SpaceX’s Crew Dragon spacecraft static fire anomaly - THREAD 3

[baldusi]

If I'm not mistaken, the problem was not Draco then Super Draco. It was that they had fired those SD before, and fuel had crept past the one way valves.

I haven’t been following this for the past few weeks and if what you say is true it’s a fairly new finding. Or I misunderstood.

My understanding from previous discussion is that because the draco(rcs) and super dracos have cross feed the previous firing of the dracos left some N2O4 in the lines(leaking check valve). When the super dracos were pressurized this caused effectively a "water hammer" when this propellant on the wrong side of the check valve slammed into it.

Do I have this right?

As I understand it the problem was due to ground handling and material incompatibility at the high pressure of SuperDracos. The mitigation actions was to convert from one-way valves to punctured burst disks. If the problem was on doing Dracos, then SD, the burst disk would not mitigate that.

[docmordrid]

My understanding from previous discussion is that because the draco(rcs) and super dracos have cross feed the previous firing of the dracos left some N2O4 in the lines(leaking check valve). When the super dracos were pressurized this caused effectively a "water hammer" when this propellant on the wrong side of the check valve slammed into it.

Do I have this right?

The issue was an unexpected (by everyone) ignitability of titanium in the presence of a highly pressurized oxidizer like NTO.

Space News...

...implicated a leaky valve that allowed nitrogen tetroxide (NTO) oxidizer into part of the propulsion system, which, when pressurized, was hurled into a titanium check valve, igniting it..."even NASA wasn’t aware of the “compatibility issue” between NTO and titanium components at those conditions."

[FlattestEarth]

Knew there was migration, but didn't realize interaction

Hmm that is interesting.  I'm still confused as to what conditions are necessary for ignition because it was known that impacts can cause ignition with titanium and nto.  Firing a liquid slug at a check valve seems like it could definitely create a secondary impact between metal parts.  But if they are saying previously unknown materials incompatibility at high pressures it implies the absence of an impact.  Hopefully some report will go into depth on the subject.

[edzieba]

Knew there was migration, but didn't realize interaction

Hmm that is interesting.  I'm still confused as to what conditions are necessary for ignition because it was known that impacts can cause ignition with titanium and nto.  Firing a liquid slug at a check valve seems like it could definitely create a secondary impact between metal parts.  But if they are saying previously unknown materials incompatibility at high pressures it implies the absence of an impact.  Hopefully some report will go into depth on the subject.

I can't find the paper to link at the moment, but the existing literature was about tests where impactors were fired at Titanium while in the presence of NTO, and that fires started this way did not propagate. Finding that NTO itself can ignite without a metal-on-metal impact, and that this would lead to a vigorous explosion, is the "unknown unknown" encountered here.

[LouScheffer]

I can't find the paper to link at the moment, but the existing literature was about tests where impactors were fired at Titanium while in the presence of NTO, and that fires started this way did not propagate. Finding that NTO itself can ignite without a metal-on-metal impact, and that this would lead to a vigorous explosion, is the "unknown unknown" encountered here.

Here is the link to the report, from above:

This is reinforced by a more detailed report about titanium reacting with N2O4, up to and including rifle bullet strikes: "Propagation of the reaction does not occur even though sufficient N2O4 is present to allow complete oxidation of the metal", and  "In no cases, however, has ignition been observed to cover more than a small fraction of the impact area as a surface fusion only". 

Most of these reports were not done under high pressure, but you'd think the 30-06 bullet impact (about 3000 joules) would create some extremely high pressure, at least temporarily.  Even in this case, "no propagation of the ignition was observed, and neither container was damaged to a greater extent than the empty impacted container."

[DistantTemple]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

[DigitalMan]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

I was waiting to see how long it would take for someone to notice that and post a comment.  My wife does this to me when she gets her hair done, she sets a timer to see how long until I notice.

The thing is I have serious doubts about finding these kind of things if you were only doing simulations instead of tests with real hardware like SpaceX is doing.  It may cost more and take more time but my instincts tell me it is the right way to go in the long run.

I'm interested to find out what folks think about this.

[edzieba]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

Those research notes were linked in the post above yours. The known risks were for a scenario different to the one encountered. Compounding this is that Titanium has been a standard material for making valves for hypergolics for many decades, without encountering the failure mode SpaceX did.

[docmordrid]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

Those research notes were linked in the post above yours. The known risks were for a scenario different to the one encountered. Compounding this is that Titanium has been a standard material for making valves for hypergolics for many decades, without encountering the failure mode SpaceX did.

The legend is that gunpowder was discovered by alchemists attempting to create an immortality recipe, then one day someone got singed eyebrows.

Lesson: some days poo happens.

[Vettedrmr]

The thing is I have serious doubts about finding these kind of things if you were only doing simulations instead of tests with real hardware like SpaceX is doing.  It may cost more and take more time but my instincts tell me it is the right way to go in the long run.

First, I fully agree with you.  Second, it's not and either/or situation.  You have to do *both*.  You can't do many hardware failure scenarios without actually destroying the actual system (e.g. can a wing still get me on the ground if it takes damage that takes out a wing spar?).  For that test, a LOT of simulations with perhaps a single actual test sample.

I teach Aerospace Engineering to high school students, and I've been going through some of Tom Hanks' excellent "From the Earth to the Moon" series.  One is on Apollo 1, and at the end of the episode Frank Borman is asked during a hearing what his opinion was of the primary failure that led to the fire.  His answer: that it was a failure of imagination.

This is, IMO, what happened here,  NO ONE thought that this was any kind of risk.  If it had there's no way NASA, Russia, or the US government would have allowed DM-1 to dock at the ISS.  It's extraordinarily fortunate that this failure occurred on an unmanned mission, plus being on a test stand where all the evidence was easily recovered.

Here's hoping that both Boeing and SpaceX have good tests this week!

Have a good one,
Mike

[DistantTemple]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

Those research notes were linked in the post above yours. The known risks were for a scenario different to the one encountered. Compounding this is that Titanium has been a standard material for making valves for hypergolics for many decades, without encountering the failure mode SpaceX did.

HMMM... Yes I read that post (above mine). The posts I was referring to were a month or so back, and I didn't think they were about a rifle bullet. However I'm likely wrong. I have only loosely followed the discussion.

[thirtyone]

There were several posts by experiences engineers on NSF claiming surprise that NASA and SX were unaware of this risk as it WAS previously known about. Someone posted research notes a couple of decades old!
I was thus surprised that NASA made such an issue about SX discovering something that was completely unknown and unexpected!

I think the confusion stemmed from a Twitter post by someone who had rather suspiciously misquoted the original NTO/Titanium research paper by NASA/Air Force (?) by quoting right up to the "igniting" part but entirely leaving out the note that the ignition never propagated beyond the impact point.

[DistantTemple]

The thing is I have serious doubts about finding these kind of things if you were only doing simulations instead of tests with real hardware like SpaceX is doing.  It may cost more and take more time but my instincts tell me it is the right way to go in the long run.

First, I fully agree with you.  Second, it's not and either/or situation.  You have to do *both*.  You can't do many hardware failure scenarios without actually destroying the actual system (e.g. can a wing still get me on the ground if it takes damage that takes out a wing spar?).  For that test, a LOT of simulations with perhaps a single actual test sample.

I teach Aerospace Engineering to high school students, and I've been going through some of Tom Hanks' excellent "From the Earth to the Moon" series.  One is on Apollo 1, and at the end of the episode Frank Borman is asked during a hearing what his opinion was of the primary failure that led to the fire.  His answer: that it was a failure of imagination.

This is, IMO, what happened here,  NO ONE thought that this was any kind of risk.  If it had there's no way NASA, Russia, or the US government would have allowed DM-1 to dock at the ISS.  It's extraordinarily fortunate that this failure occurred on an unmanned mission, plus being on a test stand where all the evidence was easily recovered.

Here's hoping that both Boeing and SpaceX have good tests this week!

Have a good one,
Mike

Hypergolics are so important, so frequently used, and are critical to human spaceflight including current vehicles. It is a real wake-up call that such a fundamental chemical fact about titanium and N2O4.
N2O4's properties and reactions should be intimately known, and be included in "Hypergolic Propulsion 101".
It is stunning that extensive research has not been done examining the furthest reaches of N2O2's reactions with all likely "plumbing", valve and engine materials. It is after all "a powerful oxidiser" ;-) Its like O2 and grease! Its not like oxidisers don't cause problems! Its a big issue, a major reason for running fuel-rich etc.

The unsaid shock is not just the obvious one that SX was soooo lucky this didn't kill.... but that NASA, Roscosmoss, Boeing, LM etc etc.... have been apparently unaware of this risk in their past HSF vehicles!
- unless of course it is/was known, and subsequently forgotten -

So this kind of testing "should" have bee done in sponsored PHD programmes etc years (decades) ago not be left to SX to accidentally find (due to a leaky valve) on a routine test so close to a HSF debut! The engineers should have been able to look up N2O2 and titanium compatibility.

Separately the risk of a slug of "liquid" behaving like a bullet and risking physical damage at such extreme pressures should be a standard risk to factor into design. Even builders cut concrete with water, and of course steel can be cut in a similar way! ....

Thankyou :

I think the confusion stemmed from a Twitter post by someone who had rather suspiciously misquoted the original NTO/Titanium research paper by NASA/Air Force (?) by quoting right up to the "igniting" part but entirely leaving out the note that the ignition never propagated beyond the impact point.

You posted that whilst I was "essaying". That sounds likely.
However this sort of issue doesn't mean SX should slow down to Boeing speeds.... It just means the right inputs go into the engineering.....

[rsdavis9]

So I read before in this thread the hypothesis that this reaction may have caused some probes to fail at planetary orbit insertion. Has anybody heard any more about this? Wasn't a mars probe destroyed for unknown reasons at mars orbit insertion for unknown reasons?

[ugordan]

So I read before in this thread the hypothesis that this reaction may have caused some probes to fail at planetary orbit insertion. Has anybody heard any more about this? Wasn't a mars probe destroyed for unknown reasons at mars orbit insertion for unknown reasons?

Mars Observer.

[FutureSpaceTourist]

Looks like we can consider this anomaly to be cleared:

https://twitter.com/spacex/status/1194745251480498177

Full duration static fire test of Crew Dragon’s launch escape system complete – SpaceX and NASA teams are now reviewing test data and working toward an in-flight demonstration of Crew Dragon’s launch escape capabilities

[SWGlassPit]

Looks like we can consider this anomaly to be cleared:

https://twitter.com/spacex/status/1194745251480498177

It's a positive sign, but I'd be cautious with a broad statement like that.  A report outlining the investigation and corrective actions taken will be what "clear" the anomaly, not merely a single test.  After all, prior to the anomaly, SDs had been successfully fired.

Again, this is a good thing, but this in itself isn't everything.

[woods170]

Looks like we can consider this anomaly to be cleared:

https://twitter.com/spacex/status/1194745251480498177

It's a positive sign, but I'd be cautious with a broad statement like that.  A report outlining the investigation and corrective actions taken will be what "clear" the anomaly, not merely a single test.  After all, prior to the anomaly, SDs had been successfully fired.

Again, this is a good thing, but this in itself isn't everything.

No it isn't. But the two-dozen single-quadrant hotfire tests that have been performed with the modified system PRIOR to this all-up hotfire test are not nearly as interesting....to some folks.

[Nomadd]

 I'm waiting for a test that duplicates the conditions of the failed one. Meaning a used capsule that's hit the water and is subject to extreme vibration during the test. No reason to think any of that aggravated the problem, or that test is necessary before manned flight. It just seems like a needed step to declare victory.

[rpapo]

I'm waiting for a test that duplicates the conditions of the failed one. Meaning a used capsule that's hit the water and is subject to extreme vibration during the test. No reason to think any of that aggravated the problem, or that test is necessary before manned flight. It just seems like a needed step to declare victory.

Unless I missed something somewhere, though, the modified system cannot be simply re-fired like the original system was intended to be.  You have to replace the diaphragms first.