Mr Snuggles

Some interesting comments and observations but if I may can I bring up a couple of points before somebody else does?

I think you may just get a bit of robust feedback on that, so I'll be gentle.

Most pilots on full time contracts these days are working long hours on quite punishing rosters...I know there's this idea we are all "total aviation people" but in reality many don't have time to rush down to the flying club for a bit of GH on a day off...even if him/her indoors would sanction it :sad:. It's also worth taking a look at the cost of flying in some parts of the world, and in any event many a newer pilot's cash is spoken for in repaying training loans.

Ummm.....well....you need be aware of it's limitations (and most of it is very limited) and be very very aware of negative training....I'm not sure a PC/joystick combo has much stick and rudder training value for someone flying a Boeing as a day job......

If more training is the answer then it must be done properly, using realistic equipment and not left to the individual ...and that will cost the industry money.

Gysbreght:

This MIGHT be one of those strange things that MIGHT be involved in the repeat of AF447.

As I see it: The publicity from the AF case, its widespread media coverage, the appaling truth, noone, especially not a French aviator, would have missed that report. I honestly believe that FO here had knowlegde of the causes of AF447 and I betcha he never ever thought that would happen to him. I am sure he believed he would have all the nuts in the bag if **** ever got to that. Yet he too failed. And the Capt failed. In horrifyingly similar circumstances.

There must be something more about this. MAYBE it has to do with ergonomics. MAYBE aural warnings are not optimal. MAYBE it after all is a stick vs yoke discussion. I do not know. MAYBE it has something to do with sensory feedback. The input from pilots experienced in both types of aircraft might be extremely valuable.

What do you think?

wiggy

Yes, I know, I know. Flying is expensive. But is that a reason not to do it? I mean, it IS after all the hand flying skill deterioration that is widely critiqued in a number of threads. I would hope that those critiquing it at least lives up to their own expectations.

And, press the airlines! You are the reason the planes fly at all. Now you have two eerily similar tragedies to shove in the face of any beancounter.
In Sweden, unions would be the way to go forward with this. Unions are not evil monsters, not here anyway.

Home sim software is far from the real deal. Being a humble wheel-born driver I also know this. But as with all home sim things, the procedures are what you want. They don't change; buttons actions don't change.

BTW, would anyone here happen to know how many near misses of this sort that amounts to one tragedy?

flight sim?? give me strength
i think the post above about alternate law and degradation is onto something there. i have flown with A and B types and IMO, both are excellent planes. However, when things go wrong, airbus is much much more comlex than its American equivalent.

What you didn't read in the FINAL report:

23:15:36 the fourth failure of the Rudder Travel Limiter Units triggered ECAM message....and 3 seconds later a sound similar to a seat motion is heard..

It seems also that Airbus picked up on my observations that some of the times do not match up.

The crucial point here (other than the fact that having stalled the airplane they failed to recover from it) is that "some" may believe that this report has been carefully worded to avoid stating that one of the pilots might have actually left his seat to re-set a CB that was not called on the ECAM to be re-set. Throughout the report the investigators have chosen to use the phrase "AUTO FLT FAC (1 or 2) FAULT which was triggered by FAC (1 or 2) FAULT followed by signature of erratic fluctuation of the parameters of the components controlled by FAC...".

Airbus on the other hand wanted the statement amended to " triggered by FAC 1 (or 2) being de-energised concomitant with FDR signature of unavailability of parameters computed by FAC 1 (or 2).[/QUOTE].

There is a big difference between the two statements, especially when one considers the definitions of fluctuating and unavailable.

Perhaps that is why not much of the CVR transcript was included in the report.

I have a question for those who know more than me. Why would this plane be flying around so long with this known fault? Why would AirAsia not fix this issue? Cost, Time, both?

because the exakt fault was not known. Spurious computer errors do happen frequently. If resetting solves the issue, no further action is taken. If the fault can not be reproduced on ground (e.g. due to a temperature related problem, like thermal stress on a cracked solder joint) the units are declared "serviceable" and end in an aircraft again. Just when the frequency of one specific fault indicates a real issue which can be pinpointed to one specific system, more action is taken. This would have probably been the case after this flight, on whicht the unit would have failed 6-7 times or more...

@sooty3694
It's a philosophical call regarding how much to include "based on speculation" vs. based on facts that can be proven.

E.g., the sound "similar" to seat motion was speculative (it may or may not actually have come from seat motion) -- and in any case unnecessary to be included, so it was omitted from the official transcript.

Again the Indonesians were (correctly) conservative in this case, since the FDR data actually shows the parameters to be fluctuating between min and max, not unavailable.

See for example the graph on top of Page 48. When FAC 1 was reset, the TLU1 parameter (green) fluctuated between 0 and 1. Similarly when FAC 2 was reset, the TLU2 parameter (blue) fluctuated between 0 and 1.

Similarly the Wind Shear Detection parameters 1 & 2 (purple and black) fluctuated during these periods. They did not become "unavailable".

@chefrp
In addition what Volume mentioned, compounding the issue was AirAsia's use of two different, non-integrated fault reporting systems (the MR1 technical log and the PFR post flight report).

Specifically, intermittent faults from the PFR which were "resolved" by simple resets were not being logged into the MR1. The PFR was mainly used as an aid to troubleshooting.

However, AirAsia's maintenance management system (AMOS) was only based on the MR1. So procedures in place to identify and fix recurring issues did not flag this problem.

Also, AirAsia pilots did not report many of the FAC failures. Maintenance work arising from pilot reports would have been logged in the MR1, and could have resulted in the issue to be flagged as recurring.

But some AirAsia pilots apparently had simply reset the problematic FACs in flight and not subsequently report the issue to maintenance.

Lots of holes in the Swiss cheese.

Another Hole
 

I don't know if this might have saved the airplane, but, at least twice now, the pilot in the left seat could have saved the aircraft if the stick priority system were to have been re-thought/re-programmed, i.e., why not make it much easier for the left-seat pilot to take priority, based on their (presumed) added experience? Forty seconds of holding down the button seems to be asking way too much of someone who is fighting for their life (and everyone else's).

Yeah, apparently the Captain did not realize that either (A) his forward-stick inputs were not enough to overcome PF's wrong inputs, and/or (B) did not realize that PF was pulling at all! One cannot make anything completely fool-proof, but this seems like an unnecessary gaping hole in the cheese.

I propose, without really having any idea of what the hell I'm talking about, making the left-seat stick ALWAYS have priority.

Discuss.

Volume:The classic, decades-old approach to troubleshooting such problems is to swap components between systems (L vs R engine, PCU#1 vs PCU#2, etc.) or between aircraft (Nxxx vs Nyyy). If the complaint follows the component, then the fault is isolated. If not, keep looking.

This method often works when others fails, e.g. on intermittent problems. Would Air Asia Indonesia have this depth of understanding?

@pekay

Points taken but in every report I have read I recall reading that when referring to sounds picked up by the CAM they generally refer to them as "sound similar to"....

Nevertheless, do you not think that there could have been a whole lot more of the CVR transcript published. There are an awful lot of gaps and it beggars belief that the crew were not communicating more than was quoted. Sure if they were praying, which some over there might understand and relate to, I can understand it being deleted, but other than a few 'Oh My Gods" and "pull it" or "push it" surely they MUST have had some kind of discussion and interaction related to the numerous repeated ECAM messages at the very least.

The crew reaction to these warnings IS relevant to the inquiry, but instead of using information that the crew verbalized between themselves it seems to prefer to concentrate on what the FDR recorded as their probable actions. Without the "intent" which would have been gleaned from the CVR data we are left to believe that they may not have left their seat to re-set a CB that should not have been re-set.

I'm sure you get my drift.

When there are so many similar failures (tens !!!) from a sub-system or sytem just restting it is eagerness to be at a time in critical situations! Statistically one crew or another will make a disastrous maneuver at a time, whether in crew is below standard, standard or above standard.


Again we find that fatal law change to alternate like AF447. PIO followed very abruptly.


So sad.

Cpn Bloggs
My question on yoke v sidestick was referring to the fact that inappropriate stall recovery is common to all.
Regarding actual handling time of the PF I quite agree with you.

Never assume...playing Devil's advocate for a moment: I'd hazard a guess that when performing heavy/augmented crew operations on the likes of the 330 and bigger it's possible at times to find the lower hours pilot in the LHS.....

I return to the language question.
If the Captain thought he had told the P2 to push down, then he would therefore not say 'I have control' because he thought he was pushing in the same direction as the P2. Likewise the P2 would also think, thanks to his interpreting the "pull down" command as "pull" (which he was instructed to do later), that the Captain was either not touching the controls or trying to move the stick in the same direction as himself.

The differing misinterpretations of the Captain's "Pull down" command explains the subsequent actions of both.

The differing misinterpretations of the Captain's "Pull down" command explains the subsequent actions of both.
*************
Under stress foreign language skills can go out the window and English was the mother tongue of neither

The sequence of events that led to the crash of this aircraft was initiated by a failure of the RTL system.

In the aftermath of the loss of Flight 587, NTSB Safety Board member Debbie Hersman made the following statement.

"The sensitivity of the system is critical and needs to be addressed."

The NTSB decreed that modifications were needed to nearly 500 Airbus A300-600 and A310 aircraft in order to provide pilots with greater protection from hazardous rudder pedal inputs at high airspeeds that can lead to aircraft pilot coupling (APC)

At the end of the daylong hearing, NTSB Vice Chairman Mark Rosenker said, "We've seen today a tragic coupling of the pilot and aircraft that brought down Flight 587."

Pending system modifications, the board said that pilots need to be advised in no uncertain terms that back-and-forth pedal inputs to the rudder, or rudder reversals - even within the certified maneuvering speed - pose the risk of imparting aerodynamic loads high enough to break the fin, as was the case in the Flight 587 .

Is it not a sad irony that a system designed to protect against excessive rudder inputs by the crew was the initiating cause of another accident.

Frankly, this crew would have been maxed out by practically anything.

Stall trainning.
 

I believe stall training must be done in an aircraft in which a full stall series does not present a danger (Skyhawk, Warrior, etc.)
A pilot must be able to feel his butt rise out of the seat and against shoulder harness and then learn to fight the 'reptile brain' response to "pull back" to arrest the decent.
I suspect we all remember our first few stalls in light aircraft and maybe not wanting to let the yoke move forward.
This needs to be done through out one's flying career or you become fearful of the aircraft.

I strongly disagree with this assessment. My reasons for disagreeing so strongly are, I have flown these aircraft A320/A330/A340 as well as all other manufacturer types and while these aircraft are emminently dispatchable, sometimes the MEL is invoked and we must deal with ECAM messages as-dispatched.

The A320 & Airbus types and Airbus philosophy have been around since the late 80's and is, (or should be) well-known and well-understood. Crews simply cannot carry out their own, made-up procedures on any types of transports any more. I'm sure the B777 and certainly the B787 will be the same.

The FCOM, the QRH (Abnormal Ops) and the MEL are the guiding documents by which the aircraft is operated. Deviating from them is done under the captain's authority and responsibility if the safety of the flight is known to be at greater risk than not deviating from these documents. Otherwise it is expected that crews will adhere to the books. They should be in possession of sufficient understanding fo their aircraft to know why such cockpit discipline is paramount, even in trying and annoying circumstances as these must have been.

Resetting of both FACs in the air is prohibited by the Computer-reset section of the QRH Abnormal procedures. Here are the relevant excerpts:

https://photos.smugmug.com/photos/i-...-zwRTtH5-L.jpg -------------->https://photos.smugmug.com/photos/i-...-tRkwBMM-M.jpg

I have had several students who were scared stiff of stalls after poor initial instruction, and it was my task to fix that at a professional airline training school.
Given the reluctance to teach stalling properly, and the ubiquity of low-time instructors hours-building towards their ATPLs, it is very much a case of the blind leading the blind. It is,however, extremely difficult now to earn a decent living as an experienced flight instructor, so I don't see the situation changing anytime soon.

[QUOTE=FDMII;9198758]I strongly disagree with this assessment. My reasons for disagreeing so strongly are, I have flown these aircraft A320/A330/A340 as well as all other manufacturer types and while these aircraft are emminently dispatchable, sometimes the MEL is invoked and we must deal with ECAM messages as-dispatched.

The A320 & Airbus types and Airbus philosophy have been around since the late 80's and is, (or should be) well-known and well-understood. Crews simply cannot carry out their own, made-up procedures on any types of transports any more. I'm sure the B777 and certainly the B787 will be the same.

The FCOM, the QRH (Abnormal Ops) and the MEL are the guiding documents by which the aircraft is operated. Deviating from them is done under the captain's authority and responsibility if the safety of the flight is known to be at greater risk than not deviating from these documents. Otherwise it is expected that crews will adhere to the books. They should be in possession of sufficient understanding fo their aircraft to know why such cockpit discipline is paramount, even in trying and annoying circumstances as these must have been.

Resetting of both FACs in the air is prohibited by the Computer-reset section of the QRH Abnormal procedures. Here are the relevant excerpts:

What would have been the outcome if the RTL had not failed. I would suggest it would have been a no event. The causa sine qua non.

The RTL failed and the crew were duly informed the protection it offered was no longer available. The following best describes that which became operative thereafter.

"It is found that anything that can go wrong at sea generally does go wrong sooner or later, so it is not to be wondered that owners prefer the safe to the scientific .... Sufficient stress can hardly be laid on the advantages of simplicity. The human factor cannot be safely neglected in planning machinery. If attention is to be obtained, the engine must be such that the engineer will be disposed to attend to it." Alfred Holt 1877.

My question is why make a system to protect against human misuse which then when not available renders the whole unsafe in the hands of a human operator. Which then brings me to pose the questions, what would the outcome have been if this was a pilotless robot aircraft and how would the computer controlled robot would have resolved the issue of flying without an RTL. Perhaps the robot would not have had any need for an RTL system. In which case the RTL system was designed solely for protection against the human operator and it must therefore follow that the weakness is the human operator. Not the sort of conclusion that any one of us would wish to come to I would have thought. But there again it is always easy to blame the pilot, especially if he is dead.

When I first moved to Denmark (from the UK), I was confused by the signs on doors.
One read TRYK and the other TRAEK.

How confusing!

Then I realised how confusing PUSH and PULL might be to someone not brought up on English.

Chronus, you're making this out to be far more complicated than it is. We deal with these kinds of things as a matter of routine. The "whole" is most certainly not rendered unsafe when there are degradations in the flight control system. Seen it, done it hundreds of times in the sim...no big thing. But you have to know your airplane, and you have to respect the books, period.

The RTL problem did not cause the accident. No crew action was required, even though the warnings were a distraction and a nuisance.

An attribution of original cause is not "blaming" a dead crew - they needlessly paid the ultimate price - the airplane would have continued safe flight just fine until the 2nd FAC CB was pulled, with the captain out of his seat, which is contraindicated by the AOM as described.

A decision to execute a non-standard procedure plus the inability to control the result caused the accident.

At present, a "robotic" aircraft has far more serious shortcomings than one with competent crews but pilotless flight is another thread entirely.

No accident
 

The starting point is that that the aircraft actually was an A320 with side sticks and fly by wire. It is too late to permanently ground all side stick equipped fly by wire airliners and in my opinion this is not needed.

I clearly see the need for better training, including the requirement to follow FM procedures. The WARNING about resetting more than one computer at a time was either ignored or not known. This was a major part of the accident, but even so, it was still not inevitable.

After that, in spite of the language difficulty, the situation was still manageable if the captain had pressed and held his takeover pushbutton.

What can we do to prevent similar situations? The position can be like vaccinations giving humans "herd immunity". If enough pilots consistently report malfunctions in the aircraft technical log, it becomes more difficult for airlines to ignore. It isn't enough to see a previous report and assume that the company knows about the problem.

This particular FAC problem escaped correct reporting as a repetitive defect, so insufficient attention was taken. It was always a collective pilot responsibility but apparently pilots failed to act. A starting point would have been to write up every single instance of this malfunction. After a few instances of successful BITE tests, each subsequent write up should include reference to page and item numbers of all known previous reports re this malfunction.

Even in a significant airline, this might not get the full attention of management. Obviously they need more incentive to act. Finally, it is still up to the captain. And finally the captain must decide if the instances of this malfunction are more frequent on his flight and return to base for rectification. Now we have the attention of the company.

Do it once and maybe captain still has a job. Company will become more proactive and maybe same flight deck crew will be rostered for next day flight with same aircraft, maybe with 2 new or replacement FACs. Crew should critically quiz engineers about rectification work before deciding to fly. Assume they claimed 2 replacement FACs fitted and ask for critical comments after the flight. Assume crew accept the aircraft.

If the fault remains, return to base. Company might be very upset at captain decision. Crew might even be grounded pending investigation. The good news is that the aircraft will probably be also grounded pending investigation. In a week or two, when the problem is fixed, the same crew will be rostered for the first flight.

No accident.

Would a few additional (and specific) aural warnings be helpful?
 

Instead of repeatedly blaring 'STALL, STALL, STALL' until the plane impacts the ocean -- would it make sense to briefly interrupt that alarm to convey other critical information, so long as the stall is at a high enough altitude that there is time to convey vital additional information?

For instance, might 'ANGLE OF ATTACK 48 degrees' have been far more helpful to the 8501 crew than yet another round of 'STALL, STALL, STALL' ? (With the stall warning then resuming, so the crew isn't misled into thinking the stall has been remedied).

(If 48 degrees is too late to recover, then substitute an appropriate number to trigger the first alert.) The computers may distrust such high AoA numbers, but the warning hopefully will have alerted the pilots to a potentially excessive AoA, which they may be able to quickly verify and ensure they are making appropriate inputs.

Also -- With 8501, the Cptn may have been aware of dual inputs -- CVR excerpts have him giving instructions to the FO-- yet neither may have realized their inputs were in conflict.

Perhaps a "CONFLICTING INPUT - PITCH" aural alert (between the rounds of 'STALL, STALL') would be helpful, when the input deviations are substantial (and possibly if the computer also recognizes that the plane's current pitch/AOA is very abnormal).

Such a warning should immediately alert the flight crew to conflicting inputs, and also that one pilot may need to assert control. (They already "know" that, but bitter experience shows in very stressful situations some basics can get overlooked.)

I understand the importance of not overloading pilots with too much input. However, am I wrong to think that interposing a few specific aural warnings might be more useful than just repeatedly blaring "STALL, STALL"?

It would require some software and perhaps even hardware modifications (and careful testing to ensure new problems are not introduced). I don't know if there would be legal issues, such as certifications.

But something must be done. More training is vital, but only one part of addressing this problem (especially if under extreme stress some pilots seem to be overlooking basics). Unfortunately, the reality we face is a rapid global expansion of aviation coupled with a strong emphasis on beancounting and relatively little experience handflying.

Given this, would it make sense to add a few very specific alerts as a last layer of defense, at the critical moment when a fatal accident may be imminent?

The NTSB, etc., have been trying to get AOA indicators available to airline pilots for DECADES. But so far the regulators and industry remain opposed to the idea.

It could also be reasonably hypothesized that the captain had something along the line of a gushing head wound or even a broken finger. This would help explain some otherwise inexplicable actions on the part of both PM & PF.

It wasn't just inappropriate stall recovery. It appears that neither PF recognized they were in a stall even with the stall warnings blaring. That screams for change and not just training or SOP.

@G-CPTN
 

I remember reading an article on AW&ST about USAF IPs training Portuguese crews on low level, "high"-g maneuvers on C130s.

During high workload/stress flight regimes, the IPs noticed that the trainees weren't responding to "Push" commands, despite being fluent in English. Eventually, they realized that "Push" sounds almost the same as "Puxe," which is Portuguese for "Pull." So, the IPs switched to "Unload" instead and solved the problem.

Experience
 

Total flying hours: 12000
Hours on type: 8000
Hours operating in Alternate Law: ....Houston, erm I mean Toulouse, we have a problem.

No, they are not opposed to the availability of AoA indicators, otherwise boeing couldn't offer it as a customer option on the 737 (and probably other types). But apparently they do not see the need for them or they would be a requirement.

AoA information of all three AoA vanes is actually available on the bus, however only via the CDU and in numeric values, not something one would do in a dynamic situation like that. And i guess displaying the difference between where the aircraft is going versus where it is pointing, although readily available via a simple "bird on", is beyond the reduced mental capability of the pilots during an event like that.

As customer option blah blah blah. Is passenger safety optional?

NTSB recommendation to the FAA, 1996, after the AA965 crash near Cali, Colombia:
Nearly 20 years later we still don't have AOA indicators on the vast, vast majority of transport-category aircraft. Why is that?

FAA's response to NTSB on this recommendation, dated October 2000:

:ugh:

Simply, they were doomed there.
Less than a minute after AP disconnect, at 38000 ft.

Nobody could save that plane using the primary controls only (stick rudder and thrust) :}
(hint: abnormal attitude law)

But that doesn't explain all the actions of the captain. If he thought the FO was doing the right thing, the captain wouldn't have touched his own sidestick. He would have let the FO fly the plane.

The captain's problem was in trying to take control without telling the FO to hand over control.

The three words "I have control" might well have saved this aircraft.

I am sorry to put it so bluntly but this crash would NOT have happened had the Captain not tried a 'smart fix' of the ECAM by pulling the CB's. Let's not forget that this EXACT procedure is expressly and clearly forbidden in the aircraft flight manuals.

This was the origin of the event, the poor manipulative skills of the crew were only exposed by this fool hardy action. The ECAM wasn't even serious! Could have written it up and left the reset to the Engineer, yes it had been faulted on numerous occasions but as the Captain if you have a problem with this then don't accept the aircraft!

This is up there with the crew who crashed the aircraft into the Everglades whilst trying to replace a landing gear light bulb, a minor tech problem that has been poorly managed which has led to total hull loss.

The question of the crews inability to recover the aircraft is an industry wide problem, in our attempts to remove the human element from incidents we have created a generation of pilots who are now causing crashes due to piss poor manipulative skills e.g. Air France, Air Asia, Asiana .........

What is it about the Abnormal Attitude Law that "dooms" the airplane at FL380, and prevents recovery if the correct response to the stall is achieved?

Cross cockpit comms
 

One thing that puzzles me is why the CAPT did not take control assertively. He is 100% responsible for a/c. Did he feel FO was more situationally aware? Perhaps another similarity with AF447 is CAPT being out of his seat at the time of zoom climb by FO .

Contrast this Air Asia incident / AF447 incident with the very public Hudson incident (quote from flight 1549 NTSB report):

Note CAPT disables FO SS inputs after verbally taking control. This is pro-active action taking control immediately and without waiting for any dual input warning messages.

I'm quite sure you or any bus driver here know the answer. Ok I'm gonna say it loudly, if you wish.
See Airbus 319-320-321 FCTM or in this final report at page 78:
Pitch alternate with load factor protection (without auto-trim)
With THS at maximum nose-up position, the correct recovery action on primary controls only, is not sufficient: