Hacker News Reader

> The report is actually a little cagey about whether the locks were properly installed on these switches. Said locks are supposedly optional.

The locks/gates on the switches are definitely NOT optional. There was an SAIB about some switches that may have been installed improperly. It didn't result in an AD, which likely means the extent was limited or potentially even nil.

The switches were moved to cutoff with a one second delay between the first and second switch. That's pretty suggestive of deliberate movement. I've flown a Max9 simulator, which has the same switches. Moving one of them by accident would be impossible, let alone two of them.

I agree with not jumping to conclusions about the pilots and possible motives or circumstances, but I will bet a lot of money that the switches were just fine.

The CVR will likely have audio of the switch movement to confirm as well.

> one second delay

did the report say a one second delay or that the two switches were turned off at consecutive seconds? The latter is what I remembered, but I'll check again.

The report can be found here: https://aaib.gov.in/What's%20New%20Assets/Preliminary%20Repo...

> did the report say a one second delay or that the two switches were turned off at consecutive seconds?

The report states, on page 14: The aircraft achieved the maximum recorded airspeed of 180 Knots IAS at about 08:08:42 UTC and immediately thereafter, the Engine 1 and Engine 2 fuel cutoff switches transitioned from RUN to CUTOFF position one after another with a time gap of 01 sec.

"with a time gap of 01 sec" seems fairly clear. The final report will have more granularity, but I don't think that's very ambiguous.

You are right, I was remembering the description of turning the switches back on.

> seems fairly clear

Not at all. First of all writing it as "01" begs the question if there is a typo (is that really supposed to mean 1 or 0.1?) but even without that only specifying the interval to a second precision makes is much more likely that the difference is between samples of the switch state that are recorded at that rate.

The switch must be lifted and turned. The optional posts block you from inadvertently knocking the switch (that you must pull up and turn).

I second that it’s not an accidental motion, you must actively manipulate the switch. But just like your turn signal in your car, it is muscle memory when you use it. I just wonder what action the pilot mistook the fuel cutoff for. Looking around the cockpit shows just how unique those switches are and not something you mistake with another common activity.

Pretty sad day if this was an intentional action

> Moving one of them by accident would be impossible

Improbable, not impossible.

If you were correct, the only situation it would happen in is when the pilot flying asks for X to happen, and the pilot monitoring instead does Y. Pilots don’t just randomly reach over and screw with the controls. Everything is called out, and as far as I know there were no callouts here (e.g. “gear up”).

This is a bit like someone parking their car, pulling the handbrake, turning off the car and putting their keys in their pocket, then arguing that it’s a practiced maneuver because it happens at the end of every car ride.

Look at a 787 cockpit. There are no other switches like that anywhere in the vicinity of those cutoff switches.

This is like pulling the handbrake on the highway when you’re trying to use your high beams

> Pilots don’t just randomly reach over and screw with the controls.

Plenty of crashes have resulted from Pilots not acting 100% by the book.

> and as far as I know there were no callouts here (e.g. “gear up”)

Unfortunately we don't have a full CVR transcript.

Turning the fuel off seems roughly equivalent to turning the ignition off when you've parked your car. It's really something rather unlikely to do as a brain fart during takeoff.

Most commercial aircraft have quite a few more buttons, knobs, levers, dials, etc than a car.

You have to lift and turn these knobs. Hard pressed to think of other knobs like this one. Especially that there are none like it anywhere near those cutoff switches. It seems intentionally designed this way

Commercial aircraft pilots are also much better trained than car drivers.

They do make an effort to make it hard to do safety critical stuff by accident though in cars, small planes and jets. Like in a car it's easy to mix indicators and windscreen wipers which are on various stalks but turning off the engine and locking the steering is very different action, on traditional cars turning the key and taking it out. Similarly here the fuel cut offs are obvious levers that have to be pulled out before they can be moved.

Putting your phone in the fridge seems like a pretty crazy thing to do, but we do it sometimes. Plane crashes are very rare.

I don't fly a lot of planes, but I do know what muscle memory is like, and sometimes it misfires. Misfires are rare, but then, so are plane crashes.

The emergency fuel cutoff should not be a practiced maneuver.

But we can already conclude that something unlikely did in fact happen, otherwise there were more fatal 787 crashes. I don't think "it's unlikely" is a good argument for dismissing potential causes here.

These switches are locked into place with a spring loaded detent switch. It's impossible to accidentally switch these off.

You say muscle memory. No muscle memory is involved in fuel cut off for both engines seconds after take off. There is no procedure remotely similar.

> These switches are locked into place with a spring loaded detent switch. It's impossible to accidentally switch these off.

The report specifically mentions SAIB No. NM-18-33 which would be a very odd thing to do if the investigators didn't at least consider this to be a possibility.

> You say muscle memory. No muscle memory is involved in fuel cut off for both engines seconds after take off.

It doesn't matter when the action is normally done - muscle memory means that the complex action of flipping a guarded switch doesn't really need a bigger brain fart than any other input because once the action is initiated you're acting on autopilot.

you make some good points but i thought your example was funny. No, i have never left my phone in the fridge. And i’m in my 30s, so not that young.

I HAVE sometimes “lost” my phone in my own backpack though (lol).

Here’s my take as a non-pilot. Takeoff is a sequence of checklists and procedures that are repeated. Often. If it’s muscle memory, a brain fart seems very unlikely… that’s why i’m skeptical of your theory.

A better analogy would be “have you ever forgotten to put on shoes before leaving your house?” And no, i have not, even during emergencies or when i’m very tired (like when i had to rush my wife to the ER). Why? because that’s something i do daily, and is part of my checklist when departing my house.

Or “have you ever forgotten how to get to work?” that also is extremely unlikely, because one typically follows the same exact steps and route daily. And i don’t randomly turn off my engine while driving, even if i’m multitasking or im forced to take a slightly different route due to a detour.

To be clear, i’m not saying it’s physically impossible. I just find it inconceivable.

> Brain farts are a real phenomenon, and an accidental fuel cutoff most closely resembles the transcript from within the cockpit.

What transcript? We only got a few potentially re-phrased bits of what was said. That the preliminary report is very careful about not implicating either pilot is not evidence of anything except that the investigative team is not yet 100% sure on the matter.

The evidence points to inadvertent fuel cutoff switch movement rather than pilot error, or intentional pilot action. Most likely cause being locking mechanism failure combined with takeoff acceleration forces.

Investigation and preliminary report is not showing critical evidence that would help clarify. Like the full transcript of pilot conversations that is clearly already available, and if these switches had any maintenance in the last year.

The fact that both pilots denied moving the switches, combined with the extremely short timeline, makes mechanical/electronic failure the most probable cause.

Note the Critical Sequence:

08:08:39 UTC: Aircraft lifts off (air/ground sensors transition)

08:08:42 UTC: Maximum airspeed of 180 knots reached

08:08:42 UTC: IMMEDIATELY after max speed, both fuel cutoff switches transition from RUN to CUTOFF (1 second apart)

08:08:47 UTC: Both engines below minimum idle, RAT deploys

08:08:52 UTC: Engine 1 switch returns to RUN

08:08:56 UTC: Engine 2 switch returns to RUN

So only 3 seconds between liftoff and fuel cutoff. Extremely short window for deliberate pilot action...

That doesn't make sense. If a pilot wanted to deliberately cut off the engine there's no reason they couldn't do at that time. The time difference between the switches being one second shows it was more likely to be deliberate, not less. An accidental hit would result in both happening at the same time, not with a time delay.

The two switches changed in sequence, but within the same second and at max speed. Dont think a human can have such a timing. If you look at the timing it looks more and more like electronical or software triggers.

Another possibility is a foreign object like a personal item sliding back during acceleration.

I looked at both pilots background, and unless a story of medical depression on the part of the captain emerges, I dont see the pilot suicide as plausible.

FAA just announced they examined the switched and found no mechanical issues. The switches are not software controlled.

Seems like the evidence is starting to become fairly one sided…

I agree the evidence points to the Captain, but the report did not provide the full text of the pilots interaction. The timestamps shows they were cut, one after the other, but within the same second as max vspeed. Doubt humans can have that precision.

>> The switches are not software controlled.

That is what unethical people, like Captain Steeeve - Who changed his "accident clear cause" story four times since the accident - keep saying.

But on the 787, fuel switches are part of the electronic cockpit interface and toggling one sends a command via AIMS to the Engine Interface Unit, then over AFDX to the FADEC. The FADEC actuates the spar and engine fuel shutoff valves who are both electrically controlled with no mechanical linkage.

The switch itself doesn’t cut fuel, it initiates a command interpreted and executed by software. Additionally, software logic can override or block the command based on conditions like fire, overspeed, or sensor faults so its software-controlled end-to-end.

A faulty sensor could trigger even if so far evidence points to the Captain.

"Understanding the Boeing 787 Fuel Control Switch: Function, Operation, and Safety Mechanisms" - https://aeropeep.com/boeing-787-fuel-control-switch-explaine...

"How do the fuel cutoff switches on the Boeing 787 work?" - https://aviation.stackexchange.com/questions/110938/how-do-t...

"Full Authority Digital Engine Control (FADEC)" - https://skybrary.aero/articles/full-authority-digital-engine...

"Full authority digital engine controls have no form of manual override available, placing full authority over the operating parameters of the engine in the hands of the computer.

- If a total FADEC failure occurs, the engine fails.

- Upon total FADEC failure, pilots have no manual controls for engine restart, throttle, or other functions."

If it were possible for a trained pilot to "brain fart" cut fuel to a 787 as it is taking off that would be a huge design flaw.

Did I hear "Design Flaw"? On a Dreamliner??

Well I'd Never sputter Good Ma'am or Sir, ARE you implying something like a "MAIN BATTERY EXHAUST"¹ retrofitted to a flagship product because management was impatient and wanted the lighter, better, faster, stronger lithium-ion-battery for their rushed 787-sized baby?

¹) https://media.cnn.com/api/v1/images/stellar/prod/14011413401...

Brain farts are literally not a real phenomenon.