As sloppy as it gets without crashing...

[3WE]

1. So, pilot overload = no CRM?

2. But really, what needs to be fixed here? The means of control or the discipline of control?

1. I'd remove the word pilot and add a list of 20 things that go out the window... Like the correct altitude selection, the right level off inputs and the ability to hear the crossed control warning or see the red light.

2. MCMs point is interesting. The non linked control sticks begged the question way back in the design days, and now we have a list of incidents where it seemed to be an issue.

[MCM]

I only agree with you that the Airbus sidestick control is more dangerous when CRM goes out the window and panic takes over and that isn't go to happen if you are properly trained and stay ahead of the approach.

No arguments there - except we see accident after accident with the common theme that 'DUAL INPUT' is screaming in the background while they happen.

'Taking Over' from another pilot is an extremely rare event. Most pilots would have had more engine failures than times they've used the priority button to establish priority.

Airbus gives you a verbal "dual input" alert, a warning light on the glareshield and a priority take-over button. I mean exactly how idiot proof do we have to get here?

Once again - its clear that the light and the alert aren't working. You can call people idiots all you want, but the idiots are those who sit around wondering why we keep having accidents with dual input, and try to change it by invoking 'protocol' and SOPs, when it is the lack of this awareness and protocol that got them into the situation in the first place!

But really, what needs to be fixed here? The means of control or the discipline of control?

We have been working on discipline of control for what, 20 plus years? The point is that the Dual Input alert is there solely for the situation that those protocols break down, except it seems that when you get to the situation that you need it, the overload factor is so high that the pilots don't perceive the warning.

[Gabriel]

MCM, I agree with you, but what you suggest in place of red lights flashing on the glareshield and a voice shouting "Dual input!"?

Being these sidesticks a simple spring-loaded joystick with no mechanical linkage, it will be extremely difficult to do things like linking both sidesticks either with a real mechanical linkage or with an artificial feedback. These things don't have even a stickshaker. Anything beyond a cellphone-like "vibrate" seems implausible.

Ok, I have an idea. An "mild" electroshock in the stick?

[3WE]

I

MCM, I agree with you, but what you suggest in place of red lights flashing on the glareshield and a voice shouting "Dual input!"?

Being these sidesticks a simple spring-loaded joystick with no mechanical linkage, it will be extremely difficult to do things like linking both sidesticks either with a real mechanical linkage or with an artificial feedback. These things don't have even a stickshaker. Anything beyond a cellphone-like "vibrate" seems implausible.

Ok, I have an idea. An "mild" electroshock in the stick?

The ironing here is incredible.

The super crazy ultimate advanced (and cheap composite crackerbox) Piper cub contains mechanically linked controls...

Perhaps Air Bus could do some similar advanced engineering.

More ironing is that it would be pretty complicated to get the cables strung from the left to right side of the plane...you'd have to dodge all the electronics...and an artificial feed back system...HUGELY complicated.

BUT...BUT...maybe, on the floor between the pilots legs, there could be nice big 3 ft long iron pipes with a plastic hand grips on the top...there'd probably be ways to rig wires between them...and have the transducerator in the center console...

[MCM]

Being these sidesticks a simple spring-loaded joystick with no mechanical linkage, it will be extremely difficult to do things like linking both sidesticks either with a real mechanical linkage or with an artificial feedback. These things don't have even a stickshaker. Anything beyond a cellphone-like "vibrate" seems implausible.

Why difficult? The technology exists, and is active use. Gulfstream use it on the G500 and 600.

BAE make the systems, and they use them to provide stickshaker/pusher, artificial feel, and 'soft stops' to show aircraft limits (not so applicable on an Airbus, but the other features are), and most relevantly, to 'link' side sticks. Gulfstream advertise the fact that their sidesticks, while electrical, are 'linked' as a feature in even their basic advertising material, showing how much of a selling point they think it is.

I must admit, I did think of the electric shock idea too... wouldn't work in Japan though! (They fly in gloves). Probably best avoided if possible!

[Evan]

Why difficult? The technology exists, and is active use. Gulfstream use it on the G500 and 600.

MCM, I agree with you, but what you suggest in place of red lights flashing on the glareshield and a voice shouting "Dual input

Airbus should make similar evolutions to their control interface. I would like to see it on the A350 at least. But how far should they need to go?

I've no doubt Airbus could employ robotics technology to create an artificially 'linked' sidestick but this would involve undesirable complexity and the risks that introduces. More importantly, it would be a statement to the industry that the problem of dual-inputs—despite all the warnings currently provided—are a significant problem and that pilots cannot be relied upon to have the focus and discipline to avoid them. In other words, it would be an acknowledgement that the modern aircraft cockpit must be increasingly 'idiot-proof'*.

As I've said many times here, Airbus, with all of its immutable protections, never set out to design an 'idiot-proof' airplane. Quite the opposite, they assumed that the industry had reached a level of sophistication that could handle the next level of flight control—the same one used on current spacecraft.

But the industry, being an industry and placing growth and revenue as its highest priority and therefore always looking for places to control spending, seems to have never raised its pilot-sophistication expectations to that level. What Airbus had assumed—that the industry would parallel the new technology with a new level of training involving higher standards with regard to systems and protocol (as found in the space program)—seems to have faltered as operators see the new technology as merely an opportunity to undertrain pilots instead. You have all these envelope protections in a streamlined, automated cockpit... just train pilots to push the buttons, follow the displays and move the stick around in brief, critical phases of flight. That is how the industry let Airbus down.

Instead of being able to recite autoflight mode transitions, control law reversions and systems architecture by heart, we have pilots asking "What's it doing now?". It's demoralizing.

AIrbus created their flight control system around a core idea: we now have fly-by-wire; we've eliminated the mechanical linkages; we can now do away with a significant part of pilot workload, i.e. the need to constantly retrim for changes in speed, attitude and configuration. Reducing pilot workload there would allow pilots to focus on 21st century priorities instead, things like navigating ever more crowded airways, managing more options in automation and maximizing efficiencies. By removing the need to trim, they also removed the need for the feedback that creates. Of course, pilots need awareness, so that awareness was provided in a new way, one that requires mental rather than physical attention.

In short, a new means of flight control required a new pilot sensibility and a new piloting discipline. But instead, we have seen an alarming level of technological ignorance and lack of protocol discipline as 'seat-of-the-pants' pilots continue to favor a physical cockpit experience based on the jet-age core competencies of a 'stick-and-rudder' instinct. The sleek Airbus cockpit seems to have a tendency to lull such pilots into forgetting to fly the plane at all.

You could argue that this is because the aircraft isn't well suited for pilots or you could argue that pilots are not being trained for the new requirements and the new demands of the new technology. Either argument is strong, but it comes down to the direction we want to take. Do we want to take a conservative approach and forego the advantages of innovation or do we want to take the ambitious approach and commit to the investments and demands of creating a new piloting mindset? Right now, it is tragically obvious that the industry is not willing—nor required— to invest and commit enough for the manufacturers to safely innovate.

Aviation is a field that favors a strongly conservative approach, if for no better reason than self-preservation. It attracts stubborn minds and a 'gut-instinct' mentality', perhaps for the same reason. While automation in the space program evolved to suit the 'pilot-scientist' role of the astronaut, the aviation industry can't shake off the 'pilot-cropduster' tradition. For Airbus, who had envisioned a parallel between spaceflight technology and commercial aviation, this must be endlessly disappointing.

So, despite the superlative safety record of their aircraft, stubborn human factors continue to conflict with the human ideal of progress. I submit that this is mostly a failure of pilot selection and training, exacerbated by the fatigue of abusive scheduling and lack of manual flight practice (at least in the SIM), but I also concede that a means of flight control that returns to the physical interface of the control column and places the workload of trim back upon the pilot is more directly informative to pilots in extreme situations.

So should we increase pilot workload in 99.999% of scenarios to provide a better means of feedback for those .001% scenarios? Or should we do a much better job of training pilots to deal with those .001% scenarios in a new, less physical, more cerebral way (and keeping in mind that 99% of those .001% scenarios are entirely avoidable through better routine discipline)?

Ultimately, the human mind is what limits us. What can we expect of it? Can we expect it to maintain the necessary mental focus under the shock of extreme situations after being immersed in the tedium of routine operations? Airbus could lower their expectations by adding a more physical aspect to their control interface. We should recognize the message that sends and what it says about humanity. Even on the edge of aerospace technology we are, it seems, still locked in that 15th century debate between striving for Utopia and accepting the Machiavellian reality.

* The colloquialism 'idiot-proof' meaning 'gross pilot-error resistant', not necessarily referring to pilots who make those errors as idiots.

[Gabriel]

Evan, you lost me with that post.

How would having linked sticks and force feedback increase pilot workload?

Boeing approached FNW in a way that also reduces pilot workload a lot while keeping what is, in my opinion, a more human-friendly interface.

That doesn't mean that Boeing pilots will not make mistakes too, but in general you see less complaints regarding the cockpit design and system philosophies being a strong factor in those accidents (even when you have the Turkish and Asiana counterexamples).

There is a video out there where one pilot says something like "In a boeing, you can be flying with all the automation on, and you can keep your hands slightly resting on the yoke and throttles, and you turn back to talk to the flight attendant who is bringing you the food, and you know what the plane is doing. In an airbus, you need to be looking at several instruments and indications to know what the plane is doing." And the same can be say of what the other pilot is doing. I think there is a point there.

Humans are humans and naturally have known weak points. I think that the Airbus' philosophy doesn't address some of these week points so well.

Also, I think that, opposite to what you'0ve said, the problem is not "seat of the pants and stick and rudder pilots". Rather the opposite. Pilots that, since very early in their carreer, were trained as "system managers" and when the system isn't doing what they want (most of the times because of the fault of the system manager in the first place) they breakdown.

All companies are doing fly-by-wire these days. Boeing, Embraer, Gulfstream, Cessna (business jets) and even small GA piston singles which, being fly-by-cable, are starting to incorporate FBW features in their control systems (like Boeing-like soft envelope protections and "Level my plane!" buttons). All of them are keeping the yoke or sidestick linked. I don't know if all, but most of them also keep feedback in the thrust levers.

[Evan]

Evan, you lost me with that post.

How would having linked sticks and force feedback increase pilot workload?

I didn't say that. I said removing the need to trim reduced pilot workload. I meant that if AIrbus changed back to a system requiring trim (and thereby providing physical out-of-trim feedback) would also add that workoad back in.

I would advocate a system that provides artificial sidestick linkage if you could convince me that it would reduce, rather than add, elements of risk.

That doesn't mean that Boeing pilots will not make mistakes too, but in general you see less complaints regarding the cockpit design and system philosophies being a strong factor in those accidents (even when you have the Turkish and Asiana counterexamples).

As I said, I think that is partially because of conservative resistance to new methodologies. The "if if ain't broke don't fix it" mentality. But the pilots in those incidents were lacking the proper training for the Airbus cockpit.

Also, I think that, opposite to what you'0ve said, the problem is not "seat of the pants and stick and rudder pilots". Rather the opposite. Pilots that, since very early in their carreer, were trained as "system managers" and when the system isn't doing what they want (most of the times because of the fault of the system manager in the first place) they breakdown.

No, that's not opposite to what I've said. I agree that too many (too many being 1) pilots are not being trained properly to fly the Airbus. That means fluency on modes and system reversions. That also means the basics and every other aspect of airmanship, particularly protocol and CRM. But you can't henpick these things. ALL of these things are required. You can't simple have stellar airmanship and expect to deal with a systems issue. The Airbus pilots who keep 'having trouble with the cockpit design' are all lacking that complete training.

[Gabriel]

I think that the problem today (in all modern airliners) is that the systems became too complex, and pilots don't actively use 99.9% of it 99.9% of the time.

Airbus philosophy just makes it more noticeable by reducing the feedback and increasing the "what is it doing now" factor.

And I don't agree that artificially making the plane with zero stability in speed and AoA reduces the pilot workload. It may reduce the physical workload but it increases the mental one, trying to keep pace with what the plane is doing.

The trim is a good example. The trim wheel doesn't move anymore. But you have a scenario where the plane trims all the way up and then it gets stuck there (by design) and the pilot has to figure out this condition while trying to recover from a stall. It's very hard!!! We will agree that the plane should have never reached this condition to begin with. But if (when) it does, the next layer is weakened.

I prefer that, if the plane decides to trim up, it will turn a big black wheel with white stripes and a clacking sound as it turns.

[Evan]

And I don't agree that artificially making the plane with zero stability in speed and AoA reduces the pilot workload. It may reduce the physical workload but it increases the mental one, trying to keep pace with what the plane is doing.

Isn't that was the 'six-pack' is for? It's all one screen now.

The trim is a good example. The trim wheel doesn't move anymore. But you have a scenario where the plane trims all the way up and then it gets stuck there (by design)...

Then you have a serious pilot training issue.

[Gabriel]

Isn't that was the 'six-pack' is for?

No. The six pack doesn't tell you if the plane is increasing the power or trimming up or pulling up. The "traditional" feedback (real or simulated) does.

Then you have a serious pilot training issue.

Please elaborate.

[ATLcrew]

In short, a new means of flight control required a new pilot sensibility and a new piloting discipline. But instead, we have seen an alarming level of technological ignorance and lack of protocol discipline as 'seat-of-the-pants' pilots continue to favor a physical cockpit experience based on the jet-age core competencies of a 'stick-and-rudder' instinct. The sleek Airbus cockpit seems to have a tendency to lull such pilots into forgetting to fly the plane at all.

I disagree. In fact, I think it's almost the opposite. It is because of the lack of physical feedback that the Airbus setup forces the pilot to pay that much MORE attention to what the airplane is actually doing. Now, if you have a pilot who is of the "lullable" type, that's a separate issue. Then again, I just fly these things for a living, what do I know?

You could argue that this is because the aircraft isn't well suited for pilots or you could argue that pilots are not being trained for the new requirements and the new demands of the new technology. Either argument is strong, but it comes down to the direction we want to take. Do we want to take a conservative approach and forego the advantages of innovation or do we want to take the ambitious approach and commit to the investments and demands of creating a new piloting mindset? Right now, it is tragically obvious that the industry is not willing—nor required— to invest and commit enough for the manufacturers to safely innovate.

Two questions:

1. Who's "we"?
2. What "new piloting mindset"? What "piloting mindset" can there be except to operate the aircraft safely and competently?

Aviation is a field that favors a strongly conservative approach, if for no better reason than self-preservation. It attracts stubborn minds and a 'gut-instinct' mentality', perhaps for the same reason.

Again, I disagree and again, I think it's almost the opposite. Stubborn minds don't make it far in aviation, not this day and age.

[3WE]

...What "piloting mindset" can there be except to operate the aircraft safely and competently?...

There are two:

1. Cowboy stick and rudder improvisational fundamental airmanship with no regard for proper procedure.

2. Total well trained always correct procedural regurgitation wasting no mental capacity nor distraction on any sort of broadly applicable fundamentals.

The two are exclusive of each other.

(I am exaggerating a little but I think I've basically captured the two mindsets that Evan sees.)

[Gabriel]

What "piloting mindset" can there be except to operate the aircraft safely and competently?

There are two:

1. Cowboy stick and rudder improvisational fundamental airmanship with no regard for proper procedure.

2. Total well trained always correct procedural regurgitation wasting no mental capacity nor distraction on any sort of broadly applicable fundamentals.

+1 / Like / Thumbs up /

[MCM]

More importantly, it would be a statement to the industry that the problem of dual-inputs—despite all the warnings currently provided—are a significant problem and that pilots cannot be relied upon to have the focus and discipline to avoid them. In other words, it would be an acknowledgement that the modern aircraft cockpit must be increasingly 'idiot-proof'*.

As a statement - yes. Its a statement that needs to be made. Are you suggesting we shouldn't have stick-shakers? After all, we have visual warnings, and pilots should have the discipline to avoid the stall. Should we not have overspeed warnings? After all, the pilot should have the focus and discipline to avoid that scenario too.

Can I ask that you please turn off the 'focus and discipline' argument, its getting very tiring. Surely you can see that it doesn't matter how much you train crews, nor how carefully you select them, you are always going to have moments of loss of both of these things, and that is where clever system design comes in to help protect the pilot. We're not talking about day to day design issues here - we're talking about instances when we have a breakdown of these things, and the aircraft can be better designed to help protect us.

You have an unrealistic ideal about just what a pilot can, and should, be. We're not talking about selecting astronauts, where you have a large pool to choose from, and a very small number required. We're talking about hundreds of thousands of people being selected, and not all of those are going to be the ace of the base, nor are all of them going to always be performing at the top of their game. We are HUMAN, and we have failings and limitations, and always will. We need to design cockpits and systems that work with our strengths and weaknesses, not against them.

The Airbus pilots who keep 'having trouble with the cockpit design' are all lacking that complete training.

Or they struggle to cope with a cockpit that is not intuitive for them to use.
Training is always going to be limited. Its just the world we live in. If the cockpit design is such that it can't be learned thoroughly in the equivalent time that another manufacturers cockpit can be learned, then there's got to be questions about its practicality in the modern world.

Aviation is a field that favors a strongly conservative approach, if for no better reason than self-preservation. It attracts stubborn minds and a 'gut-instinct' mentality', perhaps for the same reason. While automation in the space program evolved to suit the 'pilot-scientist' role of the astronaut, the aviation industry can't shake off the 'pilot-cropduster' tradition. For Airbus, who had envisioned a parallel between spaceflight technology and commercial aviation, this must be endlessly disappointing.

We favour conservatism because it is not old aircraft types that are having incidents, its the new systems with the new technology that are!

You've hit the nail on the head - the space technology 'evolved to suit the 'pilot-scientist'' - so why would you then try and force the same technology onto a good old fashioned airline pilot? Cockpit technology needs to evolve to suit the airline pilot, not some elite space pilot that takes years and years to train.

Lets be realistic here - airline pilots are going to become less skilled as airlines push for lower wages and the shear demand for pilot numbers increases. Cockpits DO need to become more idiot-proof, because we're going to be putting lower and lower standard pilots into these aeroplanes, whether you or I like it or not.

In short, a new means of flight control required a new pilot sensibility and a new piloting discipline. But instead, we have seen an alarming level of technological ignorance and lack of protocol discipline as 'seat-of-the-pants' pilots continue to favor a physical cockpit experience based on the jet-age core competencies of a 'stick-and-rudder' instinct. The sleek Airbus cockpit seems to have a tendency to lull such pilots into forgetting to fly the plane at all.

So what you're saying here is that we've created a brand new type of aeroplane control system, and then blaming the training and the pilots for not understanding it?! Perhaps the 'lack of protocol discipline' is because that is an inherent weakness in humans? The fact that we KNOW that it is a weakness in humans, and yet design an aircraft for which safety is fundamentally based on maintaining that discipline, could be considered questionable design, rather than poor training!

Airbus could lower their expectations by adding a more physical aspect to their control interface. We should recognize the message that sends and what it says about humanity.

How is this 'lowering expectations'? I find this a bizarre sentiment.

What message does it send about humanity? That we are UNABLE to concentrate on repetitive data for hours on end, and that our attention wanes? We KNOW THIS ALREADY!

Adding a more physical element to their controls shows due consideration for the very nature of our humanity - that we have multiple senses, and that using non-visual, and instinctive, methods of linking the pilot to the aeroplane is a very strong way forward. There's a reason thats the direction the military is taking, and why feedback through side sticks has been developed. They need the pilot to be linked to his aircraft through more than visual/active mental means.