SpaceX Falcon 9 - AMOS-6 - (Pad Failure) - DISCUSSION THREAD (2)

[Bubbinski]

But there was a big wildfire at VAFB, in the Worldview thread there was discussion about people checking on possible damage to infrastructure (not the Pad but other things needed to support launches). Don't know when they'll be able to resume launches from there.

[Kaputnik]

If the He storage switched to liquid, what impact would that have on the loiter time of the upper stage?

[Jim]

If the He storage switched to liquid, what impact would that have on the loiter time of the upper stage?

High pressure is needed, and not just gas

[yokem55]

If the He storage switched to liquid, what impact would that have on the loiter time of the upper stage?

High pressure is needed, and not just gas

Turbopump spinup?

[Stranger]

If the He storage switched to liquid, what impact would that have on the loiter time of the upper stage?

High pressure is needed, and not just gas

high pressure is need to COPV before filling with helium?

[mulp]

"SpaceX is knocking out the traditional Congress favored corporate cronies by coming in much lower in cost/price."

But it looks like there is additional risk

Are you saying Ariane 5 is too high risk? It's failure rate at 6 years was higher than Falcon 9 is at 6 years.

Challenger loss was the 25th flight in the 6th year of Shuttle program.  It merely made totally clear the design tradeoff of Shuttle.

On the previous Falcon 9 failure, it is believed that the initial crew safety system in Dragon would have saved the cargo if the code to activate it were enabled. I haven't seen any discussion of Dragon crew escaping safely with the crew in this case, if it were a crewed mission, but this is one of the failure modes it's required to meet.

If safety means sticking with 80s technology because it requires 20-30 years to get the bugs out, the budgets for everything space need to be much bigger, or the demand for space needs to be much lower.

I'll simply mention Takada as an example of sticking with a design with cost-safety tradeoffs. I'm sure many more can be mentioned to point out this is common across everything involving technology.

[Jim]

"SpaceX is knocking out the traditional Congress favored corporate cronies by coming in much lower in cost/price."

But it looks like there is additional risk

Are you saying Ariane 5 is too high risk? It's failure rate at 6 years was higher than Falcon 9 is at 6 years.

Challenger loss was the 25th flight in the 6th year of Shuttle program.  It merely made totally clear the design tradeoff of Shuttle.

y tradeoffs. I'm sure many more can be mentioned to point out this is common across everything involving technology.

Ariane and Shuttle have nothing to do with this discussion

[Navier–Stokes]

Update from Gwynne Shotwell at National Academy of Engineering 2016 Annual Meeting:

Jeff Foust @jeff_foust 1:09 PM - 9 Oct 2016

Shotwell: “homing in” on cause of Sept. 1 pad accident; not pointing to a vehicle issue. Hope to fly a couple more times this year.

Note: emphasis mine.

[Stranger]

As part of the mechanism of both the two accidents (this accident and 28 June 2015) think you do not need anything. Their mechanism is simple and obvious.

This accident ruptured or torn off the bottom of the upper helium balloon. The balloon under the action of reactive force helium tore down intertank stitched the bottom like a bullet metal sheet, and crashed into PR-1. Carrying over a stream of oxygen and a cloud of sparks punched metal. All mixed up and exploded.

In an accident in flight in 2015 was the same, only in reverse. To break through or torn off the lower bottom of the helium balloon. The balloon under the action of reactive force helium rushed up with a decent layer of liquid oxygen. The network has a lot of video, where the usual bullet when fired flies in water only 2-3 meters and with a shot at the zero axis speed goes down. So it is here - the balloon has slowed down in the liquid oxygen before reaching and damaging the Dragon. Intertank the bottom of the bottom of the tank and PR-1 also survived. A tank of oxygen under the influence of excess pressure and shock waves destroyed. As a result, we saw a light air accident (up to undermine the surviving part of the missile after 8 sec.) - Was the release of oxygen and helium, nothing more. It is possible that traces of the balloon upward movement objectively developers pulled out of the telemetry that gave them the version of a broken suspension. He really broke up, not because of weakness in the Archimedes force, but because of the action of the huge reaction force under 1000 tonnes. From the release of helium. The failure of the version of the Archimedes force no one, of course, I did not believe. But when testing some hangers really broken with a load lower than the estimated, the version was ready to break the deadlock and resume starts. What went wrong? Suspension. Why? They were poorly made. The prove? Some of them broke down during the test load specified in the requirements. What is done to fix the problem? Replaced suspensions to the relevant requirements. Everything, all thanks, all free, no questions. Really, then, it believed that the cause of the accident was something so improbable, that, perhaps, no longer happen. But the second accident not long in coming, and clarified the situation.

The root cause - as it is written above. Complex phenomena in COPV. Which were not taken into account in the design and manufacture, combined with errors led to the fact that the operating pressure with some probability may exceed the actual burst pressure. If you do not replace the balloon or redesign completely, then it explodes again.

[Herb Schaltegger]

The root cause - as it is written above. Complex phenomena in COPV. Which were not taken into account in the design and manufacture, combined with errors led to the fact that the operating pressure with some probability may exceed the actual burst pressure. If you do not replace the balloon or redesign completely, then it explodes again.

Not according to Ms. Shotwell (see the Updates thread). As noted by several folks, and indicated by prior info in L2, the root cause here seems to be operational in nature - SpaceX blew up the rocket by doing something different in the Hotfire on September 1 that they had not done in prior tests and launches. Either they changed procedures or there was a procedural error.

[Toast]

Ariane and Shuttle have nothing to do with this discussion

He was replying to the assertion that Falcon was riskier than the competition. Clarifying what you are defining as competition is very much relevant to the discussion--if the answer is the previous US crewed vehicle then it's not clear that Falcon is any riskier.

[ccicchitelli]

"Shotwell: “homing in” on cause of Sept. 1 pad accident; not pointing to a vehicle issue. Hope to fly a couple more times this year."

https://twitter.com/jeff_foust/status/785210649957789698

[CameronD]

Shotwell:   ....not pointing to a vehicle issue.....

That does not mean the issue didn't occur on the vehicle... and doesn't tell us anything we didn't suspect already.

Ariane and Shuttle have nothing to do with this discussion

He was replying to the assertion that Falcon was riskier than the competition. Clarifying what you are defining as competition is very much relevant to the discussion--if the answer is the previous US crewed vehicle then it's not clear that Falcon is any riskier.

The point here is, as Herb pointed out above, SpaceX are still learning optimising procedures here and in this case something got 'improved' wrong.  If so, then all they need do once they understand exactly what went wrong is roll back their procedures to within the known bounds of risk and they're good to go.  As mature systems, Ariane and Shuttle don't fit this picture.

[Toast]

As mature systems, Ariane and Shuttle don't fit this picture.

Just because they are (or were, in the case of the Shuttle) more mature platforms doesn't mean we can't compare the levels of risk in each. Based on a quick Monte Carlo simulation, I estimate that there is a ~26% chance that Falcon 9's long-term reliability will exceed Ariane 5's, and a ~9% chance that it will exceed the Space Shuttle's. That's not incorporating improvements in reliability over time (I haven't gotten around to analyzing how much of an impact that has), so there's a chance that those odds could be even better. Though that's not including some things that are difficult to quantify, like the differences in launch abort between Dragon and the Space Shuttle.

[pippin]

And let's not forget Shuttle was eventually shut down after the inherent risk in its design made further operation outrageously expensive.
Ariane got fixed and it cost them lots of work on the Vinci 2 engine to do so and pretty much testing, too.
Things that SpaceX will now have to do, too, and which will obviously add to their cost structure as well.


Sent from my iPad using Tapatalk

[spacekid]

And let's not forget Shuttle was eventually shut down after the inherent risk in its design made further operation outrageously expensive.

I believe Shuttle was shut down because a review board wanted the shuttle to be re-certified which made it expensive.

[ccicchitelli]

Shotwell:   ....not pointing to a vehicle issue.....

That does not mean the issue didn't occur on the vehicle... and doesn't tell us anything we didn't suspect already.

Yes it does? (Assuming her statement proves true)

We know the tank burst, but that doesn't mean it shouldn't have burst. It's not indestructible. Something not related to the vehicle going out of spec in ways the vehicle isn't designed to handle doesn't mean the vehicle is at fault.

[GerryB]

Why is the COPV filled before the oxygen tank is? After the oxygen cools the helium, you'd need to do much more He filling since the tank would have depressurized with the lower temperatures. Unless the He is precooled to the expected  60K tank temperature even before LOX is loaded. But if that's true then all the theories here mentioning  heat stress, thermoacoustic hammers, and temperature gradients are  invalid.

[HMXHMX]

Why is the COPV filled before the oxygen tank is? After the oxygen cools the helium, you'd need to do much more He filling since the tank would have depressurized with the lower temperatures. Unless the He is precooled to the expected  60K tank temperature even before LOX is loaded. But if that's true then all the theories here mentioning  heat stress, thermoacoustic hammers, and temperature gradients are  invalid.

It's possible they fill some helium to keep the thin aluminum liner in tension, to prevent it debonding, wrinkling or cracking as the LOX cools the COPV.  Then once the COPV is cold, you top helium to max pressure. 

[envy887]

Gerry, the LOX and helium tanks are filled more or less simultaneously according to the countdown. The LOX fill finishes first and the helium tank is topped off shortly after.