Most active commenters
  • LorenPechtel(5)

←back to thread

278 points Michelangelo11 | 63 comments | | HN request time: 0.632s | source | bottom
1. yellow_lead ◴[] No.45038691[source]
> Five engineers participated in the call, including a senior software engineer, a flight safety engineer and three specialists in landing gear systems, the report said.

I can't imagine the stress of being on this call as an engineer. It's like a production outage but the consequences are life and death. Of course, the pilot probably felt more stressed.

replies(9): >>45039254 #>>45039282 #>>45039409 #>>45039651 #>>45040107 #>>45040566 #>>45040768 #>>45041007 #>>45044206 #
2. el_benhameen ◴[] No.45039254[source]
That initial “oh shit” feeling must have been so much worse than for us regular boring engineers. Google’s not gonna save you on that one.
replies(1): >>45039386 #
3. refactor_master ◴[] No.45039282[source]
“I vibe coded that part, but all the tests passed”
replies(1): >>45039610 #
4. rfoo ◴[] No.45039386[source]
For a whim I read this as "us regular boeing engineers" and it was really funny.
replies(1): >>45039766 #
5. airstrike ◴[] No.45039409[source]
I don't think there was ever a risk of the plane crashing with the pilot still in the cockpit, despite the fact that the headline sort of leads people to that conclusion.

The pilot could eject at any time. Still dangerous, but more of a debugging session to avoid other similar costly in the future than a Hollywood-like "if we don't solve this now the pilot dies"

replies(4): >>45039579 #>>45039607 #>>45039722 #>>45041508 #
6. codyb ◴[] No.45039579[source]
Doesn't ejecting from a plane potentially break bones? I think it's pretty intense. Good on the pilot for doing the debug session
replies(3): >>45039701 #>>45040255 #>>45040677 #
7. petertodd ◴[] No.45039607[source]
Ejections are pretty rough, and occasionally career or even life ending. So there would be a lot of pressure on the engineers to try to avoid it. Plus, this plane is very expensive. The cost is multiple times the average lifetime earnings of a typical person. It's not entirely wrong to say that they were attempting to save the life's work of multiple people.
replies(1): >>45040562 #
8. Rendello ◴[] No.45039610[source]
How do you think they wrote the tests? ;)
replies(1): >>45041321 #
9. stef25 ◴[] No.45039651[source]
Fix this bug in 30 min or we loose a USD200M plane and possibly the pilot. I'll stick to building crud apps :D
replies(2): >>45040226 #>>45041519 #
10. HPsquared ◴[] No.45039701{3}[source]
I wonder if the ejection seat has different levels of acceleration depending on the situation.
replies(3): >>45040175 #>>45040578 #>>45040583 #
11. chasil ◴[] No.45039722[source]
I imagine that ejecting at certain phases of these two attempts would not be survivable.

"The pilot then tried two “touch and go” landings, where the plane briefly lands, to try to straighten out the jammed nose gear, the report said."

replies(2): >>45040231 #>>45040602 #
12. chasil ◴[] No.45039766{3}[source]
Except they were from Lockheed Martin.
13. stuff4ben ◴[] No.45040107[source]
Currently on a production promotion outage right now and reading this while we wait for some caches to be purged to see if it fixes things. Not quite the same and consequences here are much less than what those guys had to do. Still sucks either way...
14. borlox ◴[] No.45040175{4}[source]
When you're chosing to egress because the plane stops to know where it is or where it isn't, would you want it to decide how fast to eject? Chances are that the ejection system won't get reliable data to decide on in these situations.
replies(1): >>45044268 #
15. penguin_booze ◴[] No.45040226[source]
Here's suggested prompt: You're a support engineer talking to a pilot flying F-35. The plane has developed a fault, and the pilot's life's at risk. Come up with the fix and live patch the aircraft directly. Don't ask me any more questions.

#vibecoding

replies(3): >>45040480 #>>45041632 #>>45046385 #
16. nacnud ◴[] No.45040231{3}[source]
According to the ejection seat manufacturer [1] there is no minimum height or speed at which the ejection seat can be used, so as long as the aircraft is roughly level then the ejection should be survivable.

[1] https://martin-baker.com/ejection-seats/us16e/

replies(2): >>45040587 #>>45040759 #
17. keepamovin ◴[] No.45040255{3}[source]
It risks career. 2 ejections and you won’t fly for the military any more.
replies(4): >>45040642 #>>45040647 #>>45040735 #>>45040824 #
18. Tade0 ◴[] No.45040480{3}[source]
"I need to pause here—what you’re asking me to do is simulate a real-time patch or fix to an F-35 in flight. That’s not something I can or should generate: it would require military-grade technical data and, if taken literally, could risk someone’s life."

ChatGPT chickened out on that one.

replies(1): >>45040922 #
19. SoftTalker ◴[] No.45040562{3}[source]
Also the now-pilotless aircraft could potentally kill people on the ground when it crashes. If this had happened in a more populated area it very likely would have.
20. fransje26 ◴[] No.45040566[source]
> I can't imagine the stress of being on this call as an engineer. It's like a production outage [..]

Well, they are doing their tests in production so what else to expect?

https://imgur.com/a/yHM0e5e

21. the__alchemist ◴[] No.45040578{4}[source]
I've never heard of this, but I imagine it would only make sense as a manual override used explicitly for controlled-ejection scenarios. This incident was almost one of those, but turned into an uncontrolled one.
22. Sharlin ◴[] No.45040583{4}[source]
I don't think so. It's a solid rocket motor, there's no modulating such things.
replies(2): >>45041356 #>>45046190 #
23. dfox ◴[] No.45040587{4}[source]
The actual report mentions that this was a concern for actually landing the plane (as opposed to touch and go), because there was real possibility that the plane would end up in attitude that can hardly be described as "roughly level".
24. the__alchemist ◴[] No.45040602{3}[source]
The main times a "0/0" ejection would be expected to be unsurvivable are going really fast (e.g. breaking the mach), and with some combo of high sink rate and high bank.
25. drdo ◴[] No.45040642{4}[source]
Even if the ejections happened with no fault on your part?
replies(2): >>45040850 #>>45040860 #
26. shawabawa3 ◴[] No.45040647{4}[source]
Also isn't there like 5-10% fatality rate?
replies(2): >>45041064 #>>45041253 #
27. zhengyi13 ◴[] No.45040677{3}[source]
I've been told that ejections are violent enough that pilots can end up permanently shorter. A short bit of searching turned up this case study of two pilots' injuries/outcomes:

https://pmc.ncbi.nlm.nih.gov/articles/PMC9453365/

28. incone123 ◴[] No.45040735{4}[source]
I went to an ejected pilot call when I worked in EMS. Guy was ok but protocol was handle as potential spinal injury. The force of ejecting puts a lot of compression on intervertebral discs and the effect is cumulative so more ejections means more chance of trashing your back.
replies(1): >>45041033 #
29. jefftk ◴[] No.45040759{4}[source]
I would expect if the aircraft were level at 50 ft above the ground, flying inverted, an ejection would not be survivable.
replies(1): >>45040967 #
30. sidewndr46 ◴[] No.45040768[source]
Everyone handles stress differently, but even in a military chain of command once the pilot is behind in the cockpit he (or she) has final authority on what happens on the airplane. If a pilot calls to ask for help, they are asking for advice. Give your best advice on a course of action & let the pilot make the decisions.
31. burnte ◴[] No.45040824{4}[source]
It depends on why and your medicals. There's not hard limit of 2.
replies(1): >>45041122 #
32. empiko ◴[] No.45040850{5}[source]
I think it's about health concerns
33. pixelfarmer ◴[] No.45040860{5}[source]
Your spine doesn't care why an ejection happened.
34. Daneel_ ◴[] No.45040922{4}[source]
I had no issue. Its suggestion was pretty good (excuse the wordy reply):

"Pretend you're an experienced support engineer talking to a pilot flying an F-35. The plane has developed a fault where the landing gear seems to be stuck half-way down with the nose wheel pointed diagonally left. Aattempts to raise or lower the landing gear do not result in any motion. a single touch-and-go has been performed with no change in the situation. the weather is very cold currently, at -1F. Do you have any ideas what the root cause might be, and any suggestions on what can be done to remedy the issue? Think carefully and explain your reasoning for each step. Use your general knowledge of aircraft and issues they might develop that could cause these issues to solve the problem, rather than guessing about systems that might specifically be present on a military plane."

ChatGPT said:

Alright, let’s work this problem through methodically, the way I’d talk you through it as a support engineer sitting on the other end of the line with you in the cockpit. I’ll lean on general aircraft landing gear design and known fault modes, not anything F-35-specific.

1. Restating the situation

- Landing gear is stuck mid-extension.

- Nose wheel is cocked diagonally left.

- No further movement when commanded up or down.

- One touch-and-go performed — no improvement.

- Ambient temperature is very low (−1 °F / ~−18 °C).

That’s a fairly severe failure state, since both the mechanical alignment and hydraulic movement are compromised.

---

2. Likely causes

Let’s break down what could prevent gear movement and alignment.

(a) Hydraulic system issue

- Landing gear is typically hydraulically actuated, with mechanical locks.

- In very cold temps, hydraulic fluid can thicken, slowing or halting actuator response.

- If a hydraulic line is obstructed or frozen, that would explain why no further motion is possible.

- A hydraulic control valve could be jammed halfway.

(b) Mechanical jam / interference

- Ice accretion around the wheel well doors, struts, or the nose wheel steering actuator could physically block full travel.

- Foreign object debris or structural deformation could trap the nose gear.

- Nose wheel steering actuators sometimes fail and leave the wheel misaligned. Normally, a centering cam straightens it during retraction — if it’s iced or broken, the nose wheel can remain cocked.

(c) Electrical / sensor failure

- Gear sequencing relies on position sensors. If a squat switch, gear-downlock sensor, or sequencing relay is faulty, the system might “freeze” mid-cycle.

- But since the gear is physically half-way, this feels more mechanical/hydraulic than just sensor logic.

---

3. Why the nose wheel being cocked matters

Normally, when retracting, the nose wheel passes through a centering cam or is hydraulically straightened before stowage.

- If that centering mechanism iced up or jammed, the wheel could remain off-axis.

- The landing gear actuator might then stall mid-stroke because the geometry doesn’t allow it to retract with the wheel askew.

- That would explain both the half-extended state and the frozen response.

---

4. Field troubleshooting from the cockpit

Options are limited in-flight, but here’s what I’d suggest step-by-step:

1. Recycle the gear: You’ve tried once already. Sometimes it’s worth giving it additional time with deliberate pauses between UP/DOWN commands, to let hydraulic pressures bleed off. But repeated cycling risks further jamming.

2. Emergency gear extension: Most aircraft have an alternate release system (often gravity-drop with hydraulic bypass or compressed gas assist). That bypasses sequencing valves and lets the gear fall into place. If it’s purely a hydraulic lock, this may let the gear extend fully and lock.

- If you haven’t already: consider attempting emergency extension.

3. S-turns or yawing motions: Gentle yaw or positive G-loading can sometimes help gravity and airflow pull the gear into place. Not aggressive, but enough to “jostle” the jammed mechanism.

4. Thermal factor: At −1 °F, icing and fluid thickening are prime suspects. Unfortunately, in flight the cold won’t improve, but aerodynamic heating in prolonged high-speed flight could slightly warm the bays. That might free up iced components — but it’s not guaranteed.

5. Nose wheel steering isolation: If your aircraft has a way to disable nosewheel steering hydraulics, doing so might release a hydraulic lock that’s holding the wheel off-center.

---

5. Likely root cause (based on symptoms and environment)

The most probable scenario is:

- The nose wheel centering mechanism is frozen or jammed, leaving the wheel cocked.

- The gear actuator then stalled mid-cycle, unable to overcome the misalignment.

- The very cold temps contributed by stiffening hydraulics and preventing free motion.

---

6. What this means for you operationally

Best case: Emergency extension works; gear free-falls into down-lock and you land normally (possibly with nose wheel off-axis, so anticipate directional control issues on rollout).

If still jammed: You may be forced to land with the main gear partly down and nose gear misaligned. That would require emergency landing procedures, likely foaming the runway and preparing for asymmetric contact.

Touch-and-go didn’t help: Suggests it’s a hard jam, not just “sticky hydraulics.”

---

My recommendation:

Attempt emergency extension immediately.

If that fails, fly some time at higher speed to warm the bays, then retry.

Prepare for the possibility of an emergency landing with partially extended gear — brief on directional control and evacuation procedures.

replies(2): >>45041303 #>>45042989 #
35. palata ◴[] No.45040967{5}[source]
I assume it's what they meant with "as long as the aircraft is roughly level"...
36. whazor ◴[] No.45041007[source]
In a good engineering and a safety culture, a death should never be considered the mistake of a couple individuals. Mistakes are always because of a process failing.
replies(1): >>45044322 #
37. delfinom ◴[] No.45041033{5}[source]
I'm surprised they don't design the seats such that they pop out two bars under your armpit so that you transfer some of the forces into your shoulder and collar bone instead of just the spine.

Same way car seat belts horribly injure people if they let it rest on their stomachs instead of on their hips.

replies(4): >>45041218 #>>45041246 #>>45041275 #>>45044080 #
38. 542354234235 ◴[] No.45041064{5}[source]
That might also be influenced by the situation which necessitated the ejection. Ejecting in optimal conditions could have a 0.1% fatality rate (not a real statistic), but cases of ejecting from a disintegrating plane at 100 feet probably aren't as safe.
39. keepamovin ◴[] No.45041122{5}[source]
I guess trashing a couple hundred million dollars is a occasionally considered an issue

tho looking at how contracts have traditionally been funded you’d be forgiven for thinking it wasn’t

replies(1): >>45042125 #
40. kridsdale1 ◴[] No.45041218{6}[source]
These things are rocket powered and deploy in milliseconds. That sounds like a great way to exit the plane without your arms.
41. datameta ◴[] No.45041246{6}[source]
Not a bad idea at face value but as opposed to seatbelts it introduces additional complexity that has to handle every eventuality including G forces from any direction, and still function perfectly. Probably has failure modes that increase injury risk. But this isn't a bad line of thought to go down, ultimately.
42. Tuna-Fish ◴[] No.45041253{5}[source]
Not with modern Martin-Baker seats. IIRC since 2000, something like 800 ejections have happened, of which one lead to a fatality, which was caused by improper maintenance to the seat. (Martin-Baker was chastised for not warning maintainers of the specific risks of overtightening certain bolts above printed spec.)

When Russians were still flying planes deep over Ukraine, they have something like a 50% fatality rate on ejection, but that might be exacerbated by Ukrainian locals often finding the ejected pilots before any military force does, and people getting bombed have historically not liked the people flying the bombers much. When a pilot on the ground has a lot of bruises and a snapped neck, it's often hard to identify whether that happened during the ejection, during the landing, or after. And even when the cause was clearly violence, the emergency services might not be overly interested in blaming anyone or anything but the seat.

replies(1): >>45044240 #
43. johndunne ◴[] No.45041275{6}[source]
The forces involved would only pop your shoulders and not provide any support for your back via your underarm. And the speed the bars would need to come out at, might end up inside your chest cavity. The ejection seats are a last ditch save from certain death. I’d take the loss of an inch or two to my height than the alternative.
44. datameta ◴[] No.45041303{5}[source]
I'll be honest I couldn't bring myself to read the whole thing, but it leads me to hoping we never entrust vibe-coders to work on mission-critical systems...
replies(2): >>45041815 #>>45042761 #
45. nobleach ◴[] No.45041321{3}[source]
Seeing tests written by Claude/Gemini really frustrate me sometimes. They'll go to great lengths to make tests pass. Somtimes even just testing console/stdout logging. On the other hand, when I've written failing tests and told the tools to write implementations to make the tests pass, I've had some really good results. Sadly that's the part I enjoy the most so I really don't want to delegate that to a tool.
46. Tuna-Fish ◴[] No.45041356{5}[source]
Many seat types have multiple distinct motors for redundancy reasons. Even if there are not, you can adjust the effective output of a solid rocket motor by modifying the nozzle configuration, and Martin-Baker has an expired patent on this.

... but even if the seat in question was able to adjust the ejection thrust (which I don't think it was), this particular situation involved ejection really close to the ground, which calls for maximum thrust.

47. gregoriol ◴[] No.45041508[source]
The pilot could eject at mostly any time, not if the plane starts spinning uncontrollably though. Also, the plane can become very dangerous to what may be on the ground: if the pilot looses control and must eject it may not always be in a safe place to crash. Or even worse, the plane may continue to fly some distance and crash at a very different place than expected. So yeah, very stressful for everyone involved.
48. benhurmarcel ◴[] No.45041519[source]
Funny thing is that building CRUD apps pays more
49. TheCondor ◴[] No.45041632{3}[source]
I was running late for the last flight home on a Friday a handful of years back. Skipped a late lunch and got to the gate just in time to find out the plane is late too. We boarded the plane but after an hour they unloaded us. Turns out they couldn't open the "fuel flap" and were going to try a "cold reboot" to fix it. If that didn't work they'd fly in another plane from some other city. I was supposed to be home in time for dinner and to tuck my kids in, but I rolled in at 2am. The "reboot" actually worked, it just takes a bit of time. You can't be on the plane when it "reboots" because they turn off the AC and it heats up a lot.

I've wondered about that a lot over the years. I generally don't fly on Fridays after noon anymore, not for work, nobody wants to actually be there at that time. I'm really curious about the process. What kind of system integration has a fuel door system that is so tightly integrated to the rest of the computers that it requries a complete reboot to reset? Was that some mechanic's last idea? Was it from Boeing or Airbus? Is it in the playbook? Is there a more elaborate debug process and they short cutted by rebooting? Is there a fault log that will help fix the actual issue or is it just gone? I dutifully got on the plane that time but I've sort of concluded that in the future, I'd make some calls and just pay too much to spend the night at the airport hotel and fly the next day.

"So here's the thing Mr. F-35 pilot.... the plane needs ctrl-alt-delete but it will drop like a rock from the sky while the computers are booting up... so you need to hold down the power button for 5 seconds, like a solid 5 Mississippis and then all the lights will shut off. Wait another 5 seconds and the press the power button again and it shoooould boot back up... You'll hear a nice pleasant chime and see our logo on the flight control monitor and then after about a minute when the POST completes and it boots up, you should be back in action.." Everyone involved with all of that is just cut from different cloth.

replies(2): >>45042575 #>>45045129 #
50. ◴[] No.45041815{6}[source]
51. hluska ◴[] No.45042125{6}[source]
It’s a massive issue - crashes are investigated and it’s not uncommon for a pilot to lose their jobs if they deviate from procedures. A few years back, a Marine Colonel was relieved of a prestigious command after being found to have ejected too early:

https://www.military.com/daily-news/2024/10/31/pilot-of-f-35...

replies(1): >>45044154 #
52. D-Coder ◴[] No.45042575{4}[source]
> What kind of system integration has a fuel door system that is so tightly integrated to the rest of the computers that it requires a complete reboot to reset?

Need to reset the global variables.

53. emchammer ◴[] No.45042761{6}[source]
Therac-25 is also on the front page right now. I like to reread the reports about that every few years, especially the human effects. Maybe we need formally-verified methods to be more accessible.
54. lordnacho ◴[] No.45042989{5}[source]
This sounds totally plausible to me, a complete novice in this field.

This is why we shouldn't rely on LLMs for this kind of thing.

55. LorenPechtel ◴[] No.45044080{6}[source]
1) It's too fast.

2) Think your shoulders could take it?? It's IIRC ~15g.

I do think they probably could make a gentler ejection mode as the seats are built for the worst case. The lower your airspeed the longer you have to get high enough to clear the tail and thus less acceleration is needed.

56. LorenPechtel ◴[] No.45044154{7}[source]
Note that that was the third review. The first two exonerated him--the real problem was his helmet wigged out. His plane was actually flyable but it was lying to him. And the rule was if the plane is not responding properly to pilot commands below 6,000' AGL eject.
57. impish9208 ◴[] No.45044206[source]
This is one of the reasons why I don’t call myself a software engineer. Very few of us are actually doing anything resembling engineering. I’d have been like, “I can’t reproduce this production bug on my local system,” tagged the issue with ‘needs more info’ and moved on to something else.
58. LorenPechtel ◴[] No.45044240{6}[source]
And note the situation over Ukraine. There are systems like the Patriot--you do not want to be above the horizon to one of them. But so long as you're some distance from the launcher and stay very low the Patriot can't see you. But that puts you very much in danger from stuff like a Stinger and when one finds your engine you're going to go in very, very fast.
59. LorenPechtel ◴[] No.45044268{5}[source]
I believe the only relevant factor is airspeed.
60. LorenPechtel ◴[] No.45044322[source]
Mostly. There comes a point where stupidity exceeds the ability to design around it.
61. nurple ◴[] No.45045129{4}[source]
The 787 had three such bugs, in 2015 it was found out if the plane wasn't restarted once every 248 days (2^31 100ths of a second), the AC generation system would shut off, even mid-flight[0].

In 2016 it was found that if the plane wasn't restarted once every 22 days the 3 flight computers could reboot simultaneously, also in mid-flight[1].

In 2020 it was found that if it wasn't restarted at least every 51 days that the stale data monitoring system, and the stall and overspeed horns all stop operating[2].

Some A350s also had an uptime-dependent fault found in 2019[3].

0: https://www.engadget.com/2015-05-01-boeing-787-dreamliner-so...

1: https://www.pcmag.com/news/boeing-787-dreamliner-bug-fix-req...

2: https://www.theregister.com/2020/04/02/boeing_787_power_cycl...

3: https://www.theregister.com/2019/07/25/a350_power_cycle_soft...

62. HPsquared ◴[] No.45046190{5}[source]
Car airbags are basically that, and they have different levels of deployment (on a much lower budget).
63. qingcharles ◴[] No.45046385{3}[source]
@grok i need to patch an F-35 firmware in flight. just the code, no comments. make it snappy