Hacker News Reader

Comment by verelo | Hacker News Reader

Shockingly common in civilian aviation, it’s actually one of the most common causes for accidents. Not sure about commercial but it was not uncommon in the military in the 50s.

View on HN

They say there are two kinds of pilots: those who have landed gear up, and those who will. I’ve seen it happen with small planes three times.

It should be quite rare on airliners due to having two pilots and good warning systems. It’s plausible that the pilot would forget, but both of them forgetting and never noticing the plane screaming at them is unlikely.

I want to agree but it has happened before, and with the stress of some other failure it’s not going to be impossible…but it should be super rare.

It’s definitely not impossible. Pilots can get fixated. “That noise is so annoying! Never mind, gotta finish this landing.” It’s rare enough that I think it’s not the way to bet here, and there was probably another problem. But we’ll have to wait for more info to know for sure.

> It’s plausible that the pilot would forget, but both of them forgetting and never noticing

Not drawing premature conclusions here without evidence, but Asian airlines and aviation culture have a documented history of the co-pilot not questioning fatal actions by the captain in order to save face.

I know someone who had a no gear low-approach and prop strike, pulled up, got the gear down, and landed straight ahead. Not sure what to call that one, but I bet that pilot won’t do it again.

It's very easy to miss one alarm if there are several others going off simultaneously, which is highly likely if you've lost an engine due to a bird strike. Even if you hear or see the gear warning, you might be too cognitively overloaded to acknowledge and act on it.

the aviation term(probably others) for it is task saturation and loss of situational awareness.

Back in the day, there was a third crew member who was primarily tasked with navigation and monitoring systems.

Airlines decided it would be cheaper to cut that position and here we are.

Manufacturers made planes that didn't need it, making their plane's TOC more attractive.

Aviation is still much safer than back in the days where 3 or even 4 crew in the flight deck were standard. At some point the tradeoff between increased safety and increasing costs becomes unreasonable.

> Airlines decided it would be cheaper to cut that position and here we are.

It's not that easy tbh - the advent of digital monitoring systems, fly-by-wire and glass cockpits plainly eliminated the need for it.

On the other hand EASA is pushing for research into single-pilot operations [1]... now that is nuts.

[1] https://www.easa.europa.eu/en/research-projects/emco-sipo-ex...

"here we are."

Here where precisely? Civil aviation is extremely safe in 2024, and this crash made worldwide news precisely because crashes are now very rare in the developed world.

If you demand 100 per cent safety, you are bound to be disappointed forever. Not even walking is 100 per cent safe.

I’m reading from aviation experts that the fundamental heuristic of “aviate, navigate, communicate” may have been forgotten as a result of panic following the bird strike incident (two minutes prior to the landing attempt). How should this principle have been executed in this situation?

Panic is understandable momentarily, but any professional pilot should be able to overcome that within a few seconds, and remember their training, which in an emergency is something like aviate, identify the problem if possible, and run checklists. Panic doesn't solve anything, and there's a comfort in having checklists and procedures to run. Something must have gone very wrong in training or with the plane for this to happen.

There's a standard short checklist for landing that includes flaps and landing gear. There may also be an emergency landing checklist that would also include those things.

The voice recorder and flight data recorder almost certainly survived. We'll know more after those have been recovered and analyzed.

I don’t want this to come across as second guessing a cockpit that I wasn’t in, but speaking generally, recover the aircraft to a flying condition (aviate), if not on a stabilized approach with high certainty of continuing to be (mostly aviate and some navigate), go around and hold at a safe altitude (A&N) while you run the checklists and assess the aircraft state (A) and tell ATC what your intentions are (communicate).

I don’t know what the exact state in that cockpit was, but the video of the aircraft sliding down the runway sans gear at a speed that looked well higher and well longer than normal touchdown suggests that they didn’t have a stabilized approach at the end, whether for good or bad reasons is something for the investigators to figure out.

the airplane is perfectly fine to continue flying for a while on one engine so their momentary shock should not have been an issue. they've trained extensively for this kind of scenario and should probably have gone around if the bird strike happened on final approach. like the other person said... the voice recorder almost certainly survived and will give more information as to the root cause.

It may have been both engines out and there was not enough time to turn on the APU for landing gear deployment. The belly landing was executed just fine but tragically it was the wall at the end of the runway.

it wasn't the wall that got them actually it was the poorly designed ILS embankment that should have been ground level. but yeah you're right it is somewhat probable that it was two engines out as the evidence comes in

Makes Chesley Sullenberger and Jeffrey Stiles look really good, doesn't it.

Isn’t the point of rigorous pilot training and selection to pick out those who will not be ‘cognitively overloaded’ by a dozen or two alarms at the same time, even under great stress?

Not every country/airline has "rigorous pilot training and selection" -- US airlines require around 5x the number of flight hours as non-US. There are, generally, not enough pilots.

How is this relevant when the country, South Korea, is known?

It’s not a mystery where these pilots were licensed.

It is relevant that the US requires ~6x more flight hours for airline hiring than South Korea, and the accident happened in South Korea, no?

Can you link the source for the flight hour difference?

If true it does have some signal but it’s not entirely persuasive. After all there clearly are midwit pilots too with a lot of flight hours, who just manage to scrape by on each step of the way.

Edit: Who may very well perform worse in extremus than a less experienced genius pilot.

The US FAA has required 1500 flight hours to receive an ATP (airline pilot) certificate since 2010: https://www.ecfr.gov/current/title-14/chapter-I/subchapter-D...

South Korea requires 250 hours as a country (6x), and Jeju Air specifically requires 300 (5x): https://epicflightacademy.com/hiring-requirements-jeju-air/#...)

Are you taking that number out of your arsenal or from a source? AFAIK most South Korean pilots are actually doing their training in the US, since it's much easier to accumulate the hours, and there's more GA. I would think their training is about the same as US pilots.

Although the US airways system is much more developed and used than Korea's, given that Korea is a smaller country and has an extensive bullet-train network. So I could buy an argument where US pilots just fly more.

Most flights in Korea are international flights, so a pilot for Korean air probably flies as much as a pilot in the states. If the airlines need fewer flights, they could just go with fewer pilots rather than the same number of pilots flying less.

In the 1990s Korea and Taiwan had issues with accidents caused by military pilots without modern crew management cultures (“never question the captain”, which is a big no in modern commercial aviation), so they went with more career pilots trained from scratch (at American schools) rather than just transitioning military pilots into the role.

You should probably apologize for the suggestion that I am making up numbers.

The US FAA has required 1500 flight hours to receive an ATP (airline pilot) certificate since 2010: https://www.ecfr.gov/current/title-14/chapter-I/subchapter-D...

South Korea requires 250 hours as a country (6x), and Jeju Air specifically requires 300 (5x): https://epicflightacademy.com/hiring-requirements-jeju-air/#...)

Oh sure, I do apologize, just wanted to know if the numbers came from somewhere or if they were a wild guess. Thanks for bringing up the references.

The difference is drastic indeed. I almost would qualify for Jeju Air.

Flight hours are not a good metric, experience does not imply skill.

See:

- https://youtu.be/o6c3ENr_CRM?t=1731

- https://youtu.be/cUAYQTzXpsg?t=87

Flight hours are increasingly useless. The planes fly themselves enough that there are serious problems when something goes wrong and situational awareness is lacking.

Every single person's mind is going to become overloaded at a certain stress & task complexity threshold. Establishing where exactly that point is for a given person is difficult/impossible - e.g. throwing pilots into genuine life-threatening situations to test their responses doesn't seem ethical.

Of course you don’t set the bar at that exact threshold, you set it higher so even after accounting for difficulties of assessment there would still be a comfortable margin.

> It's very easy to miss one alarm if there are several others going off simultaneously

The GPWS would literally be screaming "TOO LOW, GEAR" at you, over and over again. I find this difficult to believe. Being too distracted or overloaded to respond to it I can believe.

Perhaps there are too many alarms in case of a bird strike in a modern airliner? "Hydraulic pressure LOW","Voltage in System B Out of SPEC", "Cabinets in the kitchen area Open", "and by the way,the landing gear is not down".

One would imagine there are psychology experts at Boeing and others who do nothing else all day ,but decide if one or the other alarm should be prioritised and at which volume (too low and they don't hear it, too loud and it disorients).

It is a complex subject. I think in time we realise removing the third crew member was an error.

Most of those alarms would only be shown as text on the EICAS (Boeing) or ECAM (Airbus). Very few warning systems (the most important, like GPWS, TCAS, RAAS, the engine fire alarms, and the stall warning system) are aural and/or tactile in their annunciation.

EDIT: For a practical example, low hydraulic pressure in the left-hand system in a 777 would be yellow text on the EICAS that says "HYD SYS L".

I don’t think any 737 variant has an EICAS

> quite rare

This right here is literally a super rare incident — hull loss + everyone-2 killed.

Having read quite a few Admiral Cloudberg[1] posts, task saturation seems fairly common among fatal incidents IIRC.

[1] https://admiralcloudberg.medium.com/

It has happened before due to another higher priority warning suppressing the "too low, gear" warning, so if there's a lot going on already including other warnings, it's not inconceivable that it was overlooked and the plane didn't issue a warning depending on the circumstances.

Seems quasi-impossible on a commercial airliner. This thing is going to beep and alert like hell about the landing gears being up. The most likely cause is a complex series of event that led to the landing gears not coming down. Landing gears have multiple redundancies so this is predictably going to be an unfortunately very informative investigation for the aviation sector.

It has happened before: https://en.wikipedia.org/wiki/Pakistan_International_Airline...

If your engine had a bird strike than the system is already beeping and alerting like hell.

The more likely scenario to me is that they attempted a go-around after the bird strike and failed, leading them to land with gears up, no flaps, at high speed.

If the plane can tell it's super low the ground -- low enough to sound an alarm for it -- and that's a common mistake... why not just have the plane automatically deploy the landing gear?

Doing so complicates other hairy scenarios because of the increased hydraulic/power demands, reduced clearance, increased drag on the plane, worse handling, and the fact that in some emergencies you want a gear-up landing and would expect pilots to simply make the opposite error (forgetting about the gear being automatically deployed) some fraction of the time.

Hmm. Are those situations more likely than forgetting to put the gear down? (Serious question -- I don't know the answer, and I'm curious.)

I think so, and they're definitely more severe. Gear-up landings occur in something like 1/150k flights, and they're rarely fatal. Stalls happen in roughly 1/100k, near-stalls more frequently, and if unrecovered then they are almost always fatal.

The one plane I know of with auto-retracting gear had a fatal stall because of the feature, so it's not exactly a theoretical argument, but there haven't been a lot of empirical studies.

You're talking about GA flights, not commercial ones, right? I do not believe that a gear-up landing occurs once a day on average in commercial airliners.

You do not want auto gear down. There are situations where having the gear down might makes much much worse and you might not be able to be in a situation where you can bring them back up.

this is why there are 2 pilots? doesn't add up

More pilots doesn't necessarily help..

https://en.m.wikipedia.org/wiki/Turkish_Airlines_Flight_1951

Not an aviation expert or enthusiast, but I'd imagine in a commercial airliner if the gear was not deployed and the pilot was trying to land, at a certain point the plane would start yelling at the crew something like "NO GEAR" "NO GEAR" "NO GEAR (deployed)"

So I don't think a pilot can just "forget" to deploy the landing gear in a commercial airliner.

TOO LOW GEAR

I think the 777 landing gear warning is based on flap position, eg flaps 0 you won't get a warning, but you will probably get a GPWS configuration warning instead!

> I think the 777 landing gear warning is based on flap position

It isn't. If you're within a thousand feet of terrain, a runway is nearby, and you're at an approach for landing speed, it blares "TOO LOW, GEAR" in the cockpit over and over again. If you're going faster than approach for landing speed or there is no runway, it instead blares "TOO LOW, TERRAIN".

Likewise if you deploy more than flaps 20 without the gear extended (regardless of your height above terrain or the presence of any runway), you get a master warning and "CONFIG GEAR" in red on the EICAS.

Youtube is full of examples pilots landing gear up, with the gear-up warning system clearly blaring in the background.

https://youtu.be/5McECUtM8fw?si=DwasT3T_9vHxLczn

This does not only happen to little propeller airplanes. Heres an Airbus A320 where the pilot managed to land gear up despite the presence of all kinds of safeguards and automation.

https://www.flightglobal.com/safety/pia-a320-crews-fatal-lan...

Humans are funny animals.

https://www.youtube.com/watch?v=gINDzJXpMLU

It sounds like there was an engine out due to a bird strike. The pilot might not have noticed the gear alarm because of the engine alarm.

https://en.m.wikipedia.org/wiki/Pakistan_International_Airli...

Reply on news.ycombinator.com