Dreamliner Crash in Ahmedabad: Air India flight 171

13 Jun 25 21 Comments

Yesterday, the 12th of June 2025, Air India flight 171 crashed after departure from Ahmedabad Airport in India.

Air India flight 171 was a scheduled passenger flight from Ahmedabad to London Gatwick, UK which crashed just 45 seconds after departure, impacting a medical school dining hall in the Meghani area of the city.

Ahmedabad Airport, officially the Sardar Vallabhbhai Patel International Airport (SVPIA), located in Gujarat, is the seventh busiest airport in India, handling 13 million passengers last year. The single runway, 05/23, is 3,505 metres long (11,500 feet). The airport elevation is 189 feet above mean sea level.

Air India was re-acquired by the Tata Group (the original founders) three years ago, India’s largest multinational business which owns many global brands including the aerospace manufacturer Tata Advanced Systems Limited, Taj Hotels, Jaguar Land Rover and Tata Starbucks.

Air India Boeing 787-8 VT-ANB taken June 2024 by By FNMG (CC BY-SA 4.0)

The aircraft was a Boeing 787-8 Dreamliner registered in India as VT-ANB, serial number 36279.  This is the first fatal crash of a Boeing 787. Air India has a fleet of 27 Boeing 787-8 Dreamliners; VT-ANB was their second Dreamliner, manufactured in 2013 and delivered to Air India in January 2014.

That morning, the aircraft flew from Delhi to Ahmedabad as Air India flight 423. At Ahmedabad, 220 passengers boarded for the nine-hour flight to London manned by two flight crew and ten cabin crew. The majority were Indian nationals with 53 British, one Canadian and seven Portuguese listed on the manifest.

The captain of the flight had 8,200 hours flight experience and according to The Indian Express was just a few months from retirement. The first officer had 1,100 hours flight time.

The departure was scheduled for 13:10 local time but the aircraft actually departed at 13:39.

Initial reports made much of the fact that on FlightRadar24, the aircraft appeared to enter the runway at an intersection without backtracking, meaning that they did not make full use of the runway., This is untrue: ADS-B data, which is what FlightRadar24 use, is notoriously unreliable on the ground and a quick check of previous departures of flight 171 all show the same data artefact. FlightRadar24 have since confirmed this:

Additional processing confirms #AI171 departed using the full length of Runway 23 at Ahmedabad. RWY 23 is 11,499 feet long. The aircraft backtracked to the end of the runway before beginning its take-off roll.

The weather was hot: the METAR for the crash time (VAAH 120700Z 25006KT 6000 NSC 36/19 Q1001 NOSIG) shows a slight crosswind for runway 23, with good visibility and temperature of 37°C, standard conditions for June in Ahmedabad.

This CCTV video shows the aircraft rolling and then departing from the runway. It certainly lifts off slowly; however, it was hot and the Boeing would have been heavy with a full load of passengers packed for travel to the UK and fuel for the nine-hour flight.

It is not possible to see from the video whether the aircraft was correctly configured for take-off. The aircraft climbed away and according to the ASD-B data, reached a height of 625 feet before falling out of the sky.

Business Today quotes a statement from the Directorate General of Civil Aviation (DGCA) that one of the flight crew made a distress call to declare an emergency.

As per ATC, the aircraft departed from Ahmedabad at 1339 IST (0809 UTC) from Runway 23. It gave a MAYDAY call to ATC, but thereafter no response was given by the aircraft to the calls made by ATC. The aircraft, immediately after departure from Runway 23, fell on the ground outside the airport perimeter. Heavy black smoke was seen coming from the accident site.


This video, uploaded without attribution to Reddit shortly after the accident, shows the aircraft descending in a nose-high attitude. The landing gear, which usually would have been retracted once they reached a positive climb rate, is clearly visible.

A bystander interviewed for BBC News said that they heard a loud noise before the crash. “It sounded like a bomb blast.”

Another bystander told the Times of India that he’d heard something.

I have my office 200 metres from here. As soon as I stepped out of the office, I heard a very loud noise, and smoke filled the area suddenly… When I reached the spot, I saw debris scattered here, there was a fire, and smoke was billowing here. Nothing was visible. We then came to know that the wings of the aircraft fell here and that an aircraft crashed.”

All we know for sure is that the Boeing Dreamliner came down just 1.5 km (less than a mile), crashing into the grounds of the Civil Hospital Ahmedabad, a local medical college. The explosion and fire on impact were particularly intense as the aircraft was fully fuelled, leading to many being trapped within the building.

The Times of India reported that three teams consisting of 90 personnel were immediately dispatched from Gandhinagar with another three teams dispatched from Vadodara. Their immediate focus was to set up a “green corridor” to transport the injured to hospitals.

However, a source reported that the majority of those transported to the hospital were already dead.

There was one survivor, a British national sitting in seat 11A who was miraculously able to climb out of the emergency exit with relatively minor injuries.

The Indian Express created this image from Boeing and Reuters information, showing the identifiable parts of the aircraft:

Wreckage of Air India 171

At least 50 people on the ground were taken to the hospital to be treated for their injuries. As of this afternoon,  the BBC has updated that there were eight confirmed fatalities in addition to the 241 passengers who lost their lives (four medical students and four relatives of other students).

All flights were suspended from the airport and Indian Railway put on additional trains to the airport to help transport stranded passengers to other major cities.

ANI News reported that the survivor from seat 11A also heard a bang. “Thirty seconds after take-off, there was a loud noise and then the plane crashed. It all happened so quickly.” However, a later interview for DD News, cited in the Guardian, makes no mention of hearing a sound before the crash.

When the flight took off, within five to 10 seconds it felt like it was stuck in the air. Suddenly, the lights started flickering – green and white.

The aircraft wasn’t gaining altitude and was just gliding before it suddenly slammed into a building and exploded. At first, I thought I was dead. Later, I realised I was still alive and saw an opening in the fuselage.

There are many analyses of the videos of the flight, however, the quality of the videos is of dubious quality. Although the landing gear is clearly not raised, it is in my opinion not possible to definitively tell if the flaps and slats, used to gain lift on take-off, are correctly configured. Similarly, one video has what seems to be a clear sound  of the Ram Air Turbine (RAT) being deployed, which would indicate a loss of power. However, the sound does not appear on another video taken at a similar distance. With this video quality, I don’t feel confident that it had definitely deployed, although the flickering lights seen by the surviving passenger is a further sign of a possible power fluctuation issue.

The very high nose-up attitude in the videos has been cited as the crew pulling the aircraft into a stall, but as the captain had three decades of experience, I find this hard to swallow. I think that it is more likely that the flight crew were not in a position to do anything to stop the Boeing from crashing and were simply trying to land as slowly and as flatly as possible.

Still from the airport’s CCTV footage

There are a few credible theories floating around as to the cause the crash.

Misconfigured

One theory is that the flaps and slats were not extended before take-off, which caused the low and slow departure. However, the Boeing lifts off and seems to reach a positive rate of climb, despite being heavy on a hot day. It’s possible but the quality of the videos that have been passed around so far are not good enough for me to feel confident. This also does not explain why the RAT deployed (if it did).

Another theory is that the aircraft was correctly configured for take-off but, as they climbed over 500 feet, a flight crew member retracted the flaps instead of the landing gear, causing a sudden loss of lift. This has happened before, and we just recently looked at an action slip by the pilot flying, who accidentally reduced thrust on take-off. However, this theory is more dubious. It also seems to be based on the claim that the aircraft has no flaps set in the final moments of the video, but presuming the flaps were correctly set to fully extended, reaching for the lever and suddenly retracting them would allow only one stage of flaps to be changed. It is not possible to retract full flaps in a single movement, believing that you are raising the landing gear.

Engine Failure

The loss of power could certainly be explained by an engine failure. However, the current evidence points to a double engine failure, as the Boeing Dreamliner should have been able to maintain altitude with a single engine (if configured correctly).  With the landing gear down and the flaps potentially misconfigured, the heavy aircraft may have struggled with only one engine, but in the videos, there’s also no visible sign of asymmetric power, which should be easier to spot than the position of the flaps.

There are only a few reasons why both engines would fail at the same time. There are a few instances of one engine failing and the crew mistakenly shutting down the working engine, but there was hardly enough time for such a mistake in this instance. Contaminated fuel is another potential cause, but no other aircraft suffered issues and the airport re-opened just a few hours later, which means that they did not believe there was any issue with their jet fuel. More likely is a bird strike, as the airport is in the Sabarmati river basin, where bird activity peaks during the monsoon season (June to September). But this is not conclusive: there are no signs of smoke or fuel leakage from the engines and I have seen no reports of bird remains at or near the crash site.

Something Else

But what?

The clearest evidence at this point is that the landing gear was still extended, when it should have been retracted as soon as they had a positive rate of climb. That could be a mistake. I think it is more likely that the crew were distracted by some other failure. Or it’s possible that they did not have the hydraulic pressure to raise them.

Eyewitnesses are notoriously unreliable so a key question is whether the reports of a loud “bang” are correct and related to the flight. Definitive proof that the RAT had deployed would imply a systems failure, rather than a configuration error.

We probably won’t have to wait for long. It should be possible to quickly determine whether the engines had power, despite the damage. Today, the Flight Data Recorder was recovered, which, assuming it is readable, will give us the key metrics of the take off and the descent. If the Cockpit Voice Recorder is recovered (and I see no reason why it wouldn’t be, the photographs of the tail show that it did not suffer in the fire) then that will tell us what the flight crew were doing.

For now, though,  there’s no definitive explanation and we shall have to await further updates.

Of course, that doesn’t stop the media from headlining their own theories. The New York Times immediately cited “concerns about the safety of the Boeing 787” and that the crash was “renewing scrutiny of [Boeing’s] safety record after a years-long quality crisis”, despite the fact that the Boeing in question had flown 11 years with no issues. I am not absolving Boeing of their issues but as of right now, there is no evidence that the Dreamliner is unsafe. India’s DGCA has understandably ordered enhanced inspections on all Air India Boeing 787‑8/9s with General Electric GEnx engines. At this stage, there’s no indication that the aircraft will be grounded.

My favourite headline, though, was posted as a screenshot to PPRuNe.

Unusual for ‘state of the art’ plane to crash at take-off, expert says

Seems like a safe statement to make. That particular expert went on to say that the investigation should focus on security at the airport because “basically, the aircraft these days, they sort of fly themselves”.

Prime Minister Narendra Damodardas Modi visits a crash site of the AI-171 flight.

India’s Directorate General of Civil Aviation (DGCA) has confirmed that the Director General of the AAIB is in Ahmedabad with an investigation team. The American NTSB  will lead US team in their role as the state of manufacture. General Electric, the engine manufacturer, will also be a part of that team. The British AAIB has deployed a team to assist the investigation, representing the 53 British passengers on board (including the only survivor). A preliminary report will be expected within 30 days of the crash.

Receive Fear of Landing in your inbox every Friday

    I won't send you spam. Unsubscribe at any time.

    21 Comments

    • Thanks for putting all the actual facts in one place where I can send people a link. There’s a lot of guesses, guesses stated as facts, and just pure BS out there, and I’m sure it took a lot of time and research.

      Anyway, I think this is the first aircraft crash where I haven’t even a clue what happened.

      One correction: in the caption to the last pic, the PM is actually Narendra Modi, (Damodardas is his middle name)

    • The video you posted actually has the “vroom” RAT sound—and very little engine noise.

      Any sort of misconfiguration would be apparent as the aircraft rotates. There’s video from the far end of the runway that shows what looks like an uneventful take-off run and rotation—if the aircraft was misconfigured, it would rotate but not lift off (this often leads to a tail-strike), or it would be slow to climb. But a comparison with older FR24 data shows the aircraft took off pretty much the same as the same flight did on the days before. And a photo of the wrecked wing shows the flaps deployed. I feel certain that this theory is a red herring.

      For now, the most likely theory is a simultaneous dual engine failure. The video I mentioned shows the aircraft from take-off to crash, and there’s no rudder deflection or yaw to be seen, as would be the case if the engines failed consecutively. So we can rule out bird strike, contaminated fuel, or the scenario where one engine fails and the crew inadvertantly shuts off the other engine—none of this can have happened, it’s impossible.

      As to what caused the engine failure, we don’t know. It may have occurred on rotation, as the aircraft slows down from then on. It may turn out to be a malfunction, or a maintenance issue, or a pilot spilling his coffee all over the switches in the cockpit. And we won’t likely know exactly what caused it until the final report comes out (unless we see some very specific airworthiness directives applied to 787s).

      But the chances are good that we’re going to find out eventually, because this time, the CVR and FDR (actually they’re a combined unit) had battery backup.

      • But the chances are good that we’re going to find out eventually, because this time, the CVR and FDR (actually they’re a combined unit) had battery backup.

        I wondered about this but the DGCA specifically said the DFDR and I couldn’t find any reference to the CVR so I wasn’t sure what the deal was.

    • I am certain that all who watched videos of this crash felt sick to the stomach. And, of course, our thoughts are with the family, friends, relatives and associates of the victims.
      No doubt we are all in full agreement here.

      Until the CVR and FDR are fully examined, there will be a flurry of.guesses Some from experts, but also some by less informed journalists. Since the press, more interested in a sensational story to promote the sale of their newspapers insisted calling bright dayglo-orange objects “black” I have a dose of mistrust. Sorry. I have strong feelings about it. Modern jet aircraft have a large number of “black boxes”. Really black. They are usually situated in racks below the cockpit floor and control the radio navigation and communication equipment, as well as the increasing and bewildering number of computers and other electronic gizmos.

      The bright orange CVR and FDR are located in other areas where they will be more easily accessible if the worst happens.
      In this case it did, the worst did happen. Spectacularly so, and very very sad. Shocking to watch.

      One passenger miraculously survived. But the total apparently runs close to 300 including people on the ground.

      There are a number of theories. And let us be careful: until the real cause is determined, they are pure guesses. No more than that.

      The aircraft took off from an intersection and did not have sufficient runway for a safe take-off. Rubbish.
      The aircraft was not properly configured for take-off. Were the flaps accidentally not extended? Very difficult to see from the videos.. I find that hard to believe. I retired from flying well before the current highly automated aircraft became the staple of the airlines. The A320 was very new. But even some of the much older generation aircraft that I flew had a take-off configuration warning.
      How sophisticated is the B787? Would a configuration warning have sounded that could have caught this oversight have been installed?
      An engine failure. The aircraft is seen taking off and initially climbing steeply, then very clearly stalls but continues in a straight line. No yaw into a “dead engine”. So if it was an engine failure, both would have failed simultaneously. Bird strike? Not likely, but cannot be ruled out entirely, And why was the gear still down?
      Poor maintenance, leading to a catastrophic failure? The survivor has been quoted as being critical of many things wrong with the cabin appointments and a malfunctioning air conditioning. Not comfortable, but in flight the pressurisation and cabin temperature are controlled by the packs that are fed by the engine bleed air. If the APU was unable to control the temperature in the cabin, it would have been very warm and the packs would have been working overtime to bring the temperature down once airborne. If the aircraft was unsafe to fly, would the crew have accepted it?
      Still “poor maintenance”: The aircraft was heavy and the outside air temperature high. No “flexi power” would have been used, my guess is that the required setting for take-off would have been full TOGA. Was there a malfunction of the autothrottle system that reduced the power against the will of the pilots and that they were unable to counteract?.

      I am totally out of my “comfort zone” discussing the technical aspects of these highly sophisticated aircraft. Maybe someone can educate me?
      6.Pilot error. The most chilling of them all: Did the monitoring pilot retract the flaps instead of the gear? Hard to believe, but the video clearly show a healthy initial climb, followed by a very obvious stall and a visible effort of the flying pilot to keep the aircraft in the air. Unsuccessfully as is clear to see. And equally clear are the wheels, still extended. Hard to believe: the handles are not in the same position. In most of the types that I have flown, the handle for the landing gear is somewhere in the middle of the lower part of the instrument panel. The flap switch is at the right of the pedestal. Moreover, they have different shapes: the ear lever is in the shape of a wheel, the flap lever represents an airfoil shape. Easy to identify, even without looking at them. But the videos certainly suggest the possibility of this scenario. According to an expert, the wreckage shows the flaps in the extended position. Did the crew lower them again to correct the situation? By that time the aircraft was stalling already. High pitch attitude, lots of drag and a disturbed airflow. All too late. Not a chance.
      The experience level of the crew has so far not been mentioned, but it this is found to have been the actual cause, then Air India’s training standards would leave a lot to be desired.
      I cannot believe that an airline such as Air India does not operate to high international standards,

      Crew incapacitation. There is a precedent: the 1972 BEA crash of a Trident due to premature retraction of the slats and flaps after take-off from Heathrow.
      Here the captain had been incapacitated by a heart attack after a vicious row over crew duty times.
      Could crew incapacitation have caused this crash?

      Of course, all that I wrote is conjecture, and I don’t want to sully the memory of the poor crew.
      The real experts will eventually, possibly soon, find the actual cause of this tragedy.

    • On Jun 13th 2025 the DGCA reported, that initial preliminary findings rule out a bird strike as no bird carcasses have been found. Loading of the aircraft was routine, hence a load problem is unlikely. The likelihood of a pilot error is considered minimal as is the simultaneous failure of both engines. An inspection of all Dreamliners (B788, B789) has been ordered.

      The Directorate General of Civil Aviation (DGCA) is India’s equivalent of the FAA.

      The pprune thread is nearing 1000 messages, and as always it’s difficult to distinguish who knows what, who is guessing, what observations can be trusted, and so forth. With that in mind, here’s what I found interesting:

      The distance from Ahmedabad to London is about 6800 km. The 787-8 spec has its range with max payload as 13600 km, which is twice that. This means the aircraft was not particularly heavily loaded for this flight. The take-off roll was normal.

      Would a configuration warning have sounded? Yes, the crew would not have been able to close the digital pre-takeoff checklist if the flaps had not been set correctly.

      Why was the gear still down? The gear hydraulics are operated via the central hydraulic system, which is driven by electric pumps. The available videos seem to show that the main gear has tilted forward from its take-off position to the stowed position. The hypothesis is that gear retraction had commenced when the aircraft lost electric power. (The RAT only powers essential systems; not this.) But maybe the bogeys just naturally fell down when power was lost?

      Was there a malfunction of the autothrottle system? It is said that the autothrottle is inhibited below 400 ft AGL, an altitude that this aircraft never attained. And in any case, pushing the throttle forward, or pushing the TOGA button, should easily override the autothrottle.

      Did the monitoring pilot retract the flaps instead of the gear? There is no evidence to suggest that this happened, and some evidence to suggest that it did not. In this context, I’d like to stress that “AI upscaled” means “AI guessed at details that can’t be seen in the original video”. If you’re looking at an “upscaled” video, you’re looking at what the AI imagined, not at what actually happened.

      Could crew incapacitation have caused this crash? The crew is said to have radio’d a distress call along the lines of ‘Mayday! No thrust. Losing power. Cannot lift.’ That suggests the aircraft was incapacitated, but the crew was not.

      In the past 6 months, we have witnessed 4 fatal airliner crashes in Aktau, Muan, Washington D.C., and now Ahmedabad. They all presented open questions that were not easily answered, and some remain unanswered to this day. We owe it to the passengers and crew to answer these questions thoroughly and truthfully; this is why we have agencies dedicated to investigate aircraft accidents, and because they don’t take this responsibility lightly, their investigations take many months.

      It is my conviction that in all 4 of these crashes, the airline crews acted professionally (perhaps even heroically) after they encountered improbable conditions that were not survivable. We owe it to the bereaved to make it absolutely clear that their beloved did not lose their lives through wanton disregard of their safety, or incompetence among the ranks of those tasked to safeguard it, no matter what initial uninformed rumors suggest. “News” that is shocking travels faster than news that is true.

      People having opinions about videos of the crash does not constitute evidence. Experience shows that these early opinions are often proven wrong, even if they seem reasonable. We do well to keep that in mind.

      • I’m curious about “The distance from Ahmedabad to London is about 6800 km. The 787-8 spec has its range with max payload as 13600 km, which is twice that. This means the aircraft was not particularly heavily loaded for this flight.” Is it customary for ultra-long-range aircraft on short routes (relative to their range) to take off with partially-empty tanks? Or is there some other assumption in this statement that I’m not seeing?

    • More comments on this horrible crash emerge. Some are too sensationalist, even if the author claims to be an active pilot.
      The theory of inadvertent, premature retraction of the lift enhancing devices, flaps and slats, is convincingly promoted by a man calling himself “Captain Steeeve”. B777 captain in full uniform.
      He argues that, as I also suggested, that take-off in a wrong configuration triggers a lot of warnings that are impossible to miss.
      The aircraft is seen to climb away normally before it enters a stall from which it could not recover.
      This would seem to support “Capt. Steeve’s” theory of premature flap retraction, especially as the gear is still clearly still in the “down” position.
      But another commentator, claiming to be a B787 captain,, anonymously writes that the B787 has highly sophisticated safety systems that make it impossible to retract the flaps before the gear.
      Which contradicts the theory of erroneous flap retraction.
      Another entrant, claiming to be a former fighter pilot and current airline captain, suggests that there are sounds audible on the video that would indicate deployment of the “RAT”. This is the Ram Air Turbine that will extend automatically in the event of a loss of (virtually) all electrical power. It was the RAT that enabled Capt. “Sully” Sullenberger to retain control of his A320 when it ditched into the Hudson river. A failure of this magnitude, immediately after take-off is virtually unthinkable but could indeed point to a massive bird strike.
      Even though there is no presence of what would have been a very large flock visible in the video.
      There is no doubt that the FDR and CVR will be recovered and analysed. These recordings will determine the cause of this horrible, horrible accident.
      Until then everything is speculation, even some is written by experts.
      In the meantime, I wish the survivor a good recovery and hope that he will be able to overcome the traumatic experience and loss of his brother. We all with ties to the aviation industry mourn the loss of so many people, including those who were killed on the ground.

      • From its highest point (possibly ~200 ft. AGL), the accident aircraft travelled a further 4200 ft (1250m) of distance to the crash site, which fits published “gear down” glide ratios. I don’t believe this could be accomplished while stalled. There’s a photo from the crash site that shows a wing with flaps and slats extended. I hope the data from the enhanced airborne flight recorder (EAFR) is going to convince even the last doubter that the flaps were not retracted, and that the aircraft did not stall aerodynamically until the final “flare”.

        The second video that Sylvia included (the vertical one) has the RAT very distinctly audible at the beginning, while we do not hear a pair of jet engines at take-off power. The 787 almost sounds like a propeller aircraft. This establishes that the engines failed, that the power generation failed, and that the RAT deployed; and this take is also supported by other evidence. I would be very surprised if the recorded flight data turned out to contradict that version of events.

        I agree that the simultaneous dual engine failure was not caused by a bird strike; in fact, the DGCA has ruled this out as well. Speculations center around a cause aboard the aircraft. As it looks now, that cause is likely to turn out to have been unpredictable/unpredicted. The flight recorder data is going to be instrumental in understanding how this happened.

        I’m thankful that the crew managed to set the aircraft down in a way that was mechanically survivable, even if most people aboard lost their lives in the explosion and post-crash fire; and that the aircraft did not hit the hospital a short distance away. It’s a ray of light in this tragedy.

    • FR24 has published a blog post presenting the ADS-B data received by their station at https://www.flightradar24.com/blog/flight-tracking-news/major-incident/air-india-171-crashes-shortly-after-take-off-from-ahmedabad/ (and expanded it at least once).

      They take the received ADS-B barometric MSL altitudes and convert them according to airport elevation and (then) current air pressure. For example, 575 ft. STD becomes 21 ft. AGL, which seems plausible as it’s shortly after rotation.
      The ADS-B transmitter reports the altitude in multiples of 25 ft., so it wouldn’t be 21 ft. exactly. If you graph the vertical profile of the flight, it’s not going to look as smooth as it actually was because the actual altitude gets rounded to the nearest 25 feet.

      We can also see that the ADS-B track cuts off prematurely, either because the aircraft is too low, so the FR24 station can no longer see it, or because the electric power failed.

      Looking at the CCTV video that Sylvia posted, and knowing that the wingspan of the Boeing 787-8 is 200 ft., we can estimate that the aircraft climbed to slightly above 200 ft. before sinking back down.

      The FR24 data shows that the performance was problematic right from the start: while the aircraft rotated in a similar place as previous flights, and with sufficient speed to establish positive climb, both the climb rate and speed began decaying almost immediately.

      Any theory of the accident sequence needs to take these facts into account.

      Bonus evidence: https://imgur.com/a/YE2q1e3
      This is a part of a frame of the witness video (the vertical video Sylvia posted) from about 8 seconds into the video, 2x enlarged. We can clearly see the flaps deployed, and there’s a faint trace of something sticking out underneath the fuselage.

    • This very tragic event took place soon, immediately, after take-=off and has been recorded by different viewers. As a result it has drawn an enormous attention.
      For more than one reason: it was a large aircraft with more than 200 people on board, coming down in a built-up area. The operator is a major airline and the manufacturer, Boeing, has already been in the spotlight for all the wrong reasons.
      So many people, including myself, have attempted to make sense of the information that became available.
      Two of the more authoritative experts came with a plausible scenario. Some of these were immediately contradicted by B787 captains who enlightened us about the highly sophisticated protection floor built into the 787, preventing the aircraft to take off with the wrong configuration, such as flaps not extended, and a protection to prevent the premature retraction of the flaps and slats.
      One captain, publishing a podcast calling himself “Max Afterburner”, came up with the possibility of a major, catastrophic failure.
      Another captain, calling himself “Captain Steeeve”, eventually confirmed this hypothesis. He got hold of enhanced videos that contain a whirring sound. The sound has been identified as the “RAT”, the Ram Air Turbine that extends automatically in the event of a total loss of electrical and hydraulic power. In one of the enhanced shots the RAT is indeed visible, extended under the wing. “Captain Steeeve” reports that a search for debris under the flightpath of the doomed 787 did not reveal dead birds. No bird the size of a goose is visible in the videos. Only a very large flock of small birds could have caused a double flame-out, but nothing has been found.
      The video shows clearly a good climb rate, followed by a stall. The aircraft did not suffer a single engine failure, no yaw is visible. Just a sickening heart-wrenching decreasing airspeed and a low speed stall. From that moment the crew had no control any more. The RAT would have provided some movement of the control surfaces, but the accident was unavoidable and in fact theoretically non-survivable.
      That one person survived, even relatively unharmed, is a total miracle.
      The most likely cause now emerges as a major, catastrophic failure of critical components. Did this lead to the simultaneous shutdown of both engines, or did the engines just flame out at a catastrophic moment?
      What was the cause?
      Was it a design fault?
      Was it caused by defective components?
      If so, was there a major quality problem going back to the manufacturer, or was there an underlying problem with the airline’s maintenance department?
      Could the cause have been fuel contamination?
      If so, was contaminated fuel uplifted, or had the airline neglected to prevent fuel contamination, e.g. scheduled cleaning and inspection of the aircraft tanks and fuel system?
      Pilot error can be ruled out.
      Even though the most likely cause now is a total and very serious failure, probably a flame-out of both engines simultaneously, even if so, there are still very many questions and answers will be required instantly.
      Because, depending on the outcome, there are in my opinion three suspects:
      1. The airline, its maintenance department in particular.
      2. The manufacturers: Boeing. If the engines are the cause, it is unlikely in the extreme that they will fail at the exact same moment.
      3. The suppliers of the fuel.

      Mendel, sorry but the crew “did not manage to set the aircraft down”,.and certainly not in a manner that was survivable.
      The lucky survival of the passenger in seat 11 was in defiance of all reasonable outcome.
      The crew had no more control, zero, zich. The aircraft came down in an unrecoverable stall.
      Dallman’s comment that the 787 does not use bleed air, presumably for pressurization and air conditioning is news to me.
      If this were accomplished by electrical power, the electrical system would certainly need to be very powerful, and probably heavy.
      None of the comments by b787 pilots mention this.
      The report about flickering lights, I believe, comes from the sole survivor. The most likely cause would he the switching by the system to the RAT.

      The FDR and CVR will give the answers, Hopefully sooner than later.
      This is a very serious matter, with potentially devastating consequences for whichever party (or parties) are found to have been negligent in their duty to the travelling public.

      Again, I am aware that I am speculating. My background as an airline pilot is with aircraft and systems that are obsolete or becoming outdated.
      I confess to my limited knowledge.

      But in many ways, even though the technology has evolved, the underlying principles are still valid.

    • Comments by Capt Steeve and another airline pilot, GregC, clearly point to a catastrophic mechanical failure of such magnitude that the accident was unavoidable. Enhanced videos clearly show the RAT employed, including the whirring of the propeller.
      The accident became totally unavoidable as the aircraft entered a full stall.

      I mentioned a few possible culprits, but of course the uplift of contaminated fuel is a virtual impossibility. Otherwise more aircraft would have suffered engine failure and a simultaneous total failure of both engines due to fuel contamination is highly unlikely. anyway.

      But another very disturbing possibility crossed my mind: was there any factor that was lurking in the background, but kept a secret and withheld by the FAA? And was the Directorate General of Civil Aviation of India kept in the dark for commercial reasons?
      This would have been unthinkable only a few years ago, but with the secrecy that led to withholding information for commercial reasons, leading to a number of crashes, like those of the B737 Max, leads to some suspicion.

      The old adagio: “If it ain’t a Boeing I ain’t be going” now has been replaced with “If it’s a Boeing I ain’t be going”.

      Another case of withholding crucial potential safety issues can more than likely be discounted, but even thinking of the possibility is something that would have been totally impossible only a few short years ago.

    • Comments refer to the 787 stalling. I would expect a much more marked nose down pitch if a genuine stall occurred. To me the flight path looks as though the crew were keeping the aircraft just above stalling speed after significant loss of power. It seems to have been under control in this state until impact.
      However, those commenting are more familiar with the characteristics of modern, large aircraft than I am. I’d welcome correction.

    • If this was a major technical fault, I’d expect the 787 fleet to be grounded by now, honestly.

      OTOH, if it is the RAT deployed, as it very much looks and sounds like on at least one video (and the engines weirdly quiet for that phase of flight)… that’d most likely be a technical fault. Pretty sure I could see flaps deployed both in flight and in wreckage photos too (that very slender, beautiful 787 wing is not easy to see flaps on!).

      So, I don’t know. Eagerly waiting to hear more, heartsick at the loss of these people and the aircraft they travelled on.

    • Disclaimer: I’m not a pilot, rather a software/hardware engineer with a specialism in safety critical systems and the cause of accidents. As such, I recognise my thoughts are speculative, and they are not in any way intended to detract from the horror of those involved directly or indirectly with this tragedy.

      However, it appears current evidence suggests:

      There was clearly a lot of fuel on board,
      there is no yaw associated with staggered engine failure,
      there is no obvious sign of birds, or engine damage/fire,
      flaps do appear to have been deployed,
      there was a total loss of power.

      I am therefore wondering if this accident is the result of a software/hardware fault simply causing the EEC to shutdown the two engines….

      I know there are a whole load of interlocks which are supposed to prevent this, but as with the MCAS crashes, effectiveness of interlocks is entirely tied in with:

      how they are implemented,
      whether the conditions checked are representative of the systems full operational envelope,
      whether the sensors providing the systems status are providing valid/accurate data.

      Can anyone who has flown a B787 comment on what interlocks would prevent a pilot deliberately activating the engine cutoffs (not that I’m suggesting for a moment that is the case here), and given this, what sensor errors might have allowed/caused the flight software to trigger a dual cutoff by mistake?

    • This very tragic event took place soon, immediately, after take-=off and has been recorded by different viewers. As a result it has drawn an enormous attention.
      For more than one reason: it was a large aircraft with more than 200 people on board, coming down in a built-up area. The operator is a major airline and the manufacturer, Boeing, has already been in the spotlight for all the wrong reasons.
      So many people, including myself, have attempted to make sense of the information that became available.
      Two of the more authoritative experts came with a plausible scenario. Some of these were immediately contradicted by B787 captains who enlightened us about the highly sophisticated protection floor built into the 787, preventing the aircraft to take off with the wrong configuration, such as flaps not extended, and a protection to prevent the premature retraction of the flaps and slats.
      One captain, publishing a podcast calling himself “Max Afterburner”, came up with the possibility of a major, catastrophic failure.
      Another captain, calling himself “Captain Steeeve”, eventually confirmed this hypothesis. He got hold of enhanced videos that contain a whirring sound. The sound has been identified as the “RAT”, the Ram Air Turbine that extends automatically in the event of a total loss of electrical and hydraulic power. In one of the enhanced shots the RAT is indeed visible, extended under the wing. “Captain Steeeve” reports that a search for debris under the flightpath of the doomed 787 did not reveal dead birds. No bird the size of a goose is visible in the videos. Only a very large flock of small birds could have caused a double flame-out, but nothing has been found.
      The video shows clearly a good climb rate, followed by a stall. The aircraft did not suffer a single engine failure, no yaw is visible. Just a sickening heart-wrenching decreasing airspeed and a low speed stall. From that moment the crew had no control any more. The RAT would have provided some movement of the control surfaces, but the accident was unavoidable and in fact theoretically non-survivable.
      That one person survived, even relatively unharmed, is a total miracle.
      The most likely cause now emerges as a major, catastrophic failure of critical components. Did this lead to the simultaneous shutdown of both engines, or did the engines just flame out at a catastrophic moment?
      What was the cause?
      Was it a design fault?
      Was it caused by defective components?
      If so, was there a major quality problem going back to the manufacturer, or was there an underlying problem with the airline’s maintenance department?
      Could the cause have been fuel contamination?
      If so, was contaminated fuel uplifted, or had the airline neglected to prevent fuel contamination, e.g. scheduled cleaning and inspection of the aircraft tanks and fuel system?
      Pilot error can be ruled out.
      Even though the most likely cause now is a total and very serious failure, probably a flame-out of both engines simultaneously, even if so, there are still very many questions and answers will be required instantly.
      Because, depending on the outcome, there are in my opinion three suspects:
      1. The airline, its maintenance department in particular.
      2. The manufacturers: Boeing. If the engines are the cause, it is unlikely in the extreme that they will fail at the exact same moment.
      3. The suppliers of the fuel.

      Mendel, sorry but the crew “did not manage to set the aircraft down”,.and certainly not in a manner that was survivable.
      The lucky survival of the passenger in seat 11 was in defiance of all reasonable outcome.
      The crew had no more control, zero, zich. The aircraft came down in an unrecoverable stall.
      Dallman’s comment that the 787 does not use bleed air, presumably for pressurization and air conditioning is news to me.
      If this were accomplished by electrical power, the electrical system would certainly need to be very powerful, and probably heavy.
      None of the comments by b787 pilots mention this.
      The report about flickering lights, I believe, comes from the sole survivor. The most likely cause would he the switching by the system to the RAT.

      The FDR and CVR will give the answers, Hopefully sooner than later.
      This is a very serious matter, with potentially devastating consequences for whichever party (or parties) are found to have been negligent in their duty to the travelling public.

      Again, I am aware that I am speculating. My background as an airline pilot is with aircraft and systems that are obsolete or becoming outdated.
      I confess to my limited knowledge.

      But in many ways, even though the technology has evolved, the underlying principles are still valid.

    Leave a Reply to Mendel

    Your email address will not be published. Required fields are marked *

    *
    *
    *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.