Flight AA300 Strikes Runway Sign on Departure
American Airlines flight 300 from New York to Los Angeles had a bit of a run-in last week. The aircraft, an Airbus A321-231 (registration N114NN) was on runway 31 at JFK with 109 passengers on board when they struck a runway sign showing 5,000 feet distance remaining.
As the aircraft levelled out at FL200, 20,000 feet, the pilots reported that they needed to return, as they were experiencing “a strong roll to the left”.
Well, yes, that’s not a big surprise. Here’s a photo that’s making the rounds, credited just to “Seat 16A”.
It seems that on rotation, the left wing tip struck the sign, which became embedded in the wing. The ground marks show that the left wing tip dragged along the runway for some distance.
The flight crew contacted ATC as they leveled out at 20,000 feet, saying they’d experienced a strong roll to the left as they climbed out.
AA300: We were banking…Uncontrolled bank 45° to the left.
ATC: Turbulence from another aircraft?
AA300 I don’t think so. There’s a good crosswind, but we had an uncommanded roll to the left as we rotated.
The flight crew reported that the aircraft was fine (“flying great”) but they wanted to get it checked out.
This image from AVHerald is said to give the location of the sign.
The flight landed safely at 21:08 local time, 28 minutes after take-off.
Here’s the METAR:
00:51 UTC / 20:51 local time:
KJFK 110051Z 36017KT 10SM SCT250 10/M03 A2998 RMK AO2 SLP153 T01001028
Key info: At 00:51 UTC (20:51 local time) at JFK Airport, the visibility was good (10 statute miles) and the wind was at 360° blowing 17 knots, so a crosswind component of around thirteen knots.
Does anyone know why American aviation uses statute miles instead of nautical miles for visibility?
The runway in question, 13R-31L, is one of the longest commercial runways in North America: 14,511 feet long by 200 feet (60 metres) wide, about 4,425 metres by 60 metres. The Airbus A321 wingspan is 35.8 metres, 04 117 feet.
A user on PPRuNe did the maths:
An A321 has a wingspan of 35.8 m, according to Wikipedia.
Assume it’s on the centerline of a 61-meter (200-foot) wide runway.
30.5 m, runway centerline to edge – 17.9, half the wingspan of an A321 = 12.6 m (41 feet), distance from wingtip to runway edge.
An A321 should have a clearance of 12.6 m, or 41 feet, from each wingtip to each edge of the runway.
From the runway edge to the sign that was struck looks like another 75 feet or so, just guessing from the image shown. Let’s call it about 22 m.
12.6 m wingtip to runway edge + 22 m runway edge to sign = 34.6 m wingtip to sign.
That would place the sign about 35 m (115 feet) from the wingtip when the aircraft is on centerline.
If all this stuff is correct then that machine must have been quite far off the runway when it hit that sign, about 120 feet laterally displaced on a runway that is 100 feet wide from centerline to edge. Wow!
A person posting on AVHerald claimed that he was on the aircraft and described the experience.
I was aboard this aircraft. The take off was fast, rather quick and felt short. Then we pitched down and banked right (left wing up) and then left (right wing up) and the back felt to skid out sideways, I was in the window seat just behind the left wing. Then it felt like the pilot pulled the aircraft up manually. He continued to make very strong left and right banks while in the air before we circled back to JFK. He made an announcement that we had a major computer failure, but that he had control of the airplane and that we’ll be making an emergency landing. I watched the metal flap (runway sign) above the wing the whole 43 mins we were in the air.
Making an announcement blaming computer problems while the aircraft was banking left and right seems a bit odd, especially in light of the Boeing 737 MAX issues. As the crew said at that point that the aircraft was flying fine, I wonder if they were testing to try to correct the 45° bank. Based on the ATC calls, they didn’t know their wing was broken and carrying a sign, although the passenger claims that it was visible for the full flight. Knowing two planes had nosedived after a series of unexpected manoeuvres, I wonder how many of them feared that it was an MCAS issue, as it’s been all over the news.
Certainly, plenty of people seem to believe that the initial bank or wingdrop was somehow connected to the recent fatal crashes; I’m seeing a lot of accusations that this, also, is the result of too much automation and bad software.
It seems unlikely but we simply don’t know exactly happened. The NTSB is investigating it as an accident. They’ve formed a team of six with support from the FAA, American Airlines, the Allied Pilots Association and the French BEA with Airbus as their technical advisor. The NTSB is not planning to send investigators to the scene, which surprised me, but they seem to feel the data recorders will give them all the information they need. As there are some scrapes on the ground, I’m surprised, though I guess the photographs will tell them what they need to know.
Wow! A strange story indeed.
My first thought was, just like the comment in the blog, that the aircraft must have been substantially off the runway centreline. That can be establised as a fact without these diligent calculations. Were the pilots aware of that? What seems a bit bizarre: 1. Were the pilots aware of this? How could they have failed to notice? 2. The aircraft must have banked very substantially prior to rotation for the wingtip to strike a sign? Crosswind cannot have been a factor. The wind strength should have been very comfortably within the limits. The wingtip even scraped the ground. 3. How about the engine nacelle? Did it hit the runway as well? The wings would have been providing lift, therefore already bent upwards. For them to hit the ground may have caused the engine to drag along the tarmac as well.
This raises very serious questions. Because the crew appeared to have had only limited awareness about what happened. The bank angle of the A320 is limited automatically with weight on the squat switches, this restriction (“law” in Airbus parlance) is removed when the aircraft is flying and the struts can extend under their own weight.
An Airbus came very close to grief when, during a landing in maximum crosswind, the crew decided to take off again. The switches were compressed and the roll was limited, the crew had barely enough control to make it back safely, with some damage.
If there was a technical problem that allowed the aircraft to exceed the bank angle that is deemed safe on the ground, even to the extent that the wingtip picked up a sign from the runway edge, how come did the crew fail to notice this until well after getting airborne?
If they had controllability problems, it seems that it must have happened before V1. That should have been the moment to abandon the take-off or, if it happened after V1, it should have been the moment to declare an emergency at once and return immediately.
So: Was it a crew caught napping? Or a crew that was so incompetent that they failed to notice that they were way off centreline during the take-off run, banking steeply whilst still on the ground – to the extent that the wingtip hit the deck – and yet they blissfully continued the take-off until they realised that the control problems, caused by a damaged wing were not going to go away by themselves. Even then, it seems that they were not really aware that the wing was damaged.
Even if the problem had been due to a technical (computer) fault, the crew’s reaction to the situation put them in good company and make them serious contenders for the Laurel and Hardy award: “A fine mess”.
No, with good company I definitely do NOT mean Sullenberger.
Was the problem caused by a computer “glitch”? Possible, but I doubt it.
And, to recap: No matter what the cause was, the crew’s actions and reactions make me feel uncomfortable because their responses appear to have been, to put it very mildly: inadequate. This could have ended very badly, they were lucky.
A nautical mile is a measurement of distance of air (or water) equal to one degree of latitude, which makes it congruent with celestial navigation based on latitude and longitude. A statute mile is a measurement of land that is congruent with surveying methods, which are relative to a fixed earthly datum, not to celestial objects, called a geodesic marker. Because we refer to objects on the ground to describe visibility (e.g., the end of the runway is three miles from the antenna tower), we use statute miles to maintain fidelity with terrestrial maps.
In “90 minutes at Entebbe” one of the pilots is said to have mistaken the runway edge marker for the centerline. Wonder if that happened here? Dropping a main gear off pavement and into dirt could do that, no? For now, entirely speculation.
You are quite right, the nautical mile is indeed based on a division referring to the division in degrees of latitude and longitude.
The nautical mile (1852 metres) is therefore commonly, by preference, used in navigation.
Since aviation adopted navigational methods borrowed from the seafarers, and indeed some navigators on board of aircraft once they were able to cover longer distances often had worked on board of ships, or had been taught in the same schools. Aircraft in WW2, and later civilian aircraft like the DC4, Constellation, etc., still had a glass dome in the roof behind the cockpit. The navigator could use it to “shoot” the stars with his sextant. I still have one, but it just was dropped as a subject from the curriculum when I studied for my airline pilot’s licence. The air forces took their navigation from the navy. The airline uniforms have been derived directly from those worn by naval officers. Even the title “captain” when referring to a pilot in command comes from the navy. Historically, the commander of a ship in the merchant navy was called “master”.
So generally naval terms and units have been adopted in aviation.
Nevertheless, there are a few exceptions. Countries and (civil) aviation authorities may impose their own standards.
In general, in aviation the unit of measurement for distance is the nautical mile (= 1852 m.).
The unit for height (altltude) is in feet (30.48 cm)
Air pressure is usually (in Europe) given in hectopascals, the same as millibar, only the name has changed.
In the USA, especially in general aviation, the land (statute) mile may be used in lieu of the nautical one. I don’t know it by heart. Look it up, it is about 1.6 km. The airspeed indicator in the Cessna 310 I used to fly (PH-STR, later N444ST) was calibrated in land miles. A bit complicated, because for navigation they mainly use nautical miles.
For altitude some countries, like Russia, use metres. I believe they still do.
Flying to Moscow from Berlin Schoenefeld we first had to go to Copenhagen, then south across the DDR, pick up a Russian navigator in Berlin, who would tell me: “Now climb to flight level faif tousand metres”. Hoh, that was FL 150 but yes, that was the unit over East Germany too.. Same over Hungary, Poland, Bulgaria. When I protested that the 310 was neither pressurised nor carried oxygen, I was told curtly (whilst enjoying a blast of his breath heavily spiced with garlic and Vodka): “You now climb to faif tousant meetres or you see MIG… verry close!”
And, again mainly in the USA, the unit for air pressure may be inches of mercury.
Mel gave us a very good explanation, but once we go into the dark recesses of the explanations and backgrounds it becomes murky very quickly.
No? Then you must be an experienced pilot !
Off centreline from the start of the take-off by mistake? Not all that likely, but the whole sequence of events does not seem to make sense. J may have a point. But in this case, all wheels seem to have remained on the hard runway,
BTW: This A321 has the winglet modification, not the standard triangular fences at the wingtips.
PS: As an addition to J’s comment:
The centreline of the taxiway at major airports is marked with a yellow line.
This taxiway centreline usually will continue right onto the runway and will stop at the take-off point.
Heavy aircraft will have to take a much wider turn onto the runway, but a “medium” can follow the taxi line right onto the runway centre.
So, assuming that JFK has that arrangement, it would seem yet another nail in the coffin of the professionalism of that crew if they missed it when lining up. There was neither a crosswind strong enough to have an impact, nor was the visibilty a factor, or so it seems.
Maybe the pilots should have heeded the adverts and gone to a well-know optician?
Rudy, that does appear to be the arrangement at KJFK. Here’s a satellite view of the end of 31L:
and all the taxiways leading onto the runway have that same standard marking of centreline continuing onto the runway.
The numbers do appear to be broadly consistent with a takeoff along the left edge of the runway in error for the centreline. It looks from that satellite view as though there’s a decent amount of tarmac between runway edge and rough ground, so it might not be immediately noticeable from ground feel.
Yes right, the yellow taxiway centreline that continues right onto the runway is a standard feature. So really all the pilots have to do, when cleared onto the runway for takeoff is to keep it under the nose of the aircraft. As long as they are on the correct taxiway, of course.
A “heavy” needs more room to swing around, so they may have to go over the far side before they can line up. The nosewheel is located a fair distance behind the pilots’ seats so in some cases they may be sitting over the grass. But that would be on the far side of the runway.
It is very difficult to see an excuse for taking off along the runway edge.
Even more so when the take-off took place scraping the runway with a wingtip. I cannot understand how these pilots failed to understand that something seriously was wrong and did not abandon the take-off.
Unless something of a technical nature was the cause. But then, there would have been even more reasont to declare an emergency and return as quickly as possible.
I doubt that these pilots are going to have a distinguished career in the airlines. Even less as bush pilots because they need their wits about them even more; no operations manual to fall back on in every case, these pilots have to be ready to deal with many different eventualities.
Never posted on here before but this one is very strange. So the only options are (as utterly bizarre as they all are!!) :
1. Aircraft way of centre line during take-off run. (and banking or collapsed undercarriage)
2. The sign was on the runway,
3. The sign was already embedded in wing before takeoff. (Only adding that one for the sake of completeness)
All of these need that wingtip to have been dragging the ground at the point where that sign was located at some point.
Can’t wait to see what on Earth happened here !! Obviously it can only be #1, but how have they managed it !?!?! My suggestion is that an Ace pilot couldn’t deliberately achieve this in over 100 attempts!
A nice summary.
-It seems that the aircraft was dangerously close to the edge of the runway.
-The sequence insofar as known from what has been released and described by Sylvia suggests that the take-off was commenced, and continued, along the runway edge.
-Collapsed undercarriage: that can be ruled out, see photos of the damaged wingtip. This, insofar as can be seen, is sitting at the normal height.
-Did the left gear leave the hard surface? It would seem to explain why the left wingtip came so close to the ground that it pickerd up a runway sign.
-If this is the case: It is very, even extremely hard to believe that the aircraft accelerated normally and did not swing off the runway altogether if the left gear was dragging through the turf. It is even more unbelievable that the crew did not notice an unusual bank and did not abandon the take-off.
-Signs posted adjacent to runways and aircraft movement areas are never so high that a large- or medium- sized aircraft will hit it, provided they stay on the centreline. Smaller aircraft, of course, do not have the wingspan to worry about those obstacles.
-Judging from the photo, it appears that the sign hit the wing when the aircraft was already travelling at speed, so likely during the take-off roll. AND banking suffiently to the left to hit a runway sign that under normal circumstances should have been a good distance under it. Look at the photo with the two men standing under the wingtip and you will see how much clearance it has with the ground.
-Once the aircraft gathers speed and the wing provides lift, they even bend further up.
-Did the engine nacelle scrape the runway? I find it hard to believe that it didn’t.
-If the wingtip hit the deck during taker-off, were any runway edge lights damaged?
So the mystery only deepens. Unless there is a solution to this, my guess is that the crew were grossly negligent:
1. Missing the centreline, presumably the centreline of the runway,
2. Hitting the runway sign,
3. Failing to notice that the wingtip came very close, even dangerously close to the ground.
4. Having such an abnormal bank angle (during take-off?) that the wingtip scrapes the ground, failing to abandon the take-off,
5. If the incident happened after V1, failing to declare an emergency,
6. Climbing to FL 200 still apparently blissfully unaware of the cause of a controllability problem.
7. Still failing to declare an emergency.
And why did the cabin crew fail to take notice of what happened and (so it seems) faIled to notify the pilots? Passengers did notice, it seems that the crew did not, not until much later!
American Airlines have a good reputation so what happened here? Unless we hear otherwise, this crew seemed to have lacked proper training.
Well from that report by “someone on board” it seems as though someone in that cockpit _did_ realise things were going wrong and took over control. From a flight crew perspective I am deeply suspicious of what they actually knew about the status of their flight.
What I didn’t know was that all signage and furniture adjacent to a runway has to be affixed using “frangible materials” (according to some ludicrously specific requirements from FAA). Though I imagine this is due to the potential for incoming aircraft to deviate from the tarmac – and NOT those departing!!! BUT this may have just prevented something much worse in this case though : Good job FAA!! (for a change).
Rudy — I’m curious about “-Judging from the photo, it appears that the sign hit the wing when the aircraft was already travelling at speed, so likely during the take-off roll. AND banking suffiently to the left to hit a runway sign that under normal circumstances should have been a good distance under it.” You note that runway signs are low enough that a heavy big enough to have wingtips outside the boundary lines won’t hit them — but how much higher are these wingtips than those on the A321? Sign placement is tricky — designers want them high for visibility, safety engineers want them low for clearance — so I’m wondering whether a medium in level attitude would clip a sign placed to be (just?) clear of a heavy if the medium were on the edge of the runway.
I couldn’t get RogerBW’s link to show enough detail, but the taxi lines are clear on https://www.google.com/maps/place/John+F.+Kennedy+International+Airportfirstname.lastname@example.org,-73.7823017,354m/data=!3m1!1e3!4m5!3m4!1s0x89c26650d5404947:0xec4fb213489f11f0!8m2!3d40.6413111!4d-73.7781391. If I’m reading https://en.wikipedia.org/wiki/File:FAA_JFK_Airport_map_2016.pdf correctly, the probable taxi path starts at Terminal 8 and goes counterclockwise around the outer, turning onto 31L just down-runway from the numbers, so the crew wouldn’t have seen the most obvious indicator that they were off-line — but they would have had to mistake the solid white edge line for the dashed white centerline AND not realize that they were seeing only one of each pair of touchdown markings (of which they would have passed several before rotating). I understand that takeoff of a scheduled jet is complex, but at the very least the pilot flying seems to have been too focused to notice all the signs that the plane was in the wrong place.
I wonder whether the edge lines should be more radically different — perhaps alternating white and red — but such a change would take a long time to agree on, and maybe even longer before pilots were comfortable with it.
Some good points have been posited in the comments. The wingtip underside bears obvious evidence that material (both paint and metal) was was removed by contact. This appears different than tailstrike photos I’ve seen, which have a linear appearance. But I’ve never seen a photo of the result of an airplane component contacting grass or dirt at a shallow angle at speed, in the general direction of travel. I have some background in material finishing and this wingtip underside photo (although I am viewing this on my phone) makes me think it experienced a lot of friction (I.E. grass or dirt), rather than the grinding effect of pavement. Can anyone comment further to this consideration? It seems that the wingtip likely contacted the “soft” ground outside the runway, presumably very near or during the time of the sign collision. And considering that the affected portion of the wingtip seems to be in the middle of the wing chord (I think that’s the right term anyway – the middle of the front-to-back cross section) and also not restricted to the very end of the wing proper, in accordance with earlier posts I would also be surprised if the left engine nacelle was not damaged.
And that center circular area with intact paint – is that apparent depression a design feature or does it look like a buckling-type result of the sign collision? I’ve never noticed a circular indentation there before, whether on an actual A321 or in any photos.
I have long thought that a wingtip contacting the ground or water while at speed would not only be extremely obvious to a flight crew, but could (and perhaps likely would), among other things, cause a substantial change in direction for the plane. I am trying to keep an open mind about the flight crew, but I am surprised at the overall turn of events.
After a little more observation, it does appear that the area in question on the underside of the wingtip is at the base of the transitional vertically-curved area between the wing proper and the winglet, assuming that the outboard edge of the aileron (which also looks like it could have had minimal contact with the ground) is near where the upward transition begins.
If the wingtip contact happened while a portion of main landing gear was on the ground (probably only one wheel), as long as the main gear stayed intact (which appears to be the case) and no struts were depressed too heavily during the wingtip ground contact, the nacelle might have remained unscathed, but by an incredibly small margin. As Rudy pointed out, wings deflect upward when the aircraft is accelerating or flying, which means that nacelle clearance margin decreases or disappears. If the gear was off the ground, the nacelle would be farther above the ground anyway, as long as there was a level or nose-up attitude, which, if there was an a nose-up attitude during the ground contact part of this incident, it probably couldn’t have been very much.
Drawing lines across a head-on photo of an A321 seems to show a space of maybe a few inches of clearance above the ground during simultaneous contact of an outer main wheel and a wingtip, as long as that ground plane (not the aircraft) is very flat, slightly cupped, or is higher where the wingtip makes contact. That doesn’t take into account any wing deflection, but if the plane was not fully off the ground, like at the exact moment of being in rotation and having fully extended landing gear struts but at least the outboard left main tire was in contact with the pavement, I’m thinking the nacelle could possibly have still barely avoided having its underside deformed.
I don’t know the height of standard runway light stanchions, but I’m wondering if they might have also barely escaped being ripped off, somehow just fitting underneath the wing, outboard of the engine nacelle.
I don’t know, maybe I’m just rambling. I find the minutia interesting. At the very least, this sounds like it was quite a flight. This one’a going in my cache for stories to follow up on in the future.
I find the apparent “polishing” of the undersurface extremely puzzling. If made from contact with the runway, or grass, the aircraft would have to have had something like 30 degrees or more of bank. For that to be the case, the contact area would have been further out towards the turn up of the wing tip whereas it appears to be flat and in the same plane as the main part of the wing! It is hard to imagine the huge amount of wing tip deflection that would otherwise be needed to achieve this effect at this stage on takeoff!
Yes, quite a serious angle indeed.
The higher the amount of wing deflection, the farther inboard any contact can end up being.
There is a decently large space of upward transitional curve at the base of the winglet. A comparison of photos from multiple angles shows that the affected area is at least in a reasonable spot for the incident. I believe the end of the ‘flat’ span of the wing is denoted by the forward line at the edge of the aileron. The affected area seems to cover a fairly substantial arc, which is also consistent with a dynamic surface such as grass with underlying dirt.
With the low angle that the incident photo was taken at, the trailing edge of the winglet’s upward transitional curve does make it tricky to accurately plot out what specific section of the upward curve was affected.
My considerations to this extent are primarily 1) the upward curve of the transition and 2) the swept angle (trailing edge) that makes up the rear boundary of the curve. When combined, these can give a misleading indication of the wing’s flat span continuing farther out toward the winglet. I’m also starting to think that the trailing edge was affected too. It doesn’t appear as a clean line. Does the A321 have a supercritical wing, whose cross section is continued all the the way into the winglet? If that trailing edge did experience some of the same as the main area, it could make sense with the overall appearance of the photo.
I hope I’m not sounding argumentative with all this. Im just trying to explain my thought process.
I think one thing that is evident/dawning on a lot of very good and concerned people on here is that this could have been very, very nasty. Mere inches from it. I don’t suppose anyone would care to do the calculations (highly dynamic trigonometry) as to how much more bank that wing needed for it to be a full on ground loop. Citing words above that indicate the underside of the wingtip was polished rather than scratched : it doesn’t seem very far from it to me.
And when I say “nasty” , I mean “NASTY”. This incident wasn’t initially given the gravity that it deserved imo. In fact I initially found it quite funny. I don’t anymore. Does anyone else feel that safety is retreating yet again? i.e. “tendency towards complacency”.
From another site someone mentioned something that hasn’t been mentioned (I think) on here yet : ground effect.
This bit is interesting (especially the last two sentences) :
“Unwanted consequences of wings in ground effect
The “straightening” effect of ground proximity on airflow past a wing causes the stalling angle of the wing to decrease, by as much as four degrees, relative to the longitudinal axis of the airplane. This fact is even more of a factor with sweptwing planforms, since the wingtips move downward close to the runway surface as the craft rotates on takeoff. If the nose is lifted high enough, the tips will stall due to this proximity, with either of two equally-unpleasant consequences: if both tips stall but the rest of the wing is at maximum lift, the resulting imbalance will lift the nose further, resulting in stall of the whole wing; OR if one tip stalls, the other unstalled wing will tip the plane’s wingtip into the ground. This is what caused a catastrophic accident of a Unwanted consequences of wings in ground effect
The “straightening” effect of ground proximity on airflow past a wing causes the stalling angle of the wing to decrease, by as much as four degrees, relative to the longitudinal axis of the airplane. This fact is even more of a factor with sweptwing planforms, since the wingtips move downward close to the runway surface as the craft rotates on takeoff. If the nose is lifted high enough, the tips will stall due to this proximity, with either of two equally-unpleasant consequences: if both tips stall but the rest of the wing is at maximum lift, the resulting imbalance will lift the nose further, resulting in stall of the whole wing; OR if one tip stalls, the other unstalled wing will tip the plane’s wingtip into the ground. This is what caused a catastrophic accident of a Gulfstream G650 during certification testing in April 2011.
So without the debated crosswind acting on an aircraft in GE then it’s identical (apart from the outcome )
Yes, this whole situation could very well have ended horribly! And that’s interesting about ground effect.
Well one thing is for sure : the CVR is gonna be a whole lot more interesting than the FDR!
Someone commented to me that “someone was asleep”. Seems to me that aircraft went into GE close to but not at V1 (which is kinda how it works), but neither PF or PM reacted to it crabbing (under some kinda turbulence) towards a major excursion until it was _nearly_ too late. Can’t wait for that NTSB report into this one. It’s obviously easier to be an “armchair QB” (as I am being) when no one got hurt.
Anybody know when that NTSB report is due ??
I am starting to have a bad feeling about these “winglets”. If you have a crosswind of say 10kts then how much lift does the upwind winglet offer in comparison to the latter downwind winglet (which is being pushed down?) And their turning moment is huge even given their small size (distance from com). Is this why they are starting to put winglets on both upper and lower surfaces?
All as opposed to an aircraft with no winglets : so crosswinds matter less. I might have missed a prior conversation here and please excuse me if that is the case.
13kts crosswind compared to how many knots of forward speed at Vr? That’s not going to change lift by that much.
If the winglets tended to stall unequally, wouldn’t that happen more frequently? and been seen durung test flights?
The captain talked about a computer problem. One potential problem would be the fly-by-wire software switching laws to remove protections: this might mean a strong stick movement (to counter the crosswind) that would normally be modified into an appropriately small control surface response might, via this computer glitch, suddenly become a direct input and create a strong roll, introduce a strong overcompensation by a pilot who hadn’t yet understood what was going on, and might have led to the left wing touching the ground before they caught the plane and continued the ascent.
If that’s true, some engineers are now hard at work trying to recreate this sequence of events in the simulator, to pinpoint the cause and develop a fix.
I was speculating heavily there, Apologies.
Mel’s comment is correct as to the fact that nautical miles are based on latitude/longitude, but incorrect about the measurement. Technically speaking, one nautical mile was historically defined as the distance equivalent to one minute (1/60) of a degree of arc on a circle with the same diameter as the earth. In 1929, the nautical mile was defined by international agreement to be exactly 1,852 meters, a definition the US adopted in 1954 and Britain adopted in 1970.
I realize that this is an unsophisticated comment, but in the wildly impossible case that I would be piloting an Airbus 321 down a runway for takeoff, I would think that one thing uppermost in my mind would be, “am I straight and in the middle of the runway?” If I were far enough off center to clip a sign, surely I would notice?
There was more than one person in the cockpit, yes? And none of them noticed?
A similar thought has gone through my mind. I understand that pilot workload is very high during takeoffs and landings but I cannot imagine them not noticing something so elemental as being centered on the runway, even if they had never flown into or out of JFK previously.
Yes, there should have been both a captain and first officer on the flight deck.
On a side note, after some further reading and looking at some subsequent photos of the damaged wing, that dark object emanating from the main impact point in the lower of the two above photos is evidently a runway edge light stanchion.
Apparently, the NTSB never published a final report, but the FAA dismissed it as “incorrect flight control inputs by the pilot flying during takeoff with a strong crosswind” after analysing the flight recorder data. (You can look this up at http://www.aviationdb.com/Aviation/AircraftQuery.shtm#SUBMIT using 114NN as the “N Number”.)