AirAsia plane missing between Indonesia and Malaysia

[cgzro]

The issue isn't non-linked side sticks - the issue is that a crew can't fly the aircraft safely after a relatively minor failure.

To avoid accidents you work to close off ALL the things that went wrong (thats how we make things so safe). Yes clearly they were unable to hand fly and I am a HUGE advocate of unusual attutude training and thats another huge problem but not knowing what each other was doing was a large contributing factor and lights buttons and talking are inadequate compared to instant feel.

[Meatservo]

I agree completely, though. I'm not interested in bashing this design philosophy or that. The industry just has to start placing more emphasis on keeping handling skills current. .

Or acquiring them in the first place, in the case of these "Cadet" programs...

[Eric Janson]

I've been flying airbus fly-by-wire aircraft for 11 years. I have over 8000 hours on 3 different types including 2600 hours on the A320.

Non linked Side Stick controllers

The Side Stick Controller is extremely intuitive to use. I have never had a "Dual Input" event. I have never had to take over control outside the Simulator (that includes a lot of flights with 250 hour cadets straight out of line training). I've had to issue verbal instructions to stop a rapidly deteriorating situation on a number of occasions and I have commanded a go-around after a bad bounce on landing.

I did a Simulator recently where the other Pilot deliberately screwed things up. This included full aft Side Stick on take-off and flaring at 100 feet. The objective was to be able to take control and recover the aircraft whilst avoiding a "Dual Input" situation. The correct procedure is to push and hold the red take-over push button whilst simultaneously announcing "I have control". Holding the red take-over push button prevents any inputs from the other Side Stick Controller and the verbal command indicates that I am now flying the aircraft and the other Pilot has to release their Side Stick Controller.

It is possible to see the other Pilots Side Stick Controller although looking across the cockpit isn't something that's going to happen in a high workload situation.

While this is different than a conventional aircraft it's not something to be afraid of. I'm always surprised by how many Pilots are afraid of the fly-by-wire airbus.

Air Asia 8501

Here's a link to the full report - a sad read.

http://kemhubri.dephub.go.id/knkt/ntsc_ ... PK-AXC.pdf

In typical Asian fashion they have avoided blaming anyone - even though the entire accident sequence was clearly initiated by the Indonesian Captain.

What's not clearly stated is that there are several different possible "Resets". There are push button switches on the overhead panel, a Reset panel above the head of each Pilot and a circuit breaker panel behind the First Officer.

The Quick Reference Handbook (QRH) has a list of resets and clearly states what is not allowed in flight. Normally the only circuit breakers that can be reset in flight are located on the Reset panels above each Pilot. The circuit breakers behind the First Officer are out of reach for a reason!

To pull circuit breakers on the panel behind the First Officer one Pilot has to get out of their seat to reach the panel. It's buried in the report but they do state that the Captain probably left his seat to pull circuit breakers.

At no point in the accident sequence did the Captain push and hold the take-over push button or state that he had control. Things may have ended differently if this had been done.

[cgzro]

At no point in the accident sequence did the Captain push and hold the take-over push button or state that he had control. Things may have ended differently if this had been done.

Fear and confusion no doubt. Something simulators dont simulate very well.

Anyway we have a couple of real life counter examples that demonstrate the problem where hundreds of people died at least partially because of confusion in communication with control inputs all to save a few pounds.

Fly by wire is an orthogonal discussion.

[Eric Janson]

Fear and confusion no doubt. Something simulators dont simulate very well.

Anyway we have a couple of real life counter examples that demonstrate the problem where hundreds of people died at least partially because of confusion in communication with control inputs all to save a few pounds.

I agree with the first point.

Confusion in the cockpit regarding control inputs isn't something that affects only fly-by-wire aircraft.

Here's an example of a similar event in a conventional aircraft.

https://en.wikipedia.org/wiki/EgyptAir_Flight_990

[boeingboy]

Except that was a deliberate act.

[Rockie]

At no point in the accident sequence did the Captain push and hold the take-over push button or state that he had control. Things may have ended differently if this had been done.

Fear and confusion no doubt. Something simulators dont simulate very well.

Anyway we have a couple of real life counter examples that demonstrate the problem where hundreds of people died at least partially because of confusion in communication with control inputs all to save a few pounds.

Fly by wire is an orthogonal discussion.

Suggesting FBW technology contributed in any way to this accident or the Air France completely misses the mark and worse, diverts attention away from the real cause. Poor training and inadequate pilot standards. In both those cases the flight control system provided a means of recovering had there been at least one properly trained and capable pilot sitting there. The unconnected side sticks are not an issue as a simple glance sideways will tell you whether the other pilot has his stick forward or back, and there are plenty of other clues in how the aircraft is performing.

In a full stall like they had the airplane will be shaking like crazy, the nose will be freakishly high compared to what they're used to and there will be a huge rate of descent. How can any properly trained pilot mistake that for anything but what it is and not jam the stick forward?

And FBW does much more than save a few pounds, the weight savings is the least of it.

[cgzro]

Fbw had nothing to do with the discussion of tactile feedback which should be required FBW or not.
There have been several horrific accidents where two pilots do different things but dont know it despite "training" and buttons/lights etc. Not sure what the objection to improving with tactile feedback is. Clearly it would help so why object? What exactly is gained by not providing it ? I.e we know the downside (every now confusion from supposedly well trained pilots - thats clear from the FDRs).. so whats the upside thats so important to defend?

[Rockie]

Fbw had nothing to do with the discussion of tactile feedback which should be required FBW or not.There have been several horrific accidents where two pilots do different things but dont know it despite "training" and buttons/lights etc. Not sure what the objection to improving with tactile feedback is.

I have no objection, I just don't think it's necessary. When there is no feedback there is a different mindset and awareness that is supposed to be trained into you that these fellows obviously didn't get. Would you let yourself plummet out of the sky for three minutes in a deep stall at 10,000 fpm+ assuming the other guy is doing everything properly without checking, or taking control yourself to try and recover?

Of course not. It's not the stick's fault.

Edit:

You know, the Airbus stick is just one of a very large number of systems anybody flying them needs to know about just as there are in any modern automated airplane. These systems are responsible not just for the efficiencies airplanes have today but are in large part directly responsible for the incredible safety record. But there are pilots who don't understand that knowledge of those systems and being able to use them correctly is just as important as being able to manually fly the aircraft. The old Airbus joke about "what's it doing now" is funny, but in reality represents a failure on the pilot's part. It means he better hit the books and bloody well find out what it's doing so he's driving it instead of the other way around.

[cgzro]

I just don't think it's necessary.

The FDR records disagree with you on at least two occasions involving large numbers of deaths and no doubt there are non fatal incidents with similar confusions that are never reported.

You can have ALL the benefits of FBW with links or tactile feedback so thats not a reason to not provide it. So why not? it has to be weight / cost savings.

[Rockie]

I just don't think it's necessary.

The FDR records disagree with you on at least two occasions involving large numbers of deaths and no doubt there are non fatal incidents with similar confusions that are never reported.

You can have ALL the benefits of FBW with links or tactile feedback so thats not a reason to not provide it. So why not? it has to be weight / cost savings.

You'll have to ask Airbus why they didn't include it. Regardless, scapegoating the stick system diverts attention away from the real problem which was training. You didn't answer my question - would you have sat there like a turnip for three minutes and not confirmed what the PF was doing (all you gotta do is look right) or tried to recover yourself even with a non-feedback stick?

I read one news report saying it requires a "huge" effort to look around to see what the other pilot is doing. No it doesn't - you turn your head to the right and lean forward a bit. Not hard at all, and since you're about to die in 3 minutes do you have anything better to do at that moment? Even if you don't feel like turning your head you could simply push and hold the red button to lock the other stick out and do it properly. Then you know.

[Eric Janson]

You can have ALL the benefits of FBW with links or tactile feedback so thats not a reason to not provide it. So why not? it has to be weight / cost savings.

As you add complexity you increase the failure cases - all of which have to be taken into account.

Linking the side sticks will introduce the possibility of a single failure affecting both side sticks - again this will need to be taken into account.


I'm curious Peter - what is your experience with airbus fly-by-wire technology? Have you flown any fly-by-wire airbus or tried out a simulator?

[cgzro]

You didn't answer my question - would you have sat there like a turnip for three minutes and not confirmed what the PF was doing (all you gotta do is look right) or tried to recover yourself even with a non-feedback stick?

I like to think I would have reacted rationally and used all the feedback available to me but until you are scared and confused you cant predict your reaction, nobody can. Training is very helpful but smularors for example do not show most pilots natural reactions to unusual G. Linking just makes it easier to get it right.

I read one news report saying it requires a "huge" effort to look around to see what the other pilot is doing. No it doesn't - you turn your head to the right and lean forward a bit. Not hard at all, and since you're about to die in 3 minutes do you have anything better to do at that moment? Even if you don't feel like turning your head you could simply push and hold the red button to lock the other stick out and do it properly. Then you know.

I agree unless under a lot of G or its dark or there is smoke its not a big effort.
BTW What happens when both push and hold? Who wins?

Listen im not claiming linked sticks will cure all problems of poor training Im just saying it requires extra thought when you just lost a bunch of your brain power. Seems sensible to eliminate the added confusion. So far Ive not heard a good reason besides cost and im happy to concede that better well trained pilots can survive without it but the design has to work for the lowest common denominator too.

[cgzro]

As you add complexity you increase the failure cases - all of which have to be taken into account.

Exactly, by going unlinked additional complexity required first new logic to average the inputs, then buttons, then lighted arrows. All of those add failure cases.

Note that we added stick shakers and pushers, why was that ok but feedback is not ok?

Linking the side sticks will introduce the possibility of a single failure affecting both side sticks - again this will need to be taken into account.

Its a trade off of course but we do have 100 years of experience doing it.

I'm curious Peter - what is your experience with airbus fly-by-wire technology? Have you flown any fly-by-wire airbus or tried out a simulator?

I have only mucked about in things like a 747 simulators at LHR but not airbus. I have howerver a lot of time in unusual and sometimes very scary attitudes where you cant see the other pilots stick and they often dont do what they say and think they are doing. I extrapolate from this and can well understand getting into trouble without feedbaack.

[Rockie]

BTW What happens when both push and hold? Who wins?

I've never tried it myself, but even if both pilots tried to push the button at the same time the system will electronically register one button push before the other and that person will have control until he releases the button. After 40 seconds the other stick is latched out and the button can be released, but a subsequent push of the button on the latched out stick will reengage it.

You know I actually think linking the sticks to move simultaneously is a good idea for obvious reasons given our human strengths and weaknesses, but Airbus may actually have a reason for not doing that as I've found they have a good reason for doing pretty much everything. It may be with their configuration of FBW linking sticks may make it unacceptably difficult to disable the other one which is absolutely necessary. Also except for takeoff and landing the stick works by commanding a load factor in pitch and a roll rate laterally (as opposed to simple control surface deflection), and it may be the sensors for that are incompatible with linking the sticks - I just don't know. As I said you would have to ask Airbus exactly why they did it like that.

But my main reason for arguing the point is that inadequate training is the real reason those accidents happened - not the stick configuration. I hate to see the eyes dragged off the ball to what is a minor consideration easily overcome with proper training.

[mbav8r]

Forgive my ignorance and also if I missed it if it has been brought up already but is there a "pusher system" incorporated in the FBW system? If so, how does it work, ie; can it take the stick out of your hand like the traditional yoke could? Would there be added forward pressure to the stick to indicate forward stick is needed?
Also, I'm no engineer, but to me a servo that could provide feedback could also be disabled from the lockout button. Seems so simple so I'm sure that can't be the solution, could it?

[ahramin]

The FBW logic in Normal Law gives you a 1.0 G load with neutral stick, up to max positive or negative G for full aft or forward stick until you get slow enough to enter V Alpha Prot, which turns the stick into an AOA control. Full aft stick gives max AOA. So no traditional pusher but if you get that slow, neutral stick will prevent any higher AOA by pushing the nose down as you slow down and max aft stick will push the nose down just enough to maintain max AOA as you slow further. As a side note in between V Alpha Prot and Alpha Max, the autothrust will command TOGA as well. At no point in any of this will there be any control feedback to the pilot for the stick or thrust levers. Your speed will be in the Red, and your autothrust FMA will read TOGA LK, but the controls are not pilot information devices, they are pilot input devices.

This makes escape maneuvers for terrain or windshear amazingly easier and better performing than traditional controls, but must be understood.

Keep in mind people that on anything with hydraulic controls, artificial forces to the controls are just that ... artificial. I have never flown a hydraulic powered aircraft that did a good job of reproducing the control feel that we enjoy on smaller aircraft. Maybe the 777 or 787 pilots would disagree though, I haven't heard anything either way.

[cgzro]

This is interesting, "active sidesticks" are beng used in the JSF and moving to Gulfstream. note the reasons sited include better situational awareness of what autopilot and other pilot are doing.


http://aviationweek.com/technology/acti ... cial-debut

The principal reason for using sidesticks is ergonomics. Removing the control yokes allows for larger flight displays and—in the latest cockpits—pilots can be moved closer to the instrument panels, allowing use of touchscreens. The biggest drawback of the “passive” sidesticks now used in civil aircraft is the lack of control feedback from the aircraft or the other pilot.

But the transfer of “active inceptor” technology to the commercial sector from the military is helping to overcome that objection. Active sidesticks that provide tactile and visual feedback in response to pilot and autopilot commands are used in the Lockheed Martin F-35 Joint Strike Fighter and Sikorsky CH-53K heavy-lift helicopter. Now Gulfstream is the first civil manufacturer to adopt active sidesticks, for its new G500 and G600 business jets.

I agree that training is the major problem, no question given the high profile incidents of pilots that crash perfectly airworthy planes. Clearly we have made great strides in metalurgy, propulsion, navigation etc to the point where failures are getting very rare. As to pilots clearly the systems have to be as simple and intuitive as possible and make the best use of human senses as possible. For example tactile feedback and reaction times are nearly twice that of sight feedback and reaction as measured with simple experiments, no doubt its even worse when the sight you see is confusing.

Im also not sure why airbus decided not to provide tactile feeback by linking or other means but my guess is it started simply enough, no buttons, lights etc, just simple average and the view that training was all that was required. Then faced with the occasional confusion they added the buttons, lights, lockouts as after thoughts.

Of course with any system the ultimate safety depends not only on the pros and cons of the new system but how it interacts with other factors. For example having more display visible in front of you or sitting closer to the displays could have benefits that outweigh the lack of feedback. Obviously its very hard to know without controlled experiments of different configurations for hundreds of hours etc. and I dont thinks that been done.

Anyway we are not going to solve anything here and as fun as it is to debate its the the insurance and lawyers that will provide the most significant feedback to the industry.

[trey kule]

You didn't answer my question - would you have sat there like a turnip

Well despite your anti-turnip remarks, I think the issue is a bit more than that.

Experience counts. And I don't mean dozing for dollars as one meanders through the sky.
Training will never be a 100% substitute for having the very basic flight skills implanted in the brain, so when panic arrives, one can think.
The second thing about training is that current training programs including CRM actually go against teaching pilots to think..

As long as the feeder stream is from pilots that have virtually zero flight experience before sliding into a seat on this type of equipment, and the primary training is to call for a checklist, we are going to see more and more accidents where the pilots make a rather benign abnormal situation, a fatal one.

in today's highly automated cockpits, pilot skills deteriorate. But with the new current crop of seat warmers, many never had the basic skill or experience to start with. The FAA mandated more flight experience before allowing access to the airline world. The majors in Canada, at least so far, have been able to find experienced pilots, but that seems to be changing some and less experienced pilots are moving in. In China, there is a move to restrict some FO's to that position for their entire career. Not sure how that will ultimately play out in an emergency. Other Asian countries take whatever they can get (Cathay Pacific being an exception , and a good example of how it should be done)

The simple truth is that today the aircraft are better at flying than the pilots. And when the aircraft has an oopsie and the pilots step in the real problems arise.

[Rockie]

Training will never be a 100% substitute for having the very basic flight skills implanted in the brain, so when panic arrives, one can think.
The second thing about training is that current training programs including CRM actually go against teaching pilots to think...

?

I don't know where you've been getting your training, but you either aren't understanding it or should maybe find a new place.

[trey kule]

Are you telling me Teddie's dog grooming emporium and sim training facility is below standard?
I will have you know they have the very latest flight sim X upgrades...Sub standard, indeed.


Actually, that is not just my opinion. The advanced training programs are , or were, designed for a certain entry level of experience. They are not always getting that, and as such are trying to substitue training for experience. Agree with me or not. My opinion,and the opinion of some others in the industry, is it is not working. They completely ignore, as they probably should, the fact that some pilots entering the training program are lacking in experience, and at the bottom of the curve in basic flight skills. It is an issue that is being addressed as far as I know.
Like trying to teach advanced wood working to someone who has never used any tool or built anything in their life.
CRM, from an international perspective is being reviewed for its relevance as I write this, and I think you may see a change in the future as to how we deal with the decision making process that is being taught today....and yes...they are actually teaching pilots how to make a decision.

If you have a strong, imprinted set of basic flight skills, and experience these advanced programs are meaningful. If not, when the fur hits the fan, things can go terribly wrong..Think pull back rather than push forward to recover from a stall.

I suppose, like all things, we can disagree on this, and we will have to wait a few years to see how it plays out.
BTW. I think maybe you missed the essential point of trying to do advanced training for those who do not have the basics really firmly imprinted.

[Rockie]

If you are only talking about cadet programs I'm inclined to agree with you, but I have no first hand experience with it myself. However it seems to me the first time you experience fear and the real weight of your decision making on a dark and dirty night shouldn't be with 100+ people in the back. That's a lousy time to find out your survival instinct is under-developed.

[cgzro]

Actually, the root cause is almost certainly related to GA costs. I cant imagine any young pilot not wanting to blast off for a weekend in a little plane to learn and have an adventure but its just not affordable. Like many others I made a lot of mistakes doing that sort of trip which left indellible lessons but only risked one or two people max.

Its a shame the cadet programs cant have a few 150s or whatever so the kids and other pilots can use them at very low rates.