Are there critical points of failure in a launch that no amount of checks can account for?

[intelliCom]

As title says. Also, ideally mention any existing examples of such events in rocket history.

[mikegarrison]

"Go fever" is the worst one.

[Gargamel]

2 minutes ago, mikegarrison said:

"Go fever" is the worst one.

While I agree target fixation, "Go Fever" is an issue, with proper holds and a "Stop Work for All" system in place, where any Engineer or technician, even down to the lowliest truck driver or what not, can stop any countdown or work till a problem is addressed adequately is a major fix to this issue.    Even if all but one person is go for launch, any person that speaks up and says "Hey, I think we have a problem here" (and is not held accountable for a false stoppage if they're wrong) can save a mission, and encourages people to do so.   The raised issue is then reviewed and approved/fixed, and then that work/review is required to be passed up the chain so that it can be reviewed itself.   Places that have this type of system in place often reward "Stop Work" calls, even if they're erroneous, in an effort to encourage nothing getting missed.  

While an engineer(s) did speak up prior to Challenger, the local contractor signed off on the issue and never fully communicated the issue up the chain so that other outside engineers could take a look.   In the aftermath, this type of system was implemented to some degree in NASA.  While there are probably some holes in the system, it's a far stretch better than what they had before, where contractors were compelled to make deadlines and budget at the cost of quality at times. 

[StrandedonEarth]

1 hour ago, Gargamel said:

at the cost of quality at times. 

The desire for quality vs the realities of production, I know that drill all too well. Even the franchise trainers are familiar with the demands of production...

What, we lost a tile and the only reason the Shuttle didn't burn up was because there was an antenna mount behind the fuselage skin at that spot? Nah, it's fine...

[mikegarrison]

5 hours ago, Gargamel said:

While an engineer(s) did speak up prior to Challenger, the local contractor signed off on the issue and never fully communicated the issue up the chain so that other outside engineers could take a look.

That's not quite what happened, as I understand it.

The Thiokol engineers were studying the burnthrough problem and they had started to come to the conclusion/realization that a) it was dangerous, and b) it was made more dangerous by cold soak. And then Florida was hit with extremely cold (for them) temperatures just before the mission.

The Thiokol engineers convinced their bosses to tell NASA to hold the launch. The bosses agreed. NASA pushed back hard, and the Thiokol bosses bowed to the pressure. Once the Thiokol managers decided not to take a stand on the launch needing to be cancelled, NASA said, "OK then, we're launching."

So the concern was in fact pushed up the chain, and the Thiokol management only "signed off" on the launch under pressure from NASA.

[Gargamel]

Yeah I don’t remember the exact details, but the concept is the same.    One engineer saying “Whoa, hold up here” should be enough to hold a launch until it’s properly reviewed.   

[Nuke]

7 hours ago, mikegarrison said:

That's not quite what happened, as I understand it.

The Thiokol engineers were studying the burnthrough problem and they had started to come to the conclusion/realization that a) it was dangerous, and b) it was made more dangerous by cold soak. And then Florida was hit with extremely cold (for them) temperatures just before the mission.

The Thiokol engineers convinced their bosses to tell NASA to hold the launch. The bosses agreed. NASA pushed back hard, and the Thiokol bosses bowed to the pressure. Once the Thiokol managers decided not to take a stand on the launch needing to be cancelled, NASA said, "OK then, we're launching."

So the concern was in fact pushed up the chain, and the Thiokol management only "signed off" on the launch under pressure from NASA.

it was a high profile launch. that no doubt was one of the reasons nasa was pushing so hard (likely egged on by members of congress). they simply prioritized expedience over safety and paid for it. 

[steve9728]

I don't know if I can perfectly get the point of the question but here's an example of that: the failure of CZ-5 Y2 rocket which was launched on 2 July 2017 in Wenchang.

According to the documentary, the cause of the failure was discovered on 8 January the following year after the failed rocket launch by complete 'zeroing' - an examination almost from the first screw of the rocket's original design - it was discovered that "The core stage liquid hydrogen-liquid oxygen engine one-stage turbine exhaust unit suffered a local structural anomaly in the complex force thermal environment, resulting in a significant and instantaneous reduction in engine thrust, resulting in the failure of the launch mission."

Especially with a rocket that has a lot of new technologies for you: you don't know what kind of oversights your design might have, and you don't have a reference for little "tips" from a borrower, so that a fatal problem that can't be found during inspection is, in a way, probabilistically unavoidable. Something happened inside the engine like that I don't think any 'checklist before the launch' can find it out. Unless you have the item on your list to take the engine apart and look at it.

Edited October 13, 2022 by steve9728

[kerbiloid]

A typical post-crash engineer excuse:
"An unidentified foreign body in the fuel path."

[darthgently]

38 minutes ago, steve9728 said:

I don't know if I can perfectly get the point of the question but here's an example of that: the failure of CZ-5 Y2 rocket which was launched on 2 July 2017 in Wenchang.

According to the documentary, the cause of the failure was discovered on 8 January the following year after the failed rocket launch by complete 'zeroing' - an examination almost from the first screw of the rocket's original design - it was discovered that "The core stage liquid hydrogen-liquid oxygen engine one-stage turbine exhaust unit suffered a local structural anomaly in the complex force thermal environment, resulting in a significant and instantaneous reduction in engine thrust, resulting in the failure of the launch mission."

Especially with a rocket that has a lot of new technologies for you: you don't know what kind of oversights your design might have, and you don't have a reference for little "tips" from a borrower, so that a fatal problem that can't be found during inspection is, in a way, probabilistically unavoidable. Something happened inside the engine like that I don't think any 'checklist before the launch' can find it out. Unless you have the item on your list to take the engine apart and look at it.

These types of issues are supposed to be found in long duration static fires that are typically done for longer than expected durations, at higher than expected levels (throttle, temperature, pressure, etc.) until the 99.9xxx reliability criteria is met.  Then by statistical magic, an even higher dependability measure for running at normal durations and levels is arrived at.

But of course static fires don't include the same perturbations as a real world situation, so...

Edited October 13, 2022 by darthgently

[kerbiloid]

The Apollo-13 cryotank, dropped on floor, then attached to a higher voltage power line, iirc.

[steve9728]

11 minutes ago, darthgently said:

These types of issues are supposed to be found in long duration static fires that are typically done for longer than expected durations, at higher than expected levels (throttle, temperature, pressure, etc.) until the 99.9xxx reliability criteria is met.  Then by statistical magic, an even higher dependability measure for running at normal durations and levels is arrived at.

But of course static fires don't include the same perturbations as a real world situation, so...

"You don't know what you don't know" is always the worst situation for everyone (shrug)

[magnemoe]

1 hour ago, darthgently said:

These types of issues are supposed to be found in long duration static fires that are typically done for longer than expected durations, at higher than expected levels (throttle, temperature, pressure, etc.) until the 99.9xxx reliability criteria is met.  Then by statistical magic, an even higher dependability measure for running at normal durations and levels is arrived at.

But of course static fires don't include the same perturbations as a real world situation, so...

This, then you have the weird stuff you tend to see in airplane accidents. Very high safety standard but also lots of flights so from time to time you get multiple fails at once. 
Also rockets are very performance tuned even compared to fighter jets, yes fighters are designed for combat who is much more dangerous but lots of pilots who has been saved by them doing routine training. In short Starship will not be much safer than an F-15 before 2050 if they are lucky. 
Now asking the pilots if we could remove the ejection seats to save weight and maintenance on versions who is assigned to training and other non combat roles  

[magnemoe]

Now it was an weird accident with an F-16 trainer in Norway.  One guy was given an trip in an F-16 as an reward. This was an two seat trainer version. 
It crashed, reason was that that the passenger leg had pushed on the stick with his leg during an high g turn without noticing it. 
Make me wonder why is it not an lockout switch for this, either the student panic or in combat one get injured in an two man plane its fly by wire anyway. 
Now other might want to override this but that require flipping an switch not fighting the stick. 

I was in an small plane with an friend with an pilot licence, I was in front seat and had controls and he flew over an cliff who had an serious downdraft and he was working to avoid breaking the speed limit of the plane, if he did it had to be pulled for an major check who was expensive and he would probably be kicked of the club as it as an stupid mistake. 
I made very sure I kept well away from the controls, that would be much harder at 8 g. 
He then stated that the wind was high and he did 3 test runs on the landing strip saying if wind was to high we had to divert to an national airport 70 km away and take the train back. 
He landed the guy in the back seat newer flew with him again, weird