[HN Gopher] Why was there a wall near runway at S Korea plane cr...
       ___________________________________________________________________
        
       Why was there a wall near runway at S Korea plane crash airport?
        
       Author : vinni2
       Score  : 55 points
       Date   : 2024-12-30 18:22 UTC (6 days ago)
        
 (HTM) web link (www.bbc.com)
 (TXT) w3m dump (www.bbc.com)
        
       | hilux wrote:
       | So many things went wrong, and yet the pilots brought the plane
       | down safely - and then this!? Just infuriating.
        
         | underseacables wrote:
         | They did not bring the plane down safely! The pilot failed to
         | lower the landing gear, or extend the flaps, both of which, for
         | all intensive purposes were technologically possible. Multiple
         | redundant systems for these. I think they completely lost
         | situational awareness, and panicked.
        
           | K0balt wrote:
           | When an engine blows up, it's hard to say what still worked
           | and what didn't. They aren't supposed to, but when turbines
           | come apart, there is often a lot of shrapnel that has a
           | history of taking out multiple systems.
           | 
           | But, it is possible that it was a case of poor crew
           | performance.
           | 
           | In any case, the concrete blockhouse at the end of the runway
           | was unhelpful, and it is also outside of the standard
           | guidance for runway aligned obstructions. In most cases,
           | those antennas would have been on frangible towers, and the
           | crew, at fault or not, as well as the passengers, would have
           | had a decent chance of walking away unharmed.
        
             | f1shy wrote:
             | There are 3 different hydraulic system plus one electric
             | that can be operated out of a battery... anything is
             | possible, but it seems until now they failed to lower the
             | gear.
        
               | underseacables wrote:
               | Something else that may be a factor, a lot of Asian
               | carriers teach their pilots to use auto pilot for
               | landing. American pilots almost universally SOP do not
               | use auto pilot when landing. It's possible the South
               | Korean pilot was using auto pilot to land, forced to do a
               | go around, but wasn't in the habit of manually
               | configuring the plane for landing. That's how the flaps
               | and the gear were both missed. He assumed autopilot etc
               | was handling that.
               | 
               | I think in the cockpit voice recording we are going to be
               | hearing the sirens going off about no landing gear and
               | those guys were just not paying attention.
        
               | K0balt wrote:
               | That would be a tragic void in training and procedure. I
               | hope for the sake of the families involved that it
               | doesn't turn out to be something so avoidable in practice
               | and foreseeable in the carriers operating procedure.
        
               | f1shy wrote:
               | I'm afraid it will be something like that...
        
               | wat10000 wrote:
               | There are numerous past examples of this sort of thing.
               | Automation in aviation is really hard to get right. If
               | the automation can fail then the pilots need to be able
               | to perform whatever it was going to do. If the automation
               | fails rarely then the pilots may not get enough practice.
               | But if the automation normally does a better job than the
               | pilots, there's a tension with letting them get more
               | practice on real flights.
               | 
               | A recent(ish) example is the Asiana crash in SF. They had
               | pretty much perfect conditions for a hand-flown visual
               | approach, but they were out of practice, got behind the
               | airplane, and it snowballed.
               | 
               | There's an excellent lecture about this called Children
               | of the Magenta Line. The magenta line being the flight
               | path or direction indicator on an autopilot, and the
               | discussion is about pilots who constantly reconfigure the
               | autopilot to direct the plane instead of just taking
               | over. https://youtu.be/5ESJH1NLMLs
        
               | robinson-wall wrote:
               | Which aircraft have flaps or gear controlled by an
               | autopilot? I'm just an armchair "Air Crash
               | Investigations" fan, but I've never heard of any aircraft
               | where either flaps and gear would be automatically
               | controlled by their autopilot. Speedbreaks / spoilers are
               | usually armed and moved automatically on landing.
        
               | loeg wrote:
               | There's also a backup manual gear release. But from the
               | degree of control over the airplane demonstrated during
               | landing, it seems likely they had hydraulics.
        
           | f1shy wrote:
           | They were with gear down, the retracted it, as part of the go
           | around...
        
           | kiririn wrote:
           | >for all intensive purposes
           | 
           | Always a funny one, it's for all intents and purposes
        
         | f1shy wrote:
         | Absolutely no. There is the suspicion that they even shut down
         | the good engine after the bird strike.
        
         | wat10000 wrote:
         | Bringing a plane down safely involves doing it in a way that
         | lets you come to a stop before you crash into something that's
         | going to kill everyone. You don't get points for touching a
         | runway at a point where there isn't enough of it left for you.
         | 
         | No commentary on the pilots' role here, we don't know nearly
         | enough to judge. Could be they did a great job and they were
         | just doomed. But the end result is that they didn't get it down
         | safely.
        
       | mcflubbins wrote:
       | I was wondering the same thing and suspected it was some safety
       | feature (better for a plane to smack into said wall instead of
       | crash into some populated area, etc) I had no idea he had to make
       | the approach in the opposite direction.
        
         | Towaway69 wrote:
         | This:
         | 
         | > no idea he had to make the approach in the opposite
         | direction.
         | 
         | So the wall is actually at the _beginning_ of the runway. That
         | wall was never never meant to be at the end of a landing but at
         | the start of landing.
         | 
         | I don't understand why this isn't made clear. Basically the
         | runway was used _against_ the design specifications.
        
           | notimetorelax wrote:
           | Runways are approached from both ends depending on the wind.
        
             | dredmorbius wrote:
             | This depends strongly on the airport, terrain, and
             | variability of winds.
             | 
             | There are airports in which approaches always or very
             | _nearly_ always follow the same profiles given local
             | conditions. SFO, SJC, and SAN would be three examples off
             | the top of my head.
             | 
             | SFO's major approaches are over the bay, opposite
             | approaches would involve rapid descents dictated by
             | mountains near the airport.
             | 
             | SJO and SAN are both limited by proximate downtowns with
             | tall towers. Nominal approach glide paths cut _below_ the
             | rooflines of several structures, and make for some
             | interesting experiences for arriving travellers.
        
               | notimetorelax wrote:
               | You're right. Looking at the charts, it appears that both
               | 01 and 19 can be used - https://aim.koca.go.kr/eaipPub/Pa
               | ckage/2020-07-30/html/eAIP/...
               | 
               | What's noteworthy, there's a note to use extreme caution
               | due to this wall if landing or taking off towards it.
        
           | ra wrote:
           | That's not correct. A runway can be used in either direction,
           | if you look on Google maps you can see the runway at Jeju has
           | markings at both ends including a number (denoting it's
           | compass heading) - both ends are usable.
           | 
           | You always want to land with a headwind and never a tailwind,
           | so ATC will use whichever end is favorable for the current
           | conditions.
           | 
           | In this case, if they attempted to land with a tailwind then
           | the on-heading vector component of wind velocity must be
           | added to the airspeed to get the ground speed... whilst this
           | was a contributing factor to the accident, it's not something
           | to focus on.
           | 
           | There will be a thorough investigation but it will take some
           | time to get answers.
        
             | Towaway69 wrote:
             | Thanks for the clarification :+1:
             | 
             | It should perhaps be pointed in news coverage since I
             | equated "opposite direction" with "wrong direction" - hence
             | my scepticisms about the wall.
        
             | rich_sasha wrote:
             | I read that the opposite direction had a NOTAM exclusion,
             | i.e. was excluded from use. From the professional pilot
             | forum linked a few days ago in a similar thread.
             | 
             | If that's right then OP would be correct in saying, this
             | direction wasn't meant to be used.
        
               | loeg wrote:
               | Ok but in an emergency all bets are off, the opposite
               | direction is better than a crash landing. So you can't
               | just assume 100% of landings are in one direction.
        
           | K0balt wrote:
           | Idk about this particular airport but it is nearly universal
           | that runways are used from both ends. The idea is to land
           | into the wind.
           | 
           | We don't know why the pilot elected to double back instead of
           | go around. There may have been indications of a progressive
           | failure that indicated that course of action, but it does
           | seem hasty. That haste may have caused them to not be able to
           | set up a stabilized, minimum speed approach, and may have
           | contributed to the long touchdown, which certainly was a
           | contributing factor.
           | 
           | Still, a 14 ft high concrete structure within 300M of a
           | runway end is unusual, and does not fit the standard for
           | frangable structures which is the guidance for runway aligned
           | equipment.
        
           | ajmurmann wrote:
           | Was the runway designed to only be used one way or was this
           | just the it opposite direction of how it was being used at
           | that moment? I understand that at least some airports change
           | the direction based on wind.
        
           | weweersdfsd wrote:
           | Even if the runway was only used from one direction (not
           | true), it would be dumb to build a big concrete structure
           | near its beginning. It's not unheard of for planes to come in
           | too low and touch down before start of the runway due to
           | pilot error (or even double engine failure on rare
           | occasions).
        
           | lutusp wrote:
           | > So the wall is actually at the beginning of the runway.
           | That wall was never never meant to be at the end of a landing
           | but at the start of landing.
           | 
           | Airports like this are designed to have two approach
           | directions -- in this case, 10 and 190 degrees. Either
           | approach direction would have been acceptable depending on
           | the prevailing wind.
        
         | gazchop wrote:
         | They already botched a gear down landing, which is almost never
         | mentioned. They retracted the gear and did a teardrop go around
         | from a headwind into a tailwind belly flop.
         | 
         | Stinks of bad crew resource management and ATC which is why the
         | ATC and airline for raided by SK officials.
        
           | fl7305 wrote:
           | What should ATC have done differently?
        
             | K0balt wrote:
             | lol.
             | 
             | People often have an idea that ATC actually controls what
             | happens. They just give advisory guidance to pilots, who
             | ultimately decide what to do. A clearance to land or the
             | lack of one does not absolve the pilots from making their
             | own judgments and decisions about how to conduct the
             | navigation of the aircraft, and where and when to land.
             | 
             | Usually, it's a bad idea to not follow ATC guidance, but in
             | the case of emergencies especially, pilots call the shots.
        
             | gazchop wrote:
             | Possible comms failure. ATC are responsible for reporting
             | surface wind. It may have lead to a bad decision by the
             | pilots. Go around versus teardrop etc.
        
           | K0balt wrote:
           | We don't know why the pilot elected to double back instead of
           | go around. There may have been indications of a progressive
           | failure that indicated that course of action, but it does
           | seem hasty. That haste may have caused them to not be able to
           | set up a stabilized, minimum speed approach, and may have
           | contributed to the long touchdown, which certainly was a
           | contributing factor.
           | 
           | If there were significant winds it would have compounded
           | those factors.
           | 
           | It is curious that the gear was retracted. I can only think
           | that this was due to some kind of system failure? Perhaps
           | that same failure explains the decision to double back
           | instead of going around?
           | 
           | Lots of questions, hopefully there will be answers.
           | 
           | Still, the structure does not seem to meet the standard for
           | frangibility that is indicated for objects in the approach
           | path within 300m, although it's not like it was at the very
           | end of the runway.
           | 
           | Runway over/undershoots are actually quite common, and the
           | building of a nonfrangible structure on an otherwise safe
           | skid zone is a significant error in design principles that is
           | not common or conformal to industry standards.
           | 
           | If those antennas had been placed on property designed towers
           | instead of a concrete bunker, the passengers and crew very
           | well may have walked away without a scratch, despite any
           | errors on the part of the crew or procedures of the airline.
        
             | gazchop wrote:
             | They retracted the gear after the first landing attempt. I
             | suspect they either missed it on the teardrop or had
             | secondary hydraulic failure and no time to do a gravity
             | drop. I would err on the side of crew error because there
             | were clear signs the hydraulic systems were functioning
             | (thrust reverser and that they could retract the gear in
             | the first place). Hydraulics don't fail instantly and one
             | engine was spooling still on landing.
             | 
             | That's why EASA says put the plane down if there's a strike
             | on approach. Ryanair 4102 is a good example of a close one
             | there as a reference.
        
             | loeg wrote:
             | They declared mayday and then were on the ground in like 3
             | minutes. I think they probably just forgot gear given how
             | rushed the landing was. We'll find out from the
             | investigation.
        
       | Mistletoe wrote:
       | I don't know why people keep fixating on this. Airplanes can't
       | skid out of the airport into the surrounding city. Mistakes were
       | made, but this isn't one of them. I suspect it is people trying
       | to deflect blame from pilot error, which seems by far the most
       | likely issue. They did none of the things you should do to stop a
       | plane.
        
         | looseyesterday wrote:
         | You are probably right on pilot error, dont forget Boeing
         | probably want this to be the story as well!
        
         | cenamus wrote:
         | But it wasn't there for this reason, and if it was, I'm sure
         | they would have had more than 250m of space to put it at the
         | far ends of the airport.
        
         | cm2187 wrote:
         | Like doing what things? What I read so far is that they suspect
         | the pilot had control issues.
         | 
         | I don't think people who say that it's a bad idea to have a
         | concrete wall at the end of the runway argue the plane should
         | make its way to a nearby motorway. I think most refer to using
         | EMAS, ie a crushable concrete floor in which the plane sinks
         | and stops.
        
           | Towaway69 wrote:
           | > bad idea to have a concrete wall at the end of the runway
           | 
           | but was it the _end_ of the runway? As I understand, the
           | pilot came in from the opposite direction, i.e.
           | 
           | > The pilot then aborted the original landing and requested
           | permission to land from the opposite direction.[1]
           | 
           | So that wall was located at the beginning of the runway _if_
           | the runway was used correctly.
           | 
           | From the bottom image[2], it would appear the wall is located
           | behind the point where planes begin their take-off (and I
           | assume their landing) - but I'm no aviation expert.
           | 
           | [1]: https://www.bbc.com/news/articles/ckgzprprlyeo [2]: http
           | s://ichef.bbci.co.uk/news/1536/cpsprodpb/9090/live/ab9db...
        
             | fl7305 wrote:
             | > So that wall was located at the beginning of the runway
             | if the runway was used correctly.
             | 
             | Most runways are intended to be used in both directions
             | depending on the wind. This one doesn't seem to be an
             | exception?
        
               | Towaway69 wrote:
               | Yep mea culpa, I now understand a little more about
               | aviation!
        
         | HPsquared wrote:
         | Looking at the map, there isn't much beyond the runway.
        
         | csomar wrote:
         | There is nothing beyond the embankment. Airports are generally
         | made in the middle of nowhere. And no, they are not "skidding"
         | into the surrounding city, he probably needed a few hundred
         | meters at most.
        
           | Mistletoe wrote:
           | He's going extremely fast, a few hundred meters would do
           | nothing. I think the estimates I saw were over 150 knots.
           | That's about 77 meters per second.
           | 
           | I found this comment helpful from a Reddit thread.
           | 
           | >The embankment is there to protect the road from the
           | jetblast of departing aircraft in oposite runway direction.
           | Thats why it is allowed directly in the safety area.
        
           | psychoslave wrote:
           | Maybe not all completely aligned straight forward with the
           | landing, but it looks like yes there are some inhabited zones
           | over there surrounded by wooded parcels, well before the
           | landscape change for some sea.
           | 
           | https://maps.app.goo.gl/BN15aSQ1pW6vJkzb7
        
             | csomar wrote:
             | 740m before hitting the next structure.
        
       | gazchop wrote:
       | I think this is disingenuous reporting.
       | 
       | It's too early to actually draw a conclusion. We should wait for
       | the full investigation.
       | 
       | There are a large number of compounding problems here. That was
       | just the last one.
        
         | copperx wrote:
         | Even with all of the compounded problems, the flight could have
         | been non fatal in most world airports.
        
           | gazchop wrote:
           | I wouldn't state that with any certainty. They touched down
           | 1500m down the runway with little to no braking or reverse
           | thrust. The ending may have been less violent but the outcome
           | may have been the same. I'd rather go from blunt force than
           | being burned alive if a wing ditched and it rolled.
           | 
           | Let's wait for the investigators. It's not good to fixate on
           | this outcome.
        
             | ckw wrote:
             | This is the landing of United Airlines Flight 232 in Sioux
             | City, Iowa in 1989. It hit the runway very hard, tumbled,
             | broke, left the runway, and burned, but encountered no
             | significant non-frangible obstacle. Of 296 passengers and
             | crew on board, 184 (62%) survived. It is reasonable to
             | assume that given the absence of other significant non-
             | frangible obstacles beyond the ILS mound that loss of life
             | would have been low and potentially zero.
             | 
             | https://youtu.be/sWkU6HRcOY0
        
         | thekevan wrote:
         | "... the significance of the concrete wall's location about
         | 250m (820ft) off the end of the runway."
         | 
         | "...the runway design "absolutely (did) not" meet industry best
         | practices, which preclude any hard structure within at least
         | 300m (984ft) of the end of the runway."
         | 
         | "it emerged that remarks in Muan International Airport's
         | operating manual, uploaded early in 2024, said the concrete
         | embankment was too close to the end of the runway."
         | 
         | I mean that was a pretty obvious design flaw that went against
         | common standards. I agree it isn't cut and dry yet but an
         | investigation isn't going to change the above info.
        
           | gazchop wrote:
           | I don't disagree with that. It is a compounding issue but I
           | am much more interested in what happened up to that point.
           | Many many small things or one big thing went wrong for it to
           | even get to that point.
        
             | thekevan wrote:
             | But don't forget, that point you mention is the point where
             | the people were killed.
        
               | gazchop wrote:
               | There is a whole chain of causality. They were killed by
               | multiple successive problems compounding.
               | 
               | See Swiss cheese model.
        
               | thekevan wrote:
               | I'm really not sure the chain of bird strike, belly
               | landing, mid runway landing and fiery explosion are equal
               | parts of that chain. One seems to weigh heavier than the
               | rest.
        
               | gazchop wrote:
               | There's a few more you're missing there which is the
               | point of the investigation.
        
           | rich_sasha wrote:
           | So it would appear that this structure would be fully
           | compliant if placed 50m further. That's less than a second's
           | difference. The plane would crash into it at the same speed,
           | just a tiny bit later.
           | 
           | OP is 100% right that many, many things must have gone wrong
           | for the position of this structure to remotely matter (human
           | or mechanical errors).
        
       | tedunangst wrote:
       | How much difference would an additional 50m have made?
        
         | baq wrote:
         | To be brutally honest, the plane would crash about a second
         | later, so approximately none whatsoever.
         | 
         | But this isn't the right question to ask, as it isn't the right
         | question to ask why the wall was there. Why did the pilot and
         | ATC decide it's a good idea to attempt a gear up landing in the
         | wrong direction is a good start.
        
           | EliRivers wrote:
           | I recall back in 2020, when Pakistan International flight
           | 8303 belly landed at Karachi, slid down the runway, and then
           | took off again and had a go at going around, the
           | investigation showed that between them, the pilots just
           | screwed up on having the landing gear down and had a go at
           | landing without it for no other reason than they fumbled it.
        
             | loeg wrote:
             | The PIA pilots totally fucked up the approach, missed that
             | they were very high late in the approach, dive bombed down
             | to the runway at the last minute instead of going around.
             | CRM issues -- the senior pilot plausibly couldn't actually
             | fly (he had a history of bad approaches with unsafe
             | descents) and the first officer failed to raise major
             | problems or push back at all on unsafe decisions. When they
             | first touched down gear up, they had dual stick inputs (one
             | pilot was pushing down, one was pulling up) -- this is a
             | huge no no. Communication in the cockpit was awful. Just
             | lots of awful.
             | 
             | After the incident, PIA pilots were audited and determined
             | like 1/3 had fake or suspicious pilot licenses(!!!). Lots
             | of paying other people to pass tests for you.
             | Internationally, PIA has been banned from landing by like
             | all first world countries.
        
         | __m wrote:
         | Compared to running into a concrete wall probably a lot
        
         | SideburnsOfDoom wrote:
         | 50m of runway? Not much difference
         | 
         | 50m of arrestor bed? Some difference, probably meaningful.
         | 
         | https://en.wikipedia.org/wiki/Arrestor_bed
        
         | loeg wrote:
         | They'd probably have another 1000m of relatively clear ground
         | without the localizer mound. The brick fence probably wouldn't
         | significantly slow the aircraft, then you have a relatively
         | clear easement (with ILS approach towers, hopefully frangible)
         | until a parking lot and trees on the south coast.
        
       | baq wrote:
       | The title question is like the 20th in order of importance why
       | this crash happened...
        
       | Jabrov wrote:
       | Runway's gotta end somewhere
        
         | vasco wrote:
         | Technically it could loop around the earth.
        
           | echoangle wrote:
           | You don't even have to make it that long, you can just make a
           | large circle of runway:
           | 
           | https://en.wikipedia.org/wiki/Endless_runway
        
       | ge96 wrote:
       | I was wondering what if they tried to turn whether by rudder or
       | thrust differential would the outcome have been different/worse.
       | Maybe you can't do much at that speed and so little room.
        
         | lutusp wrote:
         | I'm a pilot. The airplane was sliding on the ground and the
         | landing gear was not deployed. Too fast to stop but not fast
         | enough to use the rudder for directional control. There was no
         | realistic chance to change direction.
         | 
         | If there had been enough engine power to control direction on
         | the ground, there might also have been enough power to remain
         | airborne, but based on limited information, that wasn't so.
         | Under the circumstances the pilots would have wanted to stay
         | airborne to buy time for a more controlled descent, were that
         | possible.
         | 
         | All these speculations are preliminary and may completely
         | change once the black box information is released.
        
           | rogerrogerr wrote:
           | > not fast enough to use the rudder for directional control.
           | 
           | Sure about that? 160kt (how fast someone calculated it went
           | off the end of the runway) is way above Vs1 for a 737, there
           | should be plenty of rudder authority. Heck, Vapp is usually
           | in the 130-150kt range.
        
       | gsf_emergency wrote:
       | Korean police have raided the airport offices
       | 
       | https://youtu.be/3D9BDIH553U
        
       | psychoslave wrote:
       | Just a message thinking for anyone reading who might have had
       | some relative in this plane: sincere condolences.
        
       | Yeul wrote:
       | I have been to airports whose runway end in the ocean, a swamp, a
       | mountain and a 5 lane highway.
        
       | dominicrose wrote:
       | A plane usually crashes because of multiple reasons. The fact
       | that runway design was one of them is a big deal because it was a
       | concern for all airplanes landing there not just one of them.
        
       | kelsolaar wrote:
       | Ate Chuet made a quick analysis about the crash:
       | https://youtu.be/xUllPqirRTI. The wall is there because that area
       | is regularly flooded, it serves for the ILS system, and it is
       | unfortunately over the minimum legal distance for such an object.
        
       | blueflow wrote:
       | I'll recite an avherald comment:                 If you look at
       | "Video of aircraft after touchdown sliding along the runway and
       | impacting the fence:", you will find out that it took them ~1.7
       | sec from leaving the tarmac until they hit the construction. If
       | you measure the distance on Google earth you come up with ~140m.
       | That means they hit the construction with roughly 296km/h or 160
       | knots. If it wasn't the construction it would have been the
       | treeline or something else. That plane was doomed, concrete
       | construction or not.
        
         | loeg wrote:
         | Maybe. But maybe another 1000m of dirt would have been enough
         | to slow them before the treeline. The area south of the runway
         | is mostly an easement for the ILS approach equipment, then a
         | parking lot and finally some trees.
         | 
         | It's also definitely the case that the cement-reinforced dirt
         | mound is not best practice for a locator array.
        
         | mmooss wrote:
         | > it took them ~1.7 sec from leaving the tarmac until they hit
         | the construction. If you measure the distance on Google earth
         | you come up with ~140m. That means they hit the construction
         | with roughly 296km/h or 160 knots.
         | 
         | (Assuming the math is correct:) That's the average speed over
         | that distance. The plane would have been slowing down the whole
         | time.
        
           | formerly_proven wrote:
           | I'm not a pilot.
           | 
           | In the video it looked like the plane was only running on the
           | rear landing gears, I assume with no brakes applied, since
           | that would've caused it to violently pitch down I assume.
           | Only in the last bit did it pitch down and started scraping
           | along the runway. It certainly doesn't _look_ like it was
           | efficiently shedding speed (but looks can be deceiving).
        
       ___________________________________________________________________
       (page generated 2025-01-05 23:00 UTC)