http://www.bloomberg.com/news/articles/2...rasia-jet-probe

that may forever remain a mystery

Sad story again.
Cockpit audio recordings could eventually shed a little light on this particular point, no?

Slug

I'm starting to not like all the computer control and interconnected systems.

Makes me think that the pilots probably got confused regarding how the plane was flying, didn't trust the computers, turned them off and tried to fly manually what they thought was the right way. Wouldn't be the first time people die because pilots trust their instinct at the wrong time over computers. E.g. like the Apache pilot in Desert Storm who didn't believe his GPS when it told him he was off course, and instead went on his own map reading skills in a barren desert at night without thinking of the wind that was behind the helo being somewhere else than he thought before putting a Hellfire in an M113.

Wonder if the system was turned off and a manual reset was attempted due to a lighting or static discharge occurring? They were in weather at the time and static discharges have been known to affect electrical control systems, taking them off-line.

The situation reminds me of the Air France flight 447 that was lost over the Atlantic a few years back. Bad weather, faulty speed data that automatically disabled the normal autopilot and passed control to the pilots.

Maybe the pilots tried to reset the computer to get the autopilot back online?! But if they intended a manual reset, they would have turned it on again...

The Apache did not have a GPS in Desert Storm. It had an Inertial Measurement System and doppler that was unable to detect that the helicopter was being moved ever so slightly by the wind. The computer gave the pilot incorrect position information in this case and the pilot determined that the location of the vehicles must make them enemy based upon the incorrect information provided. If you watch the video you should note that the pilot had a difficult time pulling the trigger because he realized something was not right.
The Apaches used in Operation Normandy had to follow USAF PAVE LOW aircraft because the navigation systems then in use on Army aircraft were designed to work in Central Europe where there are lots of terrain features on which to orient one's vehicle.
The first thing Army instructor pilots tell the new pilots in the Aircraft Qualification Course (AQC) is not to trust your life to a black box. They can and do fail. Computers are there to assist the pilots not make decisions. If the computers quit function properly then it is incumbent upon the pilots to be able to safely return the plane. That is why you are paying them.
It is more likely that the pilots in question did not really now how to fly the plane like the Asiana pilots in San Francisco who crashed on a perfect day because no one seemed able to fly a manual approach. These guys probably had no idea how to fly the plane past pushing buttons on the MPDs and setting the autopilot to execute.

oh, come on that is a news report from people with little to no understanding of aviation... speculating on that is rather pointless, from things flying around in the turbulences hitting a switch to distrust in faulty stall warning a million reasons could be invented but it all doesn´t change the fact that they flew right through an enormously wild thunderstorm and that is what likely caused the crash...

Seems like an odd scenario indeed. I'm thinking that due to some issues with the air data / airspeed sensors the aircraft thought it was in an overspeed condition. It may have pitched up and the pilots were unable to lower the nose with manual input in any of the automation modes. I think they were trying to regain control and lower the AOA manually but even that didn't work, perhaps due to a glitch in the fly by wire system of the elevator that wasn't allowing raw unaugmented inputs by the crew.

I have experience on the ERJ-145 and recently transitioned to the more automated E-175 - I've asked myself "what's it doing now?" quite a few times as I learn the nuances of the automation modes. It's very important to fully understand the automation logic in all the various modes, and to not become an automation dependent "child of the magenta". If it isn't doing what you want it to, turn it off!

Well my years on the A320 feel like a lifetime ago. But my recollection is that disabling the FACs (or SECs, or ELACs) is called for only in the most oddball of non-normals, if ever. If any of these computers fail, you just do a few ECAM actions and press on. It's not like resetting any of them will get the autopilot back. And anyway, who cares if you get your autopilot back! The flight is 2 hours; go without. BUT... if you WANT Direct Law, that is the way to go. Turn of both FACs and you've turned your A320 into a 737. The plane is still perfectly fine, but now there are no limits to what you can do. That's all I'll say because anything further is just speculation. But if true it does seem odd.

Yep, lots of speculation still, but from the limited and clueless-journalist-filtered stuff in that article, it could be something similar to what happened to a Lufthansa A320 on the way from Bilbao to Munich in November, when an angle of attack probe froze, and led the computers to believe that the plane was in an ever-increasing climb, and so the protections kicked in and started lowering the nose. The pilot flying held the stick fully back, but the plane continued to pitch down uncontrollably, until the quick-thinking pilot monitoring pulled a couple of circuit breakers (Smokin_Hole or Wireman can probably clarify which ones, as I don't know the Bus that well - I think it was the air data computer CBs) and regain control, after having lost 4000 ft of altitude.

http://avherald.com/h?article=47d74074&opt=0

I wonder about the sudden gain in altitude, and the
previous request to change course. Makes me think they're
flying straight into a monster, and want desperately to
avoid it, and have to wait for clearance. If they enter
the thunderhead, they could be hoisted a few thousand
feet without any pilot action. (Isn't there any regulations
about pilot discretion when faced with extreme weather,
when the traffic controllers are far far over the horizon?)

A pilot never flies into anything he/she doesn't wish to fly into. A clearance is irrelevant in this situation (talking lateral only). You turn first, regardless of clearance, and use PAN-PAN-PAN, off track procedures until you can get proper coordination from ATC. Climbing above a storm is often a mistake unless the tops are fairly low, the convective strength is also low and you begin the climb many miles before the line.

Yes, that is what my expectation would have been.
The news reports made it sound like they were not
prepared to divert without permission, but if that
only applies to an altitude change, that seems rather
less dire. Still, I suppose one could find oneself
boxed in, a not uncommon calamity for small aircraft
at low altitudes, but would require extremely improbable
circumstances for a liner - it's unusual enough for
scattered nimbus to attain such height, let alone a
blind alley of them.

I miss-wrote that slightly. A captains emergency authority works in the vertical as well. That might be sensible in a situation like extreme icing, but generally not in storms. We're taught to go around, not above, unless 100% certain of the tops and we can clear those tops by 5000ft.