Air India AI171 Crash: Boeing 787 Crashes on Takeoff from Ahmedabad

[Famine]

Minister says black box found at crash site as families await answers

All but one of the 242 people on board the London-bound flight died when it crashed into a residential area in Ahmedabad shortly after take-off.

www.bbc.co.uk

Very early days, but thus far:

• AI171 took off from Ahmedabad for London Gatwick, gained 600ft, signalled a mayday, then came down into a residential area:
  
  
  
• 242 on-board (including crew), largely Indian then British, almost zero likelihood of survivors, probable casualties on the ground.
• Among the buildings impacted was a hostel for doctors.
• First fatal crash for 787, first hull loss for 787, first loss-of-life incident for Air India (other than subsidiary Air India Express) in 40 years - and that was terrorism.

 

[Dreadmed]

If anyone on that jet survived it would be a miracle for the ages, the verified video the BBC have acquired of its descent is nothing short of harrowing... the smallest of mercies is that it didn't crash into an even more populated area.

 

[Famine]

the verified video the BBC have acquired of its descent is nothing short of harrowing...

Aviation experts have been saying that the video shows that the wing flaps aren't in take-off configuration - basically the profile is too clean and the wings aren't generating enough lift at the slow take-off speeds - and there's a whole load of stuff that should ensure that this never happens, so the investigation should be quite something.

 

[Keef]

I spend an awful lot of time that close and closer to airports with jets taking off and landing and that jet kiiiiinda sounds like its engines aren't on. 787s aren't excrutiating but they still like a bigass jet taking off, it's kinda hard to miss and you certainly wouldn't want to live there if you didn't have to. The quality of that video is poor but the volume of the plane going by isn't any more than what you'd expect just before landing at a low power setting, and while its AOA appears similar to takeoff, its flight path looks similar to landing.

The 787 has a considerable amount of power, it's absolutely capable of taking off and climbing with no flaps. Takeoff flap settings are meant to shortern the ground distance and increase the immediate climb performance over obstacles in the safety area at the ends of the runways. We begin retracting flaps at 400 feet in most SOP and at that point these airplanes are travelling probably 160+ knots and reducing their thrust setting a couple % down to "climb thrust" which is still basically full thrust per the computer. I don't think there's any way a proper thrust setting or a mistaken flap setting would be outside the safety margin of climb performance for this plane, but it presumably was very heavy so I could be wrong. But the plane was clearly already in the air, well above 400 feet, when it decided to come back down.

I'm more interested in the volume of those two engines which should've been at near full chat at that point.

Edit: This type of journalism should be illegal. Consult experts before using words with extreme implications.


https://www.bbc.com/news/live/c8d1r3m8z92t?post=asset:e1eeabf7-6407-4fee-a91e-b11149d998de#post

"Sudden" climb to 625 feet and 174 knots? How about "normal"? Unless every person who rides in a plane things that the takeoff is some sudden miracle that can't be explained (I think most of these idiots probably do).

I've never even seen the cockpit of a 787 and I already told you all what would happen and the data shows it. Jet airplanes are jet airplanes, SOPs are SOPs and for the most part across the globe they vary very little for safety and standardization purposes. The plane was indeed above the initial flap retraction altitude and it was indeed above 160 kt cooking right along. Except it stopped cooking and with heavy speculation it sounds to me like - literally because there is no noise in that video - that its grills ran out of propane for some reason.

 

[Liquid]

My first thought upon seeing the video was a stall - whether an aerodynamic stall or an engine failure, through whom- or whatever's fault, remains to be seen - but the captain apparently contacting the tower with a mayday call suggests the flight crew knew something was wrong almost immediately, given how quickly the plane crashed.

 

[Famine]

Edit: This type of journalism should be illegal. Consult experts before using words with extreme implications.

"Sudden" climb to 625 feet and 174 knots? How about "normal"? Unless every person who rides in a plane things that the takeoff is some sudden miracle that can't be explained (I think most of these idiots probably do).

They're basically quoting what they see on FlightRadar24, without grasping that planes sometimes do that on FR24/ADS-B 'cos the signal bounces about quite a bit.

They'd have crapped themselves at this, which was a trace of a plane (with electronic warfare simulation equipment on-board) that flew over my house (20 miles away and on the coast) 30 seconds before I checked to see what it was on ADS-B:

 

[Fezzik]

British passenger in seat 11A survives India plane crash, reports say

The London-bound Air India plane carrying 242 people crashed shortly after take-off in western India.

www.bbc.co.uk

 

[Famine]

Some of the images show chunks of it intact, and it looks like it's just belly-flopped onto these buildings; the shot of the tail section (sans tail, which broke off and is in the street) is partly in/on the top level.

 

[Fezzik]

There’s a new video of the end of takeoff and entire short flight from airport CCTV. It’s weirdly undramatic, just a slow straight descent almost immediately after it gets in the air.

 

[Keef]

There’s a new video of the end of takeoff and entire short flight from airport CCTV. It’s weirdly undramatic, just a slow straight descent almost immediately after it gets in the air.

Thanks for the link, I found it on Instagram:



Really hard to tell where the runway ends but the cloud of dust looks very similar to times that airplanes haven't lifted off until the very end of the runway, blowing dirt and dust that usually is undisturbed. That airport has a very long runway, however the full length requires a taxiback and turnaround at the northern end. While it's common in the US to depart from a taxiway intersection rather than the full length of the runway, typically this sacrifice is only a few hundred feet, not a few thousand, and crews will have either briefed it already as they know their takeoff length requirement, or be briefed by ATC and then make the decision whether or not to accept the intersection departure. But I highly doubt their SOP allows them to accept an intersection departure from literally half of the runway length.

Something about that climb and descent looks like a thrust issue to me. I can't think of anything right now that wouldn't set off alarm bells in a Bombardier and that 787 is even more automated with a very integrated EICAS message system.

There have been instances before of jets attempting takeoff without flaps, however if I remember right in those cases the airplane rotated and attempted to climb at a normal flap speed except without the flaps which simply won't work. Modern computers calculate these differences. This plane did achieve rotation speed and an initial climb but then just stopped climbing.

I also note that the gear was still down and doesn't appear to be transiting - I assume the 787 still uses the manual "positive rate, gear up" callout and action, and that happens within seconds of liftoff, literally as soon as the FO observes a positive climb rate. But while the landing gear being down does increase drag and slow the climb, it doesn't reverse the climb, not even close. Most transport jets are certified to fly at 250kt and 18,000 feet with the gear down in case of a maintenance issue, and can obviously takeoff and climb with plenty of power despite the gear being locked down. It could be that the pilots identified a problem and chose to keep the gear down momentarily, or simply skipped that step due to dealing with another problem which is totally reasonable.

V1 and rotation speed on these large planes are relatively far apart usually, multiple seconds will pass between the FO calling out "V1" (continue/abort speed) and "rotate". If a serious problem occurs after V1, the takeoff must continue and chances are the plane will have at least one engine at takeoff thrust to continue the climb. If an engine problem happens after V1 the plane will eat up a lot of runway at that speed before lifting off, but again this safety margin is calculated for every takeoff and offers plenty to spare.

So even if the pilots were idiots and took off halfway down the runway at an intersection with enough TOFL left, and even if they had the wrong flap setting, and even with the gear down, and even with only one engine, that plane should still be powerful enough to lift off and climb, and it did, we can see it. But then it stopped and that's most certainly not normal.

 

[GT_brickhole]

total speculation, but could it be an air density issue? could climate conditions not allowed the assumed amount of lift?? a wet bulb type of thing...

 

[Blitz24]

Thanks for the link, I found it on Instagram:



Really hard to tell where the runway ends but the cloud of dust looks very similar to times that airplanes haven't lifted off until the very end of the runway, blowing dirt and dust that usually is undisturbed. That airport has a very long runway, however the full length requires a taxiback and turnaround at the northern end. While it's common in the US to depart from a taxiway intersection rather than the full length of the runway, typically this sacrifice is only a few hundred feet, not a few thousand, and crews will have either briefed it already as they know their takeoff length requirement, or be briefed by ATC and then make the decision whether or not to accept the intersection departure. But I highly doubt their SOP allows them to accept an intersection departure from literally half of the runway length.

Something about that climb and descent looks like a thrust issue to me. I can't think of anything right now that wouldn't set off alarm bells in a Bombardier and that 787 is even more automated with a very integrated EICAS message system.

There have been instances before of jets attempting takeoff without flaps, however if I remember right in those cases the airplane rotated and attempted to climb at a normal flap speed except without the flaps which simply won't work. Modern computers calculate these differences. This plane did achieve rotation speed and an initial climb but then just stopped climbing.

I also note that the gear was still down and doesn't appear to be transiting - I assume the 787 still uses the manual "positive rate, gear up" callout and action, and that happens within seconds of liftoff, literally as soon as the FO observes a positive climb rate. But while the landing gear being down does increase drag and slow the climb, it doesn't reverse the climb, not even close. Most transport jets are certified to fly at 250kt and 18,000 feet with the gear down in case of a maintenance issue, and can obviously takeoff and climb with plenty of power despite the gear being locked down.

Could that have been some type of stall? Definitely wasn't an overspeed.

 

[Exorcet]

Could that have been some type of stall? Definitely wasn't an overspeed.

It looks like a loss of engines. The plane doesn't appear to stall, which is when it exceeds the angle of attack of maximum lift. The right wing looks like it dips very briefly, it may have momentarily exceeded max AoA, but not for any significant amount of time. They just lost the ability to climb or maintain altitude and became pegged at the AoA limit until they hit the ground. My guess from the footage.

Very unfortunate.

 

[Keef]

total speculation, but could it be an air density issue? could climate conditions not allowed the assumed amount of lift?? a wet bulb type of thing...

The plane's computer already calculates that. Pilots manually input a few pieces of weather data and the computer uses that. Pilot should verify what it's come up with based on the list of certified takeoff data that they get from a different source, either ACARS or in my company's case Foreflight's data calculation. They could've done it wrong, yeah, and they could've had bad data, and they could have forgotten to crosscheck it. In this case a temperature error could be responsible for a massive lack of thrust. For example if the temperature was 40c but they typed in 4c the computer would think the air was much more dense than it is. This would've increased takeoff roll considerably but wouldn't necessarily cause a flag if the distances were still doable. And again, the plane did achieve flight but then lost it which tells me a thrust issue may have existed but was actively getting worse. Thrust setting should have changed at around 400 feet and the plane's wingspan is just under 200 feet so it's plausible the thrust setting lowering automatically was why the climb was halted. Typically a climb thrust setting is just a couple percent lower than a takeoff setting.

Could that have been some type of stall? Definitely wasn't an overspeed.

I don't see any signs of aerodynamic stall. The wing dip doesn't stand out. Pretty sure the 787 also has anti-stall avionics like Airbuses as well.

One of my biggest questions is why the gear doesn't appear to move at all. It's typical in Western SOP to raise the gear within seconds after liftoff, after a positive rate of climb, and the idea is that if you do lose an engine during initial climb you're already reducing drag as quickly as possible. I watch these planes take off all the time, their gear should definitely be in transit by that altitude although they do move pretty slow.

 

[GT_brickhole]

@Keef thanks for the insight. i was totally thinking mainly of lift to weight and didn't even think of it not being able to generate thrust as well. to me there are a lot of seemingly counter intuative physics in airplanes happening till someone explains the why...

 

[Blitz24]

Would it have made any difference if the crew was trying to pull a Gimli glider if the landing gear was up or down? I think the answer is you want it up but I am not positive.

 

[Eunos_Cosmo]

Stupid question - did they put fuel in the plane?

edit: The fireball would tend to imply yes...

Complete electronics failure? Double bird strike? Clearly the plane lost thrust completely....possibly before it left the ground.

Apparently a survivor managed to jump out of the plane before it impacted the ground. Which is wild.

 

[Keef]

@Keef thanks for the insight. i was totally thinking mainly of lift to weight and didn't even think of it not being able to generate thrust as well. to me there are a lot of seemingly counter intuative physics in airplanes happening till someone explains the why...

Yeah don't forget the engines are just air compressors with fire inside. Pretty much everything about flight performance can be explained by Bernoulli's Principle and the Ideal Gas Law.

Would it have made any difference if the crew was trying to pull a Gimli glider if the landing gear was up or down? I think the answer is you want it up but I am not positive.

Give their landing option was into buildings I'm not sure that was a factor lol. Otherwise that decision depends on a lot of factors, but lower drag equals better glide if that's what you're asking.

 

[Liquid]

Apparently a survivor managed to jump out of the plane before it impacted the ground. Which is wild.

Does that mean he opened the emergency exit after knowing/seeing the plane faltering? That's both wickedly brave and outstandingly ballsy.

 

[Joey D]

One of my biggest questions is why the gear doesn't appear to move at all. It's typical in Western SOP to raise the gear within seconds after liftoff, after a positive rate of climb, and the idea is that if you do lose an engine during initial climb you're already reducing drag as quickly as possible. I watch these planes take off all the time, their gear should definitely be in transit by that altitude although they do move pretty slow.

Fighting with whatever issue was happening and just didn't get to it? Which would lead me to believe the plane had issues before they pulled back on the stick, or moments after the plane went nose up.

Seems like if they input variables incorrectly in the computer, they'd still be able to throttle up. The 787 has some serious thrust behind it so I would imagine they could pull out of it, unless the computer won't let them override. I don't know nearly enough about airplanes to say if this is even a reasonable take though.

 

[Omnis]

Were any Iranian scientists on board?

Kind of worried about this one considering it's the 787. Like Keef said, a lot of stuff doesn't make sense and it's a whole cascade of things that had to go wrong. Pretty crazy for 2025

 

[MisterWaffles]

I didn’t really want to watch the clip but I did anyways because it was added to the Wikipedia article on the crash, and it’s almost completely unbelievable watching it happen.

It goes up very briefly like a normal takeoff, doesn’t change pitch or roll, and just goes down slowly and explodes. No nose dive, no uncontrollable spiral into the ground. It’s just like the plane started producing barely any thrust randomly a few seconds after takeoff.

I hope now that they have the black box that the cause comes out quickly. I’m morbidly curious about how this possibly could have happened.

It wasn’t even that long ago I was going to Amsterdam and checking whether I was going to get the Airbus A330 or Boeing 787. And I thought hey at least the Boeing 787 has a perfect record so far. Not anymore, and in the deadliest way imaginable no less. Insane.

 

[eran0004]

One of my biggest questions is why the gear doesn't appear to move at all. It's typical in Western SOP to raise the gear within seconds after liftoff, after a positive rate of climb, and the idea is that if you do lose an engine during initial climb you're already reducing drag as quickly as possible. I watch these planes take off all the time, their gear should definitely be in transit by that altitude although they do move pretty slow.

One theory I've seen mentioned is that the copilot might have accidentally retracted the flaps instead of the gear. Not only would that have caused the loss of lift, but it would also have confused the pilots as the performance would be counterintuitive. They'd know they have full thrust, they'd believe the flaps were in take-off configuration, they'd think the gear was coming up, they knew they initially had a positive rate of climb. And yet the aircraft doesn't respond as it should.

 

[TheCracker]

Does that mean he opened the emergency exit after knowing/seeing the plane faltering? That's both wickedly brave and outstandingly ballsy.

This just doesn't ring true. I can't see that he would have had time from figuring that there was an issue with the plane, to unbuckle and get out of his seat whilst the plane is still nose up and open the door which as anyone who has flown in a commercial jet knows isn't a quick or simple procedure - even for people who do it daily as part of their jobs. If he did manage to do all that you are falling straight onto the leading edge of the wing from the exit he exited from. Plus surely commercial jets have auto locks on the doors whilst in motion?

 

[Fezzik]

Does that mean he opened the emergency exit after knowing/seeing the plane faltering? That's both wickedly brave and outstandingly ballsy.

This just doesn't ring true. I can't see that he would have had time from figuring that there was an issue with the plane, to unbuckle and get out of his seat whilst the plane is still nose up and open the door which as anyone who has flown in a commercial jet knows isn't a quick or simple procedure - even for people who do it daily as part of their jobs. If he did manage to do all that you are falling straight onto the leading edge of the wing from the exit he exited from. Plus surely commercial jets have auto locks on the doors whilst in motion?

It seems he got out on the ground after the plane crashed and the door/fuselage broke

The Briton who walked out alive from the crash says he "still can't believe" he survived.

Vishwash Kumar Ramesh was the only survivor of the 242 people on board.

Ramesh said he thought he "was also going to die" but he then "opened [his] eyes".

"I pulled out the belt from under my seat and tried to escape. Then I managed to do it," he has told NDTV.

'It felt like the plane was stuck'

Ramesh said "within the first mile" after take-off "it felt like the plane was stuck" and "like something had happened".

"Then a light came on, like a green and white light came on inside the plane," he added.

"The pilot was trying to give it a bit of a push to push it forward, but it was struggling. But it went straight into a building."

He said the side where he landed was not the hostel side and as soon as the plane door broke, he "could see that it was open for [him] and [he had] a chance to get out".

 

[Keef]

Fighting with whatever issue was happening and just didn't get to it? Which would lead me to believe the plane had issues before they pulled back on the stick, or moments after the plane went nose up.

Seems like if they input variables incorrectly in the computer, they'd still be able to throttle up. The 787 has some serious thrust behind it so I would imagine they could pull out of it, unless the computer won't let them override. I don't know nearly enough about airplanes to say if this is even a reasonable take though.

For example, the jet I fly usually sets takeoff thrust at around 87%-90% depending on the weather and I believe the CRJ was about the same. There is a performance reserve and in some situations its automatic, but we can also engage that manually by simply firewalling the thrust levers rather than leaving them in the proper detent for takeoff or climb. The 787 has autothrottles which automate this process even further but I don't know example how it does that. Regardless, a manual override exists and it should have a pretty healthy reserve of power...again, this thing like all transport jets is capable of taking off on one engine.

One theory I've seen mentioned is that the copilot might have accidentally retracted the flaps instead of the gear. Not only would that have caused the loss of lift, but it would also have confused the pilots as the performance would be counterintuitive. They'd know they have full thrust, they'd believe the flaps were in take-off configuration, they'd think the gear was coming up, they knew they initially had a positive rate of climb. And yet the aircraft doesn't respond as it should.

Totally plausible mistake but I don't believe that would degrade performance below a single-engine level. When jets like these lose 50% of their thrust they actually lose 75% of their performance and yet are capable of continuing the climb in that situation. This plane didn't merely level off into a shallow climb as you'd expect in a single-engine procedure, it lost virtually all of its performance as if its engines just died or went to idle. That would be startling as a pilot but the fix is to just firewall it, come back and land, and call it a day lmao. Flaps alone couldn't cause this severe lack of performance during initial climb - I mentioned earlier that there have been cases where no-flap takeoffs have been attempted on the ground resulting in extremely long takeoff rolls, early rotations, and subsequent accidents, but this thing was already in the air climbing normally.

Edit: Reddit investigations are unfolding haha. A photo of the wing from the site suggest the plane's fowler flaps are extended out of zero.

 

[Eunos_Cosmo]

Correct me if I'm wrong, but if it was a lift issue wouldn't the plane have just hammered down the runway without taking off, crashing into whatever was at the end?

Seems like the most likely scenario was a fuel contamination issue. If the engines lost power, wouldn't it stand to reason that the plane lost hydraulics too? That may explain the landing gear not coming up, as well as the fact that the pilots were obviously very distracted...

 

[Exorcet]

Correct me if I'm wrong, but if it was a lift issue wouldn't the plane have just hammered down the runway without taking off, crashing into whatever was at the end?

The wings are always going to make lift provided that they can reach an angle of attack. If the plane is in the wrong configuration, lift off might require a greater angle or greater speed but there isn't really a way to "turn off" the lift.

Seems like the most likely scenario was a fuel contamination issue. If the engines lost power, wouldn't it stand to reason that the plane lost hydraulics too? That may explain the landing gear not coming up, as well as the fact that the pilots were obviously very distracted...

There is a backup ram air turbine on the 787, though it's typically pretty visible and I didn't see it deploy in the video. There may also be an electrical backup though I'm not super familiar with the 787 systems.

 

[Keef]

There is a backup ram air turbine on the 787

There is, and it's definitely visible in one of the videos. These videos have been hard to source but VAS has several. At 10:53 you can see a little blur on the lower right side of the plane, just inside the mains, and that is absolutely the RAT. That implies total electrical failure. I don't know if the 787's RAT is both an electrical generator and a hydraulic pump or if it's one or the other - my current Challenger 350 doesn't have one, and the CRJ series has a ADG or Air Driven Generator that creates only electrical power which thus powers auxiliary hydraulics. Some RATs are hydraulic only, allowing flight control actuation while the batteries handle the electrical load, but I assume the 787 is all fly by wire with hydraulic actuators.

Anyway, yeah that implies total electrical power failure and that implies total engine driven generator failure and the lack of thrust sorta explains why the generators failed.

Edit: forgot to post the VAS video.

 

[ScottPuss20]

I don't see a double engine flame out on take off so it obviously wasn't a complete engine failure. Which begs the question: why did the engines go dead? Was it due to an electrical failure? Initial evidence suggests that it could be the case with the very obvious sound of a RAT from the amateur video taken near the airport. But why? Why did the electronics fail? I think the issue lies on Air India's end, because some of their 787s went through a "refurbishment" which I believe was part of a large maintenance operation for some of the older aircraft. I read a comment on YouTube which said that the onboard entertainment system on one of the planes wasn't functioning properly. Something COULD have gone wrong after Air India tampered with the systems, so I think Boeing may be in the clear here. But who knows.