Updates SpaceX Falcon 9 F5 CRS SpX-2 through CRS SpX-12 Updates

[Quick_Nick]

I figure it's relevant to the thread.
Scott Kelly talks about topics including recent issues relating to ISS, NOW!
http://www.nasa.gov/multimedia/nasatv/

Edit: Nothing unexpected of course. Kelly says they're good til mid September or October.

 

[Thunder Chicken]

There must have been a hole, much larger than the usual piping because of the much higher mass flow rates.

Why so? If the O2 was still liquid at the breach, it would not need to be a very large opening to deliver the amount of gas produced. To my eye, the rate of gas escape was something on the order of the engine propellant rate (just judging by the plume), so we're looking at maybe a couple of cubic meters per second of LOX (a couple of tons per second).

If the tank was overpressurized and one or more of the lines failed, there would be very little resistance to the flow and so a considerable amount of LOX could be forced out. I don't know what the line dimensions are, but I don't see why a hole larger than a few 10's of centimeters across would be needed to release that much LOX.

 

[Urwumpe]

Why so? If the O2 was still liquid at the breach, it would not need to be a very large opening to deliver the amount of gas produced.

gaseous oxygen is invisible. All you see is the interaction of the cold gas with the air. But still, that's a lot - When a Saturn V S-IC was dumping multiple hundred kilogram LOX per second after separation, you see a much smaller plume. And that used the main feed lines at tank head pressure.

---------- Post added at 07:09 PM ---------- Previous post was at 07:07 PM ----------

We don't have any information regarding the time from the overpressure starting to the time of the observed failure, so how can you say that a particular model matches what was observed? I assume SpaceX has this information, but they're not sharing it.

1. No indication before FTS, public affairs officer was commenting on a good flight, fractions of a second later, the cloud appears.

2. SpaceX is trying to recover final telemetry frames by hex editor for finding the CAUSE. If you would have the cause seconds earlier, you would not need that in that urgency, restoring the final telemetry frames is for telling what happened last, if necessary at all. (Not all telemetry is transmitted at once and at the same rate, usually, less the important values multiple times per second, the house-keeping data every other second)

 

[RCO]

1. No indication before FTS, public affairs officer was commenting on a good flight, fractions of a second later, the cloud appears.

2. SpaceX is trying to recover final telemetry frames by hex editor for finding the CAUSE. If you would have the cause seconds earlier, you would not need that in that urgency, restoring the final telemetry frames is for telling what happened last, if necessary at all. (Not all telemetry is transmitted at once and at the same rate, usually, less the important values multiple times per second, the house-keeping data every other second)

Much of the data in the telemetry might be subtle, and would not necessarily be obvious in real time. There is TONS of data, and it's not possible to analyze it all in real time. And the public affairs narrator would only announce what's totally obvious or what he's already overheard on the main voice channel.

Lack of a commanded destruct doesn't tell us anything useful. Range Safety doesn't destruct the vehicle simply because there's an anomaly. Only when the vehicle is out of control and threatening to leave the safety corridor, or after it's 100% obvious that the vehicle won't make it to orbit, then they'll command destruct. For example, when the Merlin engine failed on a previous flight, the FTS wasn't activated. Lack of a destruct command is not evidence of no anomalies.

I don't think too much should be read into the fact that engineers were looking at the last few milliseconds of data. I'm sure engineers want ALL the data available, including the final milliseconds. But this does not exclude the possibility that there is relevant data prior to the final milliseconds. After all, Musk commented about the LOX overpressure almost immediately after the incident. So this data obviously wasn't from the final milliseconds of telemetry that needed to be reconstructed. The overpressure event could have been from ANY time prior to this.

---------- Post added 07-03-15 at 05:29 AM ---------- Previous post was 07-02-15 at 11:15 PM ----------

Thinking about it now, the LOX tank probably has 2 redundant vent valves, so it seems less likely to be related to a sticky vent valve or such, as I earlier suggested. Although I'm still not convinced it had to be a sudden failure which gave no earlier indications in telemetry or engineering video. I think it likely that SpaceX has at least some information which they haven't shared.

 

[MaverickSawyer]

I think it likely that SpaceX has at least some information which they haven't shared.

Well, one, it's an ongoing investigation, so we shouldn't expect anything different than a typical FAA/NTSB investigation, and two, ULA would intentionally spin the data to make SpaceX look unsafe, so playing it close to the vest is the smart move.

 

[RCO]

Well, one, it's an ongoing investigation, so we shouldn't expect anything different than a typical FAA/NTSB investigation, and two, ULA would intentionally spin the data to make SpaceX look unsafe, so playing it close to the vest is the smart move.

I agree that keeping silence makes sense. I am not making any kind of a moral judgment regarding their silence. My point is that some people seem to be assuming that there is no relevant data in the telemetry prior to the final milliseconds that were being reconstructed, otherwise SpaceX would have given us this data because Elon Musk is our best friend. And therefore the failure must have occurred very rapidly and not given much indication in the telemetry.

But this is not a reasonable assumption. We should not assume that the only data so far from the telemetry is "2nd stage LOX overpressure", just because this is all the information that SpaceX has given us. After doing their data analysis, SpaceX may have discovered that there were other unusual things going on with, for example, He bottles or engine LOX plumbing, but chosen not to share that information.

 

[RCO]

As I think about it and imagine scenarios that fit all the data convincingly, I've come full circle now back to my first comment, and think this could have been just a structural failure of the LOX tank, possibly related or possibly unrelated to the rumored liner issues. I don't think there's a need to resort to He bottle, LOX plumbing, or other tank pressurization issues. I think the overpressure may be a side effect, not the cause. I think it may simply be that the LOX tank wall buckled inward. This buckling compressed the tank, creating a pressure spike and spraying LOX out the vents for a split second before the tank burst. As the walls proceeded to buckle and collapse, the Dragon dumped off to the side. The whole thing then pancaked and banana peeled from aerodynamic and g forces.

 

[Donamy]

All in all, I think some valuable lessons will be learned.

 

[Urwumpe]

All in all, I think some valuable lessons will be learned.

Make sure your next COO bears the name "Launchgood". :lol:

 

[Donamy]

COO ?

 

[DaveS]

COO ?

Chief Operating Officer.

 

[Urwumpe]

Chief Operating Officer.

which currently just "Shot well" ....

 

[Andy44]

which currently just "Shot well" ....

That should be a good name for a space executive, since space launches were once called "shots" by some folks, probably because the early lanchers were all ballistic missiles. Moon missions are called "moon shots", for example.

Now if your COO's name is Otto Tankstirrer, you may have an issue...

 

[Sky Captain]

IIRC during testing one Saturn V S-IVB stage suffered helium tank explosion causing total destruction of the stage. In this case something similar may have happened. Suddenly rupturing helium tank would be sudden event with no trouble signs showing up on telemetry before the explosion.

 

[MaverickSawyer]

IIRC during testing one Saturn V S-IVB stage suffered helium tank explosion causing total destruction of the stage. In this case something similar may have happened. Suddenly rupturing helium tank would be sudden event with no trouble signs showing up on telemetry before the explosion.

There was a S-IVB that did rupture on the test stand in Rancho Cordova, but that was due to poor communication between test shifts and a miswired sensor, iirc.

 

[GLS]

There was a S-IVB that did rupture on the test stand in Rancho Cordova, but that was due to poor communication between test shifts and a miswired sensor, iirc.

The only S-IVB (that I know) to have blown up during testing, did so because of a weld.

The group finally traced the source of the explosion to one of the eight ambient-temperature helium storage spheres located on the thrust structure of the J-2 engine. The exploding sphere ruptured the propellant fill lines, allowing liquid oxygen and liquid hydrogen to mix and ignite, setting off an explosion that wrecked the stage. Further analysis showed that the sphere had been welded with pure titanium weld material, rather than the alloy material specified. The helium sphere and the weld seam had been previously tested to withstand extremely high overpressures, but repeated tests on the sphere prior to the acceptance firing sequence had created the weakness that ultimately resulted in disintegration of the sphere and destruction of the stage.

 

[RisingFury]

Expect to reach preliminary conclusions regarding last flight by end of week. Will brief key customers & FAA, then post on our website.

Hmmm... we'll see...

 

[hellotoall]

I saw some other forums are placing the blame towards the COPV helium bottles. COPV being composite overwrap pressure vessels. Ill give a brief description of this - usually pressure vessels are metallic and ductile, however composites give a great strength to weight ratio but are not ductile. During some normal limit loading or even cool down from curing they can undergo whats called micro-cracking (small crack in the resin between fibers). This can cause leaks in a pressure vessel. So the solution is to have a thin metallic liner that you can also use as a tool and wind the composite over this. So use the thin metallic liner to prevent leakage while having the outer composite carry the load, and (depending on material), strength on par with titanium but 1/3 the weight you may be able to save some weight. So obviously composites have there up and down sides. Tailored stiffness is a plus but environmental effect such as water ice ingress (opening up micro-cracks) or even the cure process (to spread resin properly and remove reaction by-products). Can affect the strength. Normally they are tested to proof before acceptance, which should be a factor on top of the limit loads. And during the development stage they will certainly be tested to burst numerous times (with technicians hiding in an underground bunker). Im not sure what kind of NDT (non destructive testing they can do on the part, i doubt they can do ultrasound, maybe c-scan or xray not sure? Anyways ive read maybe one of these gave way - but i dont know - i still think maybe something attached to the wall had a poor fit. Normally composite parts sound great in a brochure but because of unknowns people still size them beefily for applications so the weight saving is never just the ratio of density versus metallic parts. So while theres a few days left to speculate - ill say the tank wall rather than pressure vessels.

 

[MaverickSawyer]

Having a little experience with NDT, I suspect that they would use an X-ray if it were a purely composite structure. The metal inner layer complicates things, though. They probably use at least two different techniques to perform the inspection.

 

[hellotoall]

On a side note, thats the one thing about speculating which is tough. To use an analogy - many incidents are a result of pilot error. Sometimes rocket incidents are a result of operator/technician/engineer error, especially if they are pretty much single use products - but going down that road gets very political very fast...

'your saying engineers and technicians are inexperienced?'

'im sure they have skilled people you cant say that...you just hate spacex'

So take all the people in the prime company out of the debate and we are left with all the lowly suppliers hehe, vendors and zeus and the gods.

So always have to watch the wording carefully.

---------- Post added at 04:29 PM ---------- Previous post was at 04:13 PM ----------

I guess why i was thinking about the tank wall. Was because i thought the aluminum used was relatively brittle plus the welding process plus any rework needed if bolt dont fit properly. Plus that type of aluminum would have whats known in the composites world as bearing/bypass interaction. For example a plate of this material under a huge amount of tension wont necessarily then have the ability to carry the same bolt load if it was not under tension. Brittle skins are like that - of course that would be thought of but coupled with other failures e.g doubt wall mounts are sized with margins that include missing or loose fasteners.

So with perhaps ultimate factor of 1.25, some local wall thinning, some rework to drill out and install next-size-up fasteners upsetting load distribution, some increased vibration amplitudes of slightly loose structure. Who knows....guess have to wait until end of the week.