explosion event at 1:11.721, explosion shockwave arrives at 1:23.817. Total sound delay is 12.096 (with some fake precision).
« Last Edit: 09/01/2016 09:08 PM by mlindner »
Internal combustion engine in space. It's just a Bad Idea.TM - Robotbeat

Offline intrepidpursuit

  • Full Member
  • ****
  • Posts: 624
  • Orlando, FL
  • Liked: 409
  • Likes Given: 243
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #608 on: 09/01/2016 09:08 PM »
The biggest thing I don't understand is what is the ignition source? Even if the O2 tank had an overpressure, underpressure, or any other failure mode, why would the first thing we see be blinding fire? Some static or something would probably catch it eventually, but without the engines running you could mix O2 and RP-1 and it would still not self detonate.

Am I missing something? I realize my optimism toward SpaceX might make me favor the least damaging outcome, but I just don't see how anything other than an explosive charge could have detonated so quickly. TEA-TEB or hypergolics could easily cause a bad day, but they don't appear to have been involved unless they were being loaded into S2 by the umbilical at the time.

If the fuel-oxygen mix is right, it only takes a tiny spark.  Electrostatic discharge, a momentary arc from a switch opening or closing, really anything...
Indeed. You might call this a *static*fire*. Heh.

That is exactly my point though, the mixture isn't right. They are totally separate until something goes wrong. It seems a stretch to think the O2 tank popped, which popped the RP-1 tank, and static happened at the right time and place to cause a spark, all within a few milliseconds. We should have seen a tank pop before it all went up.

Offline jgoldader

  • Full Member
  • ****
  • Posts: 718
  • Liked: 272
  • Likes Given: 155
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #609 on: 09/01/2016 09:09 PM »
Anybody familiar enough with S2 to know what's near the center of the "X"?  I believe ugordan was speculating on page 24 that the horizontal feature you can see extending to the left across S2 from the cradle was the clamp that holds S2, which is located near the common bulkhead for the LH2/LOX tanks.

I'm trying hard to remember that even the apparent location of the brightest part of the image (at the center of the X) is only a 2D projection of a 3D event.  The actual problem could have started on the unobserved side of the vehicle and what we're seeing is something propagating around to this side, or it could be a weak point where the event found a place to exit the vehicle, or it could be a secondary result of a primary failure that occurred far away.

So sorry for all those affected.  But I have faith in SpaceX; what they're trying to do is hard, and they only make it *look* easy.
Recovering astronomer

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32484
  • Cape Canaveral Spaceport
  • Liked: 11259
  • Likes Given: 333
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #610 on: 09/01/2016 09:12 PM »
I heard from one of the reddit that some believe a hydrazine leak from the satellite may of caused the initial explosion.  Given the location of the explosion being close the top of the oxygen tank, and the initial size explosion it seems like this is certainly possible.  I know from chemistry that hydrazine is particularly reactive, volatile, and very dangerous.  I am not an expert on launch systems and satellites, but what is the likelihood that this was the case.


The fairing would not have remained intact then

Offline DarkenedOne

  • Full Member
  • ****
  • Posts: 954
  • Liked: 58
  • Likes Given: 8
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #611 on: 09/01/2016 09:13 PM »
I heard from one of the reddit that some believe a hydrazine leak from the satellite may of caused the initial explosion.  Given the location of the explosion being close the top of the oxygen tank, and the initial size explosion it seems like this is certainly possible.  I know from chemistry that hydrazine is particularly reactive, volatile, and very dangerous.  I am not an expert on launch systems and satellites, but what is the likelihood that this was the case.

Probably near zero. The payload was well above the plane where the first flash occurred, encapsulated in a fairing that remained intact long after the initial explosion. Also there's no way for hydrazine to get *into* the LOX tank, which appeared to rupture, suggesting the event may have started inside the LOX tank.

The explosion appears close to the base of the payload and the top of the LOX tank.  There is just nothing inside the LOX tank that can cause an explosion as far as I can tell.  It has to be something that mixed with the oxygen outside of the tank.  I think they vent oxygen that boils off right there. 

Offline Norm38

  • Full Member
  • ****
  • Posts: 1161
  • Liked: 538
  • Likes Given: 774
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #612 on: 09/01/2016 09:13 PM »
Wonder how the Dragon2 Trunk/Capsule would have dealt with this 2nd Stage RUD in a Pad Abort scenario?

It'll depend on if the software is active during a static fire.  Kind of how the CRS-7 parachute deploy software wasn't running.  I hope after this they'll have the Dragon2 primed and ready to abort once the rocket is vertical and into propellant loading.

Offline mlindner

  • Software Engineer
  • Senior Member
  • *****
  • Posts: 2090
  • Space Capitalist
  • Silicon Valley, CA -- previously in Ann Arbor, MI
  • Liked: 748
  • Likes Given: 289
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #613 on: 09/01/2016 09:15 PM »
Crank up the volume, the first sound could be the creak of metal bending. If that's the case, we might be hearing a strut bend and break, a COPV rupturing or helium hose popping, and the loud boom is the visible explosion.

Please... This is an incessant problem on the internet to attribute new failures to previous issues. In engineering its the exception rather than the rule that a new failure is the same failure as before. If something fails its almost certainly something else that failed unless your engineers have no clue what they're doing or root cause was not found. This failure is NOT going to be related to struts. Forget the struts exist. That's a solved problem. Different metal suppliers, different stronger design, additional struts, impossible to be the same problem.
« Last Edit: 09/01/2016 09:30 PM by Carl G »
Internal combustion engine in space. It's just a Bad Idea.TM - Robotbeat

Offline iamlucky13

  • Full Member
  • ****
  • Posts: 1660
  • Liked: 102
  • Likes Given: 93
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #614 on: 09/01/2016 09:18 PM »
FTS is designed to cease acceleration and disperse the propellants.  This is often accomplished by a linear shaped charge running the length of the vehicle.  I suspect the Falcon is the same way.  You might not see an immediate fireball, but you would see the whole vehicle unzip basically at once if it were FTS, not the several seconds it took to completely come apart.

It would be several shaped charges, not one running the full length across both stages. The second stage disappears instantly. The rest of the rocket took a while. That said, the payload took several seconds to fall off, but cutting line through the length of the tank does not necessarily make the whole tank buckle instantly.

An FTS activation by the controller should have destroyed both stages at once, but it is not entirely out of the question that something else caused a single detonator to go off. Just extremely unlikely due to how carefully tested FTS hardware is.

Offline Firehawk153

  • Member
  • Posts: 74
  • Liked: 19
  • Likes Given: 2
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #615 on: 09/01/2016 09:19 PM »
Crank up the volume, the first sound could be the creak of metal bending. If that's the case, we might be hearing a strut bend and break, a COPV rupturing or helium hose popping, and the loud boom is the visible explosion.

Please... This is an incessant problem on the internet to attribute new failures to previous issues. In engineering its the exception rather than the rule that a new failure is the same failure as before. If something fails its almost certainly something else that failed unless your engineers have no clue what they're doing or root cause was not found. This failure is NOT going to be related to struts. Forget the struts exist. That's a solved problem. Different metal suppliers, different stronger design, additional struts, impossible to be the same problem.

I dunno, ever heard of Taurus, OCO, and Glory? Just saying...
« Last Edit: 09/01/2016 09:31 PM by Carl G »

Online eeergo


Sorry, but there is oversaturation and then there is oversaturation. According to you, those spikes should simply be as wide as the whited-out region, which is clearly not the case. My point is that it takes even more light intensity to produce spiking than it takes to saturate the detector at a region and that it was reasonable to assume that the highest physical brightness point corresponds to the point of origin. The spikes convey additional useful information about the brightness profile above the whited-out region precisely because they are more tightly constrained.


The reflection off the T/E structure is most certainly biasing any correlation between saturation and 2D projection of the explosion initiation point, I would think. The right third (roughly) of the oversaturated area should be weighted by a function of the T/E reflectivity. In essence, I would argue the true point lies a few pixels more to the left.


Further, look at the symmetry of the expanding LOX cloud a few frames later. The right side is a bit muffled because of the interference with the T/E, obviously, but the near perfect oval shape for most of its circumference suggests the failure point was mostly toward the camera, not facing directly at the erector.
« Last Edit: 09/01/2016 09:20 PM by eeergo »
-DaviD-

Offline joncz

  • Veteran
  • Full Member
  • ***
  • Posts: 372
  • Atlanta, Georgia
  • Liked: 73
  • Likes Given: 157
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #617 on: 09/01/2016 09:23 PM »
I heard from one of the reddit that some believe a hydrazine leak from the satellite may of caused the initial explosion.  Given the location of the explosion being close the top of the oxygen tank, and the initial size explosion it seems like this is certainly possible.  I know from chemistry that hydrazine is particularly reactive, volatile, and very dangerous.  I am not an expert on launch systems and satellites, but what is the likelihood that this was the case.

Probably near zero. The payload was well above the plane where the first flash occurred, encapsulated in a fairing that remained intact long after the initial explosion. Also there's no way for hydrazine to get *into* the LOX tank, which appeared to rupture, suggesting the event may have started inside the LOX tank.

The fairing remained with the payload as they fell free of the strongback.  When the payload impacts the ground, you can see another "explosion," presumably of the satellite's hypergols.
« Last Edit: 09/01/2016 09:24 PM by joncz »

Offline jongoff

  • Recovering Rocket Plumber/Space Entrepreneur
  • Senior Member
  • *****
  • Posts: 6214
  • Lafayette/Broomfield, CO
  • Liked: 2355
  • Likes Given: 790
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #618 on: 09/01/2016 09:23 PM »
One thought re: speculation about a COPV failure. The COTS-7 failure wasn't thought to be a COPV letting go per se, but a massive helium leak that led to an overpressure. A COPV flat-out failing might happen faster, and could lead to a more energetic explosion. For instance, isn't that how the S-IVB 503 upper stage was lost during a ground test (in very similar circumstances--during fueling before a hot-fire test)?

~Jon

Offline MattMason

  • Full Member
  • ****
  • Posts: 931
  • Space Enthusiast
  • Indiana
  • Liked: 645
  • Likes Given: 1145
Re: SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD
« Reply #619 on: 09/01/2016 09:27 PM »
Wouldn't the color of the initial detonation (orange) suggest an RP-1 leak?
"Why is the logo on the side of a rocket so important?"
"So you can find the pieces." -Jim, the Steely Eyed

Tags: