Failed Avionics a Possible Cause of BA038 Crash

Muhammar writes "As you may have heard by now, both engines of the Boeing 777 aircraft flight BA038 suddenly cut off without warning at very low altitude and low speed during autopilot-assisted landing at Heathrow. A prompt reaction of the pilots prevented the stall and saved all lives aboard. The crash landing short of the runway tore off the landing gear on impact, and the fuselage plowed a long, deep gouge in the grass. With the investigation ongoing, the available information points to an electronic control problem as the most likely cause of the sudden engine power loss."

Comments

Software?

[Marcion]

If it is a software problem, then expect more public scrutiny of software based machinery. Especially after the US Senate vs UK debacle over the source code for the new joint-combat fighter.

Re:Software?

[AlecC]

Actually, they have given up creating multiple implementations of the code. There were only ever two implementations, scattered across several computers. However, when developing the systems for this very aircraft type, Boeing decided that they now have tools which can verify precisely that the software matches the specification, and where they actually need to put the effort in is in checking that the specification makes sense. Rather than wasting effort in having two teams implement implement the specification, and verify that using automated tools, you use the extra effort to look closely at the specification.

Re:Software?

[ObsessiveMathsFreak]

That is not likely.

Yes it is likely. We are expected to believe that a single consumer grade device caused the simultaneous failure of both engines? Or from high powered sources which the planes must be built and certified to withstand. Give me a break. A Computer/Hardware glitch is a far more plausible cause.

That said, my paranoia meter says this could have been caused by some nut near the airfield with a HERF Gun.

Re:Software?

[Troed]

or a very near lower powered source such as a cell phone inside the farady cage

While already moderated funny, I'll just clarify that this is a myth. A more likely explanation for the cellphone ban on planes is due to the networks not being able to handle several hundred clients moving at 800km/h in view of tenths maybe hundreds of base stations.

Re:Software?

[TClevenger]

A software glitch of this type (if that's what it was) has never happened in aviation history. Certainly not in the 10 year history of the 777, with more than 500 of them flying around the world, but not to any other type either.

It's certainly not without precedent. No case of air/fuel mixture explosion was found in 747's until TWA 800 in 1996, and 1,396 of those were built since the 747 started flying commercially in 1970.

Possible autothrottle problem

[bananaendian]

With the investigation ongoing, the available information points to an electronic control problem as the most likely cause of the sudden engine power loss."

What I've read is that the pilots observed a relatively gradual loss of power symmetrically on both engines. This tells me that I can rule out engine problems with FADEC and fuel. It all points to the auto-throttle. Autopilot tells where it wants the plane to go and autothrottle calculates how much throttle is needed. It then commands both engines FADECs via the bus system which is doubly redundant. What I'm thinking is that auto-throttle is supposed to be backed up, bypassed by a manual direct control to the engine FADECs from the cockpit throttle control?

Any B777 avionics mechanics around - I only know military jets...

Are the pilots heros?

[XMLsucks]

I've read several summaries, such as this one, which state that the pilots did something to save the lives of the passengers. But I've never read a news article that provides the information that supports this claim. I'd like to read about what the pilots did to save the situation. Can anyone point out a news article that is actually coherent, and tells more than how many 777s are in service around the world?

Re:Errrrr..

[Thilo2]

You can be sure that the autopilot did not need to warn the pilot. Even a relatively unexperienced pilot will notice when
a) The airplane drops below the glide path
b) The airplane flies at too high of an angle of attack
during landing.

Re:terrists?

[Anonymous Coward]

Its surprising boeing have not been at this already. After the Long Island crash they were quite keen on the terrorist missile theory because it helped "explain" the fact that their fuel tanks could explode when a light was switched on.

In this case, however, in the absence of any explosions it's surely far simpler for them to fall back on their old favorite "pilot error" explanation. Like, maybe the pilot carelessly forgot that he hadn't landed yet and switched the engines off too soon? Sounds far more plausible than the possibility that a Boeing might have been faulty.

Re:terrists?

[R2.0]

A little bit of perspective here.

First, there were MANY credible witnesses that swore they saw a missile shoot into the sky before the explosion. Of course, it turned out to be the different trajectories of the airplane pieces, but that was only figured out after a detailed analysis of radar records.

Second, prior to Flight 800 the terrorist explanation WAS more likely - I don't think a modern airliner had EVER exploded by itself before that, but there had been a few that did it with outside help.

Finally, the intelligence and police agencies were careful NOT to peg it on terrorists as the only theory. It was the news media that ran with the "Arabs and Stingers and Bombs Oh My" stories incessantly. Yeah, the government floated the idea - because it was a definite possibility. What are they going to say? "We have some eyewitness acounts of what looks like a missile launch, but we have definitely ruled out terrorist involvement."

As an aside, where are the Flight 800 "Truthers"? Why isn't anyone blathering about the conspiracy to hide the REAL reason Flight 800 blew up?

That Is Brilliant!

[hax4bux]

Please post this at every /. article on aviation.

People, if you are so interested in aviation then get off the couch, take lessons and get some first hand experience. I know little airplanes are not completely the same as big airplanes, but you will be closer to some factual opinions.

Interesting Australian Passenger Recollection

[Hynee]

Here's an interesting account from an Australian passenger, he says:

... just as we touched down, a piece of debris punctured the wall and slammed into my leg. ... My disbelief at the sound of rushing air through the hole was soon overtaken by a sickening crunch as the plane hit the ground hard and all too quickly we had stopped.

Yep, the plane was actually punctured and he was hit, you can see the hole on the RHS of the aircraft behind the wing, just under windows.

Anyway, his recollection indicates that the plane was punctured before it touched the ground. If that were the case, his "hole" would probably be the point of failure.

I think it is more likely that the puncture happened after the plane hit the ground, caused by debris from the right landing gear ripping away. It would be like this--plane touches down on grass (he thinks they're still smooth in the air); wheels dig in rip off, and punctures hull in quick succession (he has been hit); the plane starts scraping along the hull and engines (he feels the plane "hit the ground hard").

So it's probably just a slightly misleading passenger recollection, but something to think about while we're guessing about the control systems.

From various articles on the incident

[einhverfr]

The right and left engines are controlled by different computers. The only single points of control are the pilot and a central engine control system. Thus in the absence of pilot error, the only single point of failure is that specific avionics system.

Now the root fault may be due to some sensor or processing system failing and causing a cascade failure to other portions of the system. This sort of thing *has* happened in other 777's (an accelerometer failing in a way as to cause a cascade error into flight control software). In the end the most careful proof of software accurate operation must make certain assumptions about unerlying hardare states. Once hardware starts to go bad, all bets are off (for example, sensors could fail in such a way as to provide apparently valid but wildly inaccurate data to the software which would then return incorrect results (and hence take wrong actions).