Posts about: "Fuel Cutoff Switches" [Posts: 827 Pages: 42]

No experience of the 787 but lots on the 75/76.

The possibility of a perceived stab trim problem allied to an associated eicas message could have induced the PNF to go straight to the stab trim switches mindful of the problems associated with the 737 Max albeit a different type and a system not fitted in the 787 but an issue of which the Training Captain would have been well aware especially if he had been briefed on the stab messages from the previous sector. If, repeat if, he simply made the wrong selection and hit the fuel cut off switches then the holes have suddenly lined up. Unlikely, unfathomable perhaps but by no means impossible. Other than that....deliberate act ?

I've studiously avoided commenting on this accident, for good reasons. And looking at the history, many others would have benefited from the same restraint.

And I won't comment now. But this article, about another classic "no one would do that" scenario is pertinent and worth reading in its own right:

https://www.washingtonpost.com/lifes...a52_story.html

It's precisely the ones who loudly proclaim "I would NEVER do that", "that could NEVER happen to me" who are most at risk. And that is relevant to aviation too. IMHO.

R1

no experience of Boeing at all here but would a stab trim issue have been highlighted by EICAS that early? I assume Boeings have take off inhibit logic to hide certain less than emergency conditions until above ~1500AGL? Would a Stab issue have been serious enough to immediately pop up?

Placing the fuel control switches to cutoff is not \x93hard\x94 but you certainly need some force to do it.
They certainly can not jump across the safeguard by themselves. (in turbulence for example)

this whole scenario is simply far beyond my comprehension.
what happened to crew coordination when selecting fuel switches? Even in a perceived emergency situation, the switcher needs to confirm the perception.

This puzzled me as well and got me thinking about the quick diagnosis of the problem.

If one of the two pilots had preplanned this action, and didn\x92t want to be blamed for it, he will have known the conversation would be recorded, so he moves the switches to cut off, then immediately challenges the other pilot about the action, which of course he denies. One of them resets the switches to run, but by then it is too late.

Language:

Having lived in the Middle East for over 20 years, I am somewhat used to the Indian use of the English language. Therefore, I wouldn't look too closely at the usage of "transitioned"; it simply means moving from one condition to another.

The conversation of the pilots, on the other hand, will be very interesting. It should be analysed in the language they were speaking, considering whether this was their native language to start with. I assume that the conversation was in Hindi, translated to English for the report. In that case alone, a lot of nuance might have been lost in translation. That's without even considering the tone, volume etc. of the conversation.


Procedure:

I am an Airbus driver, so I am not familiar with the B787 EICAS. On the Bus, the ECAM would generate an ENG FAIL, followed shortly thereafter by an ENG ALL ENGINE FAILURE. I assume the B787 would have a similar event on the EICAS. Now, considering the time frame of how quickly things happened, there is absolutely no reason for the pilots to assume something was wrong with both engine cutoffs, unless they were physically switched off by someone. We are not trained to consider a fuel cutoff switch as the main reason for an engine failure, especially on the takeoff roll.


Question?

When I was a young F/O, some of the captains I flew with had the (annoying) habit of resting their hands just behind the thrust levers on their PM (PNF) sectors to 'be ready' to reject the takeoff. This would put the hands in the vicinity of the Fuel Cutoffs, which would, in turn, increase the chances of an unintended action on these switches. Is this a possibility, or am I way off?

Agreed: you would not see all contacts failing the same way, at the same time, in a multi-pole switch like that. Nothing is impossible, of course but the odds of that happening in one switch are extremely low. Two switches, at the same time?.
Ditto for the idea of the physical detents failing, causing the switch lever to drop at the worst time. Not impossible, again, but both switches within a second of each other? Very low odds.

Am I correct in thinking that engine cutoff would in flight would trigger a warning on EICAS?

It's established fact both fuel control switches moved to CUTOFF 4 seconds after Vr, a deliberate guarded action, not easily done accidentally.
The CVR records one pilot asking “Why did you cut off?”, the other replies “I didn’t”. Then a bland Mayday attributed to the Captain “engine failure, returning” in the middle of the crisis.

The language, if reported correctly, feels strangely detached. No confusion, no urgency, no clear troubleshooting. Not drawing conclusions, but does anyone else see signs of performative behaviour, that is saying the right things outwardly, while being at odds with the underlying cause?

I appreciate that both crew members lost their lives, however if we avoid discussing uncomfortable patterns, we miss the point of investigation and learning.

20 minutes between engine start and take off clearance. That\x92s enough time for the cockpit to cool down even if the packs were not turned on previously.

Really? It suggests to me and I would imagine the vast majority of us who have flown modern Boeings that they were physically moved, by one of the crew, one at a time, the question is why.

We can be sure this switches and their locking mechanism were carefully inspected in the wreckage .

since there is no grounding of the 787 fleet i think nobody considers the switches were defective but it seems to be clear for the investigators they were pulled and switched to cutoff by pilot action .

The investigation probably now concentrates by whom of them both and to find out if it was a catastrophic operation error in a mental blackout or a concious , deliberate action .

The accident aircraft was written up for a status message of ”STAB POS XDCR” on the previous flight, which is a message relating to implausible data from the stab trim switches. It was released from maintenance (according to the preliminary report) at 06:40UTC ahead of an 07:40UTC departure (the crash flight) with ”no fault found”.

On the 787-8, as all modern planes, switches are not cabled as dry closing contacts all the way from the switch poles to the affected end devices (FADECs in the case of fuel cutoff switches), but rather connect locally to an analogue/digital converter to encode the switch position data onto the digital comms bus ARINC629 which allows all aircraft systems to talk to one another.

Are the fuel cutoff switches, which are positioned adjacent to the stab trim switches, connected to the same ADC module which produced the error message on the previous flight, which maintenance was unable to resolve before the accident flight took off? I do not know, but it must be worthy of being looked into.

Previous media reports (from supposedly well-placed sources) talked about ‘voltage spikes’, ‘rapid power disruptions’, ‘faulty power-panel components’ and ‘EEC glitches’ inferring an uncommanded electronic state change in the fuel switches. This report sadly seems to reverse the causality.

The Mods have stuck to this principle, which I shall try to adhere to.
This preliminary report is just that, but maybe consider the issues the Indian AAIB have had to address in publishing it. They will have a similar concern to the pP mods, maybe more so since any apparent accusations directed at the pilots may lead to physical retribution.
I therefore conclude great care has been taken to "sanitise" what the AAIB know, or at least strongly suspect, (from EAFR) into the report. They have conspicuously failed to identify which of the pilots was each half of the conversation they have not repeated the exact words, there's a lot missing (was positive rate ever called, was rotate ever called, any discussion about putting FC back to Run, who/how flying aircraft meanwhile). As a result we, the reader, should step back and not over-interpret this sanitised report.

Secondly, given the mod statement above, if a criminal act is suspected by the AAIB, this will likely trigger all sorts of 'primacy' issues in the investigation i.e. police? AAIB? or joint? and all the history that involves (SAS Linate?) - in Europe we have 996-2010 Article 12 para 2, but India?

Summary : For good reason I believe this report has been very carefully worded, sanitised with great care, and as such easy to inappropriately speculate what went on.

From my reading of the AAIB report, the delay in noticing the cut off switch position could not have been any later than 08:08:47 because the reference to the verbal exchange was written before the reference to the RAT deployment and I assume the written narrative follows the chronological sequence of events. The delay in resetting those fuel switches might therefore have been no more than 5 seconds. It is a pity that the pilot making the challenge didn't simply reset the switches to run instead of challenging the other pilot. I frequently tell people who blow their horns at pedestrians or other motorists that blowing the horn isn't going to help them avoid an accident and that what they really need to do in the limited time available to them is concentrate on applying the brakes and turning the steering wheel.

Spoken to a rated mate whom is current on 78s and he told me that inadvertent operation of a fuel shut off in flight has happened with a switch being knocked past its detent (lock) by the switch being caught by a crew bag.

Probably not during Takeoff I wouldn\x92t have thought.

I cannot imagine a situation where any pilot would be moving a crew bag within seconds of getting airborne.

From the report it is clear that the switches were selected to cut off, we will likely NEVER know why, maybe the CVR gives a few more clues but we will never know if it was an intentional or accidental act. What is clear is that during rotation you are never fiddling with bags etc.

Was the RAT deployed manually?
The report says, This was 5 seconds after the fuel was cut off.

It suggests to me that the RAT deployment was initiated while the engines were still above idle and generating electrical power. Obviously one of the pilots could have done it via depressing the switch, as it's a "dual engine failure/stall" memory item (see Air India Ahmedabad accident 12th June 2025 Part 2 ) that won't hurt anything.

Is there a way for the RAT to deploy while the engines are still above idle?