Air India Crash

esmozz said:
No, I understood what you where saying and it makes sense - it could be that.

To be honest it could be pilot suicide and if it is then the industry is going to have to face up to it. I don't like the idea either.

But one of the things that seems really quite odd to me the more I think about it, is just how remarkably vague the preliminary report is. For example - I can see no reason to leave open the question of who said what to who. I have no doubt a full transcript of the cockpit conversation covering the event exists. And things like thrust setting, aircraft configuration seem to be missing.
Click to expand...

Oh it’s massively redacted - part of me thinks it’s probably intentional as they definitely know more. I find it odd they chose to include the SAIB if it’s not relevant, and the words of the pilots without more detail. I get they have to be right in their eventual report and I don’t want it rushed as the consequences of being proven wrong at a later date don’t bare thinking about, but it feels like they should’ve given more or less information to prevent speculation, rather than this halfway house approach.
 
One of the pilots blatantly did it... If you watch any of the blow by blow brake down videos I posted posted previously, they are basically saying as such.

The chances of it being a mechanical issue are about as remote as losing a hand of poker with 4 aces.
Technically possible, but mathematically so improbable it's just not really a consideration.
 
Penfold101 said:
As much a I’d like to say ‘read the report’, there’s a high chance one of them did it. Intentionally or unintentionally is what the investigators haven’t released yet but I suspect they know.
Click to expand...
I wonder how they could ever tell for sure, obviously they are going to do extremely detailed checks into the personal lives and metal health of both pilots but there's essentially four different ways it could go:

1: Pilot A did it intentionally
2: Pilot A did it unintentionally
3: Pilot B did it intentionally
4: Pilot B did it unintentionally

Even if they manage to prove who most likely did it, they will still never be able to prove intent.
 
The question is: could a short circuit have caused this whilst the switches remained in position? The previous flight had significant electrical problems.
 
gary996 said:
Did the pilot turn off the fuel to the engines??
Click to expand...

It seems that the copilot took the plane off and the pilot switched the fuel off at a time when the copilot couldn't really do anything about it. It seems rather the perfect time to switch the fuel off, to crash the plane.
There was a one second gap between the first switch moving to off and then the second, which is consistent with a deliberate action rather than an accident. Also, it is unlikely the switches could be moved accidentally. That would be classed as faulty switches, which, again, seems unlikely that both were faulty at the same time!
The copilot asked "why did you cut off" and the pilot replied "I did not" (although who said what was not identified by the report), and and they were turned back on again. However, it was too late.
Draw your own conclusions but it looks very much like the pilot chose to turn them off at a very critical moment to deliberately kill himself and everyone onboard.
 
Last edited:
One thing I wondering regarding polling rates? Maybe someone will know more than me about this. If the polling rate is 1Hz and the switch is rapidly moved between on/off in much less than a second. Is it possible that the switch would not be recorded, or leading on from that, is it possible only half of the switching action is recorded i.e from on to off, but not the preceding off to on (or vice versa)?
 
As an avid aviation enthusiast, some things about this have raised my own questions, as we continue to receive an incredible amount of poor journalism, poor investigation insights (the prelim report raises more questions than answers), and high speculation. Whether there was human error, malicious or otherwise, it does raise questions of Boeing, particularly in the DM era, of their systems.

Things that I find unclear:
1. The switches are mechanical, but send electrical signals to FADEC. These switches are available for pilots to engage with for many reasons, safety and emergency. It's proposed that they were physically moved, but is it not possible that an short, or some other system failure led the FDR to believe they'd been moved from RUN to CUTOFF.

I've seen comments in here saying the 787 fuel switches are mechanical, which is only true in operation, like your light switch in your house. The rest is electrical signal sent to FADEC which is software. Another analogy to be considered is 5 people go into a data centre and one of the most important business critical servers shuts down. Who did it? Physical switch, software controlled.

Based on this, we can never know with 100pc certainty, which pilot pulled the switches, if at all. Just because FO was rotating, doesn't mean he couldn't have also changed the fuel switches. "Both hands full" remarks are speculation at best.

2. If the fuel switches were physically moved, why would the RAT deploy? There must have been another system responsible for realising sustained flight was not possible for the RAT to deploy. Simply moving the switches from RUN to CUTOFF would not deploy the RAT.

3. Focus is on the captain. However, if I was planning on bringing down a jet, knowing full well it was being monitored by FDR/CVR I would be clever enough to make an ambiguous comment such as "why did you turn off" knowing full well, it would throw someone else into the spotlight.

4. In relation to point 1, why where thrust levers set to idle despite FDR showing take off thrust? Seems like some possible system discrepancy cannot be ruled out.

5. Seems to me that FADEC should be gateposted in a more efficient way, ie, "hey, dummy, we're in the rotation and climb phase, are you really really really sure you want to starve both engines of fuel?" (For instance, bird strike, engine failure)

Boeing, under the DM era, have history of covering up findings, cost cutting and lack of compliance with regulation, I sincerely hope they aren't doing that here, because if another 787 comes down in a similar fashion, and they've been pushing the human error narrative, they're going to have a lot of explaining to do.
 
Last edited:
Django x2 said:
As an avid aviation enthusiast, some things about this have raised my own questions, as we continue to receive an incredible amount of poor journalism, poor investigation insights (the prelim report raises more questions than answers), and high speculation. Whether there was human error, malicious or otherwise, it does raise questions of Boeing, particularly in the DM era, of their systems.

Things that I find unclear:
1. The switches are mechanical, but send electrical signals to FADEC. These switches are available for pilots to engage with for many reasons, safety and emergency. It's proposed that they were physically moved, but is it not possible that an short, or some other system failure led the FDR to believe they'd been moved from RUN to CUTOFF.

I've seen comments in here saying the 787 fuel switches are mechanical, which is only true in operation, like your light switch in your house. The rest is electrical signal sent to FADEC which is software. Another analogy to be considered is 5 people go into a data centre and one of the most important business critical servers shuts down. Who did it? Physical switch, software controlled.

Based on this, we can never know with 100pc certainty, which pilot pulled the switches, if at all. Just because FO was rotating, doesn't mean he couldn't have also changed the fuel switches. "Both hands full" remarks are speculation at best.

2. If the fuel switches were physically moved, why would the RAT deploy? There must have been another system responsible for realising sustained flight was not possible for the RAT to deploy. Simply moving the switches from RUN to CUTOFF would not deploy the RAT.

3. Focus is on the captain. However, if I was planning on bringing down a jet, knowing full well it was being monitored by FDR/CVR I would be clever enough to make an ambiguous comment such as "why did you turn off" knowing full well, it would throw someone else into the spotlight.

4. In relation to point 1, why where thrust levers set to idle despite FDR showing take off thrust? Seems like some possible system discrepancy cannot be ruled out.

5. Seems to me that FADEC should be gateposted in a more efficient way, ie, "hey, dummy, we're in the rotation and climb phase, are you really really really sure you want to starve both engines of fuel?" (For instance, bird strike, engine failure)

Boeing, under the DM era, have history of covering up findings, cost cutting and lack of compliance with regulation, I sincerely hope they aren't doing that here, because if another 787 comes down in a similar fashion, and they've been pushing the human error narrative, they're going to have a lot of explaining to do.
Click to expand...

2. The RAT will deploy automatically under certain conditions, in this case probably due to loss of hydraulic pressure. In addition, the checklist for loss of thrust of both engines calls for the RAT to be deployed manually.

3. This seems plausible - it could also be the result of some kind of weird dissociative phycological state to avoid the implications of what he's just done. I'm absolutely no expert here though.

4. Most likely the thrust levers moved to idle during the impact.
 
Big friday morning post coming up!

Harlequin said:
The question is: could a short circuit have caused this whilst the switches remained in position? The previous flight had significant electrical problems.
Click to expand...

I don’t know the exact wiring of the switches, but I would expect them to be entirely separate, and not even exit via a common plug to the aircraft wiring to prevent this very thing. A short COULD explain it happening to one engine, but not both together. They were also able to switch them back again and the engines attempted to restart, which suggests they were working fine internally.

esmozz said:
One thing I wondering regarding polling rates? Maybe someone will know more than me about this. If the polling rate is 1Hz and the switch is rapidly moved between on/off in much less than a second. Is it possible that the switch would not be recorded, or leading on from that, is it possible only half of the switching action is recorded i.e from on to off, but not the preceding off to on (or vice versa)?
Click to expand...

This is another question I’ve seen raised elsewhere, and it’s generally accepted that a 1Hz polling rate for a switch like that would be likely, and thus puts the timing anywhere from 0.1 seconds to 1.9 seconds. I don’t knowing this has been confirmed though.

Django x2 said:
As an avid aviation enthusiast, some things about this have raised my own questions, as we continue to receive an incredible amount of poor journalism, poor investigation insights (the prelim report raises more questions than answers), and high speculation. Whether there was human error, malicious or otherwise, it does raise questions of Boeing, particularly in the DM era, of their systems.

Things that I find unclear:
1. The switches are mechanical, but send electrical signals to FADEC. These switches are available for pilots to engage with for many reasons, safety and emergency. It's proposed that they were physically moved, but is it not possible that an short, or some other system failure led the FDR to believe they'd been moved from RUN to CUTOFF.

I've seen comments in here saying the 787 fuel switches are mechanical, which is only true in operation, like your light switch in your house. The rest is electrical signal sent to FADEC which is software. Another analogy to be considered is 5 people go into a data centre and one of the most important business critical servers shuts down. Who did it? Physical switch, software controlled.

Based on this, we can never know with 100pc certainty, which pilot pulled the switches, if at all. Just because FO was rotating, doesn't mean he couldn't have also changed the fuel switches. "Both hands full" remarks are speculation at best.

2. If the fuel switches were physically moved, why would the RAT deploy? There must have been another system responsible for realising sustained flight was not possible for the RAT to deploy. Simply moving the switches from RUN to CUTOFF would not deploy the RAT.

3. Focus is on the captain. However, if I was planning on bringing down a jet, knowing full well it was being monitored by FDR/CVR I would be clever enough to make an ambiguous comment such as "why did you turn off" knowing full well, it would throw someone else into the spotlight.

4. In relation to point 1, why where thrust levers set to idle despite FDR showing take off thrust? Seems like some possible system discrepancy cannot be ruled out.

5. Seems to me that FADEC should be gateposted in a more efficient way, ie, "hey, dummy, we're in the rotation and climb phase, are you really really really sure you want to starve both engines of fuel?" (For instance, bird strike, engine failure)

Boeing, under the DM era, have history of covering up findings, cost cutting and lack of compliance with regulation, I sincerely hope they aren't doing that here, because if another 787 comes down in a similar fashion, and they've been pushing the human error narrative, they're going to have a lot of explaining to do.
Click to expand...

1. The switches do a lot more than just the FADEC - there are 3 or 4 poles inside sending signals to different components, including the FDR for switch position. That the endings shut down and the signal was seen on the recorder would suggest they did physically move.

2. Automatic RAT deployment has a variety of conditions - in this case more than likely the generators were all disconnected at the moment of fuel cutoff and that triggered it: You’re in the air, all your generators have quit and you’re down to battery power.

3. Indeed - if you’re talking life insurance, you’d have to be convincing, including pretending to try and restart the engines.

4. The FDR recorded them at the fully forward position the entire flight, so more than likely they were impacted in the crash. The reverser levers were also bent backwards.

5. Definitely a conversation that will be had - Embraer don’t allow shutoff unless the throttles are at idle, for example. That would be a huge industry wide undertaking and not a quick job I suspect.
 
Penfold101 said:
Big friday morning post coming up!



I don’t know the exact wiring of the switches, but I would expect them to be entirely separate, and not even exit via a common plug to the aircraft wiring to prevent this very thing. A short COULD explain it happening to one engine, but not both together. They were also able to switch them back again and the engines attempted to restart, which suggests they were working fine internally.



This is another question I’ve seen raised elsewhere, and it’s generally accepted that a 1Hz polling rate for a switch like that would be likely, and thus puts the timing anywhere from 0.1 seconds to 1.9 seconds. I don’t knowing this has been confirmed though.



1. The switches do a lot more than just the FADEC - there are 3 or 4 poles inside sending signals to different components, including the FDR for switch position. That the endings shut down and the signal was seen on the recorder would suggest they did physically move.
Click to expand...

Ah that answers my question, thanks - so the actual position of the switches is recorded. And I guess polling is like a time 'bucket', where every event between 0.1s - 1.9s goes into the same bucket.

Well there goes my theory....

Edit: going back to your previous post about 3 seconds not being enough time to call 'positive rate' - 'gear up'. Upon reconsideration, it's strikes me as possible that 3 seconds might exactly coincide with that call.
 
Last edited:
esmozz said:
Ah that answers my question, thanks - so the actual position of the switches is recorded. And I guess polling is like a time 'bucket', where every event between 0.1s - 1.9s goes into the same bucket.

Well there goes my theory....
Click to expand...

I’d imagine they can also see the valve position in the engine which the faded is operating, and it all comes together at the same time so they can be certain the physical switching caused the shutdown.

People were concentrating on the ‘1 second’ aspect but that won’t be the exact timing difference, just that they moved at a different big enough to go into the next bucket, as you say. Realistically there wouldn’t be any reason to poll them any more often than that - you couldn’t dream up this sort of scenario.

I still have hope they can prove the switch locks were both defective and something fell down to hit them both on rotate, because I really dont want it to be pilot error or malicious intent. It’s a small amount of hope though.
 
Penfold101 said:
I’d imagine they can also see the valve position in the engine which the faded is operating, and it all comes together at the same time so they can be certain the physical switching caused the shutdown.

People were concentrating on the ‘1 second’ aspect but that won’t be the exact timing difference, just that they moved at a different big enough to go into the next bucket, as you say. Realistically there wouldn’t be any reason to poll them any more often than that - you couldn’t dream up this sort of scenario.

I still have hope they can prove the switch locks were both defective and something fell down to hit them both on rotate, because I really dont want it to be pilot error or malicious intent. It’s a small amount of hope though.
Click to expand...

Sorry - I edited my post whilst you where quoting mine.

I'll repost here so as not to be confusing.

'Edit: going back to your previous post about 3 seconds not being enough time to call 'positive rate' - 'gear up'. Upon reconsideration, it's strikes me as possible that 3 seconds might exactly coincide with that call.'
 
Penfold101 said:
I still have hope they can prove the switch locks were both defective and something fell down to hit them both on rotate, because I really dont want it to be pilot error or malicious intent. It’s a small amount of hope though.
Click to expand...
I think at this stage it's probably unlikely. But I feel the same way about it.
 
esmozz said:
Ah that answers my question, thanks - so the actual position of the switches is recorded. And I guess polling is like a time 'bucket', where every event between 0.1s - 1.9s goes into the same bucket.

Well there goes my theory....
Click to expand...

Unless you know the physical implementation, the circuit design, then we're still guessing.

Usually critical systems don't poll for this very reason. It makes the ensure system very unpredictable. The events are handled immediately. However an audit system, such as the flight recorder, may still be polled.
 
esmozz said:
Sorry - I edited my post whilst you where quoting mine.

I'll repost here so as not to be confusing.

'Edit: going back to your previous post about 3 seconds not being enough time to call 'positive rate' - 'gear up'. Upon reconsideration, it's strikes me as possible that 3 seconds might exactly coincide with that call.'
Click to expand...

You’re probably right - enough time for the rate check, call for the gear and the hand movement though?

I’m clutching at straws still if I’m honest.
 
You must log in or register to reply here.
Back
Top Bottom