AF447

I agree very much with @Raven. I have never flown an Airbus, but I have experienced fatigue at the end of a long flight with time differences.
And I have experienced a tunnel vision or brain overload when nothing seems to make sense, anymore.
I fear, many on here are in danger of pre-judging a case before it had occurred.
Hindsight is a wonderful thing but what the courts will be looking at are things like could X, Y, Z have been foreseen and corrected.
If they could have been foreseen then the judges will find against AF and/or Airbus.
Sadly some pilot errors cannot be foreseen and therefore can not be corrected with SOPs. Pilots are human and sometimes even multi crew training cannot totally eliminate errors. We know that.
Aircraft, are designed, constructed, tested and tested again, then computer modelled to identify the what ifs and problems are designed out.
But as an American politician famously once said “we don’t know what we don’t know”.
In my experience of Air France, knowing many of their flight crew and test pilots. All were trained in flight sims on the particular aircraft they would be crewing on. They would be rechecked every few months and unusual situations drawn up to train on. All accidents and incidents around the world on that particular type/model of aircraft would have been fed into the training programmes.
Does that mean that all situations have been trained for? No, because there are situations which no one has foreseen.
It’s the same with aircraft manufacturers and maintenance companies.
But despite all this, things do go wrong, stuff happens. We know from experience/history that most air accidents/incidents are down to the holes in a Swiss cheese lining up. Hence all training, aircraft design, maintenance etc is designed to stop the holes lining up. Sometimes, if they do line up it results in no more than a minor accident/incident. The airline industry, worldwide, should and does (I trust) investigate these matters without prejudgement of blame.
The media including social media tend to apportion blame without all the facts or without bothering to find them out.
Whether any of the posts here hold the key to avoiding such accidents in future I don’t know.
Meanwhile the families of the victims of that aircrash have questions and a need to blame someone. It is now up to the courts to get answers to those questions of at all possible and if there is blame to apportion, apportion it.
Will the families, get the closure they are looking for? Possibly not.
How many on here, have never got it wrong?
How many ponitificate on the causes of an accident, (eg the footballer going from Nantes to Cardiff) only to be proved completely wrong. The cause being something they hadn’t thought of.

kwlf wrote:

Take the Voyager incident as a counterexample. My understanding is that the captain got his camera wedged between the seat and the joystick. When he moved his seat forward, the seat pushed the camera which pushed the joystick and the aircraft went into a bunt. The captain pulled back but the aircraft didn’t respond. He assumed the problem was with the autopilot and tried to disengage it. Luckily the copilot had an independent sidestick and saved the day, but had the joysticks been connected, it seems to me that this might not have been possible. You win some, and sadly you lose some.

I’m not sure it’s a great example of the benefits of the Airbus philosophy.

Had the aircraft had large centrally-located control yokes which were connected together and gave force-feedback from the controls, the idea that a misplaced camera could cause such a significant inadvertent control input becomes absurd. A stick which gives no force-feedback and isn’t connected to the other is at risk of accidental deflection in a way which will not necessarily be immediately obvious – same point as was made earlier about the importance of throttle lever positions.

It is a shame that redacted report doesn’t give the findings, because it’s interesting reading.

This isn’t really an aviation issue, it’s about overly-arrogant design philosophies which think they’ve captured every possibility and everything will always be fine if only all the ignorant people in the world would think the same way as the designers. I see this every day at work – even when the most fundamental issues are raised these folks will argue black is white in telling you that your problematic scenario only arises because you aren’t thinking ‘correctly’ like them. I prefer to remain humble enough to admit, in work and in aviation, that scenarios will arise which I haven’t thought of and when that happens I want to go back to basics, not kid myself that my design philosophy captures it and deals with it.

Airborne_Again wrote:

Does that mean that you will get less authority from each individual control column if they are disconnected from each other?

THis is type-dependent. On the mechanical types I know (737, 717 and MD-80) yes, you get reduced authority but otherwise demonstrated sufficient for limiting circumstances like a landing flare all other things being nominal. For example, on the 717 and MD-80 the LH+RH elevators are driven by their own servo tabs. RH tab linked to RH control column and LH to LH. Normally movement of one column will move both elevators, but in a disconnected situation only one of them.

The idea is that regardless of where in the control “line” there is a jam (ie camera on the pilot’s feet, ice on the elevator, misplaced spanner in a tail compartment pulley, broken hinge or cable ….) the pilots should still have some reasonable measure of control.

On fully FBW aircraft (ie Boeing 787) I guess it depends where the fault is but in theory you should be able to have full authority. I do not know the details, but will try to find out how it works in such case.

Graham wrote:

Do they? I thought the tab in the leading edge is being pushed upwards – and the circuit completed and the buzzer going off – by the airflow as the aircraft descends at high alpha?

If the forward airspeed becomes very slow, does the tab stop being pushed upwards? I’d assumed it would always be pushed upwards with the airflow detached and the wing descending, unless of course the aircraft becomes inverted.

This has nothing to do with the aircraft descending. It only depends on the AoA. You can have the stall warning going off in a climb. The stall warning is not related to airflow detaching as it is a stall warning which goes off before the wing reaches critical AoA and not a stall detector. (You don’t really need any equipment for that.) The PA28 tab is activated when the stagnation point moves down as the AoA increases. I guess the “organ pipe” on a C172 works in a similar way.

Both the electromechanical sensor in a PA28 and the pneumatic one in the C172 depend on sufficient airflow. If the speed is too low, you will not get a stall warning regardless of the AoA.

Airborne_Again wrote:

This has nothing to do with the aircraft descending. It only depends on the AoA. You can have the stall warning going off in a climb. The stall warning is not related to airflow detaching as it is a stall warning which goes off before the wing reaches critical AoA and not a stall detector. (You don’t really need any equipment for that.) The PA28 tab is activated when the stagnation point moves down as the AoA increases. I guess the “organ pipe” on a C172 works in a similar way.

Both the electromechanical sensor in a PA28 and the pneumatic one in the C172 depend on sufficient airflow. If the speed is too low, you will not get a stall warning regardless of the AoA.

I didn’t express myself well re aircraft descending.

All I mean is that upward airflow relative to the wing leading edge pushes the tab up and completes the electrical circuit. I have not tried it, but I bet you could point a hairdryer at it from some angle below the horizontal and it would go off. I’m thus not convinced the PA28-type has any minimum forward airspeed – if the aircraft were falling vertically with zero forward speed, with the nose anywhere between vertically up and vertically down, I believe it would be sounding – just as surely as it sounds when you push it up with your finger. I don’t see why it needs forward speed.

Graham wrote:

if the aircraft were falling vertically with zero forward speed, with the nose anywhere between vertically up and vertically down, I believe it would be sounding – just as surely as it sounds when you push it up with your finger. I don’t see why it needs forward speed.

You’re right that the PA28 stall warner probably doesn’t need forward speed, but it does need speed, i.e. airflow. I don’t believe the C172 stall warner will work at zero forward speed no matter the sink rate.

Airborne_Again wrote:

You’re right that the PA28 stall warner probably doesn’t need forward speed, but it does need speed, i.e. airflow.

AF447 never lacked speed, it just lacked forward speed – hence the comparison. The point is I believe a simple PA28-type stall warner will continue to sound in a very deeply stalled condition, but the Airbus one did not because the computer thought it knew better. I suppose the computer could interrogate the rate of descent but without airspeed indication it would have no way of distinguishing, on RoD alone, between a steep dive and a deep stall. It could check the angle of attack though, or the attitude indication. Those would tell you clearly whether your high rate of descent was as a result of a nose up or nose down condition. So why didn’t it cross-check an attitude gyro/INS and conclude “yes, this is actually a stall. unbelieveably deep, but it is a stall.”?

Or the wheel spin-up, which is used on some types to auto deploy thrust reversers.

Is this something new? 🤔 Or do you maybe mean speedbrakes and autobrakes (if armed)?

Probably very old. A DC10 crew told me about that, on a cockpit visit, pre 9/11.

Antonio wrote:

Habsheim was a bitter warning

Had Habsheim been flown with the same profile with a B737, the same crash would have happened.