Re: OT: air india crash

It sounds as if the engines weren't developing full power from the start if, as reported, it took the full length of the runway to get airborne. It had obviously just been re-fuelled because it was carrying a full load of fuel. Could the new fuel have been contaminated - maybe with water?

This is my understanding, too,

Pilot comments, which I haven’t seen contradicted, about the fuel load state that a full load is circa 220,000lbs, but that for the flight to Gatwick, including plenty for diversions, would have been around 120,000lbs or nearly 50 tons short of full tanks.

If that’s true, then the long take-off run, with an aircraft that wasn’t near its maximum weight, suggests the engines were not producing the required thrust, possibly gradually failing in some way even during that take-off run and conking out early in the climb-out leading to the ram-air turbine being deployed.

The air temperature at the airport was 37C.

The pilot did the right thing, by realizing he had a full fuel load, a full passenger complement and a derating to deal with at 37C. That's why he jogged the plane down to load at the very beginning of the runway.

One available video, shot from a smartphone, shows "exhaust colour" on the left engine, less or no exhaust color on the right engine, and a yaw as if the left engine was pushing the left wing forward.

It could be that one engine was out, and the remaining engine was constrained by the situation. While the airplane is supposed to be able to lift out on one engine, there are an awful lot of things weighing against it in this case.

Boeing has some recommendations for operation, that are counter-intuitive to a pilot taking control and rescuing an aircraft. There are also some maths to do, when determining what throttle setting to use. Which presumably is a reason to not be screwing with the throttles while in the early part of flight.

The pilot having to perform mental gymnastics, while the ground is coming up towards them, that's not a good design particularly. That's why you want to practice that particular situation, over and over in the behavioral simulator, until it is second nature.

Only time will tell, which control surfaces were operational in the last seconds of flight.

The exhaust colour was not the emission of an explosion. It was the spreading pattern of an engine at a high setting. The right engine, I couldn't see the same exhaust colour as the left engine. The angle of the video might have affected what we can see.

Given that the pilot purposely used the entire length of the runway, logically the engines are going to be set to as high a setting as is allowed by the derating. The throttle setting is not arbitrary, like pushing both knobs against the stops at the top. It's a machine with a performance curve and for some reason, the pilot is exposed to the math. So rather than being a virtual throttle ("slam to firewall"), it's a physical throttle ("adjust for known engineering limits").

Paul

Apparently it was very hot so would need a longer runway to get airborne.

The pilots had masses of experience. So, I think we can safely ignore the possibility that the engines weren't working properly at the start of take-off. Likewise, an overloaded plane, they'd have noticed and aborted take-off.

Is 37C temperature way out of the ordinary for that area? The forecast suggests not.

Thank goodness the AAIB has access to invaluable resources like Usenet to fall back on!

Some experienced pilots do a check on the time taken from opening the throttles to e.g. 100kt, just as a confirmation that the aircraft is configured and operating as expected. Any significant extension of that time will result in either an aborted take off or the throttles being jammed to the firewall.

And why do you speculate that the aircraft was overloaded? With a fuel load sufficient to get to Gatwick and a reserve to get to a raft of alternates, it was approximately 50 tons below max. Airlines hate hauling dead weight, it costs for no benefit.

I now can’t recall the finer details, but once upon a time there was an aircraft crash in France. The black box was recovered and showed everything was normal. Later, someone looked at the earlier data and found it referred to flights in North Africa, where the crashed aircraft had never been.

Think about that.

Airliners do not use maximum power on takeoff if they don't need it.

I thought that was airport CCTV? anyway yes. I saw that. Starboard engine failed pretty much as it rotated Would still have taken off OK

Lol!

-- There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.

Mark Twain

Perhaps..

All airports ought to have a high definition recording of every take off as some extra evidence of what went wrong?.

If it went wrong of course.

But considering the number of successful take off's and landings every day?.

Its waay down in the noise..

The same thought had occurred to me: high-def videos of all take-offs and landings, for the infinitesimal chance of it all going horribly wrong.

I remember an after-dinner talk in about 1980 "What goes up might come down" by a Birmingham air-traffic controller called Dave Gunson in which he said (I may be paraphrasing slightly)

"Take-offs are easy - it's brute force over ignorance - but landings

And "at rotate speed it *will* go up... and the editor's decision is final on that", which the Air India crash sadly proved not to be true...

At rotate it did go up. It just didn't keep going up.

Andy

Preliminary report says both engines fuel switches turned to cut-off.

The switches cannot be accidentally knocked to that position.

An observation from a pilot is that working engines take maybe 30+ seconds to spin down and the RAM wouldn't be deployed until they had spun down. If the aircraft is not on the ground and both fuel switches are in the cutoff position the RAM deploys immediately. The RAM was deployed a lot sooner than 30 seconds of flight.

Flame out of an engine when the aircraft is at height, based on the information given by a pilot commenting on the report. If there is a flame out the switches will be set to cut-off and then immediately turned back on but this would not be the first action on the check sheet.

Its speculated that they wouldn't have had enough time to consult a check sheet.

The (toggle type) switches have to pulled up using a thumb and finger against spring loading to be moved over a mechanical detent. They have to be lifted to put them back. They cannot accidentally be knocked out of position. It also takes two deliberate actions to put both switches in the cut-off position.

engine fire/fan explosion/whatever

Apparently both switches were turned back to run, one engine had started to spool back up, the other didn't have time.

Apparently you don't just "knock" the switch from on to off, you have to deliberately lift it past an over-centre point.

The report

It indicates that the fuel switches were subsequently changed from cut-off to run and one engine began to spin up again before the crash.