Hacker News new | ask | show | jobs
by bell-cot 599 days ago
When you're seeing multiple major failures from an overall system that complex, and don't have 10K spare hours to trace through all the wiring harnesses and docs before you gotta decide - then it may be reasonable to suspect that there is a common upstream cause (say, an undetected fire eating through critical stuff) and decide to bail.

My LLRV point was that even the fabled St. Neil bailed when things were obviously headed south. And that LLRV was a vastly simpler system than the F-35 - for Neil to, in theory, quickly figure out what was going wrong and land safely.
1 comments
db48x 599 days ago
No, you’re misunderstanding. The F–35 has three displays, the heads–up display in the helmet, the primary display which is a large LCD panel that takes up the entire width of the cockpit, and a smaller standby display below that. (You can find plenty of pictures.) The HMD and primary display both turned off, then rebooted and came back on several times. The standby display didn’t; it kept displaying the critical flight data and artificial horizon the entire time. There was no faulty data, there were no broken sensors, nothing. He could have simply flown the plane using the data on the standby display alone, and should have. That’s the mistake he made.

It says this directly in the report:

    24. MP’s decision to eject was ultimately inappropriate, because
    commanded flight inputs were in-progress at the time of ejection,
    standby flight instrumentation was providing accurate data, and the
    MA’s backup radio was, at least partially, functional. Furthermore,
    the aircraft continued to fly for an extended period after ejection.
Analogies are not always useful, but if your PC has three monitors and two of them are blinking on and off, do you automatically distrust what is shown on the third? Do you toss your whole computer out the window? Probably not.
link
bell-cot 599 days ago
> ..if your PC has three monitors and two of them are blinking...

Not a good analogy, because dying in a FPS game != dying IRL.

I've dealt with computers that had bad power supplies, bad upstream power, or ground loop issues. When things get weird, you really shouldn't trust the parts of the system that aren't yet acting wonky. Not with any data that matters to you.

That said, I'm neither an aerospace engineer, nor a pilot. In those fields, standard are probably different.

link