And how is this useful practice and instruction to be consolidated?

Very well said, Chris.

Even better-said!!

And I think that there are primarily 2 reasons for what we see today.

First – airlines are in the business of making money – and, as almost anyone can understand, training does not make money … it only costs – and those costs are both direct (instructor salaries, salaries of students, equipment and facilities costs, etc.) and indirect (loss of line flying – which is the only source of income, transportation costs of moving the “students” into and back home from the training site, costs of per diem, lodging, etc.).

Second – airlines are loathe to provide more training than is required by the regulations. Why? Certainly cost is a significant driver, but I’m of the opinion that conducting training beyond what is required by rule could be seen by some as management recognition, or at least suspicion, of the existence of a sub-par capability, attitude, or ability on the part of existing crewmembers. So, simply by having regulatory required standards, lessens the willingness of some to continue to provide training beyond that required by rules. Some airlines in the US have adopted training under what has been described by the regulators as a voluntary program which has a unique characteristic … and that characteristic is taking advantage of building a training program with “alternative” training goals, “alternative” standards, and uses “alternative” training equipment, which can be extended to “alternatively approved” intervals. Personally, I think this situation is one that has resulted in airlines being able to save significant amounts of money, prompting the appropriate industry training managers to go to great lengths to compliment the regulatory authority for such “forward thinking” … which, of course, does not hurt the professional aspirations of those few regulatory officials ultimately responsible for such authorizations. The problem is, as I see it – and I’m not alone - is that having multiple sets of “alternative” standards applicable to “alternative” programs, each using “alternative” equipment, to complete “alternative” task assignments, on “alternative” recurrent schedules results in a “drifting away” from standardized expectations of performance and capabilities of these crewmembers, particularly pilots … all based on the economic value being contributed to the airline bottom line.

I am not about to apologize for believing it appropriate that airlines pay their employees a just wage for the skills and abilities they must possess and demonstrate; that airlines should provide their employees frequent and meaningful training, both initially and on an appropriate recurring basis – using equipment and facilities that are appropriate for the tasks that must be accomplished; that airlines should provide equipment and facilities to provide transportation services to those who choose to use this particular method of transportation; that airlines should charge those who use those services a fee that will provide an adequate income to meet expenses AND meet whatever profit margin they deem appropriate. It simply cannot be expected that the employees, or the skills of those employees, should be expected to be sacrificed for competitive advantage over other similarly situated competitors. Of all of these components, the only ones that should be governed by rules and regulations are those of competency and capability – and they should be equally applied to all who choose to participate in this industry. To me, this means that the regulatory authority should require that all participants have, and be able to regularly demonstrate, that the required capabilities and competencies are, indeed, addressed and that each appropriate person has satisfactory knowledge and experience to regularly and correctly exhibit those competencies and capabilities whenever and where ever called upon to do so. Clearly, this cannot be done through the authorization of individually approved deviations from established standards – particularly when those specifics (and any differences) are maintained as “confidential.” The requirement should be – no MUST be – a single set of demonstrable proficiency standards, using demonstrably appropriate equipment, for all who are depended upon to execute the duties and responsibilities of appropriate crewmembers (both cockpit and cabin) involved in delivering these services to the traveling public. If the regulatory authority should NOT do these things … I wonder why it is that a regulatory authority is involved at all.

Chris
It is Entree Level Training/Jet Familiarisation/Multi Crew Cooperation. It can be consolidated during Refresher training and online but unless you know the symptoms and the correct procrdure of UAS you may not handle it competently and safely. Let us say a pilot is competent in hand flying but does not remember correct procedure of TCAS he can mess it up.

I'll try different approach this time
 

What limit?

What do they measure?

100 kt in which direction?

Really, why?

Really, why?

What inconsistency?

What occurrence are you discussing here?

Improved in what way?

How would you simplify it?

Why is it so?

Really, why not?

What occurrence are you referring to here?

What MFDs are crowded with data and tapes? How do you propose to reduce crowding?

Why every aeronautical power that be does not see that?

How many medium and heavy jets vent below 60 kias while airborne, ever?

TWA843

Cheap provocation, eh? Think again, gentlemen. All of these questions are answered many times during the course of 11 threads we had on subject yet same old opinions resurface again and again. Please put some effort in answering these, I would really like to know what you didn't understand last time and time before that and before that...


You were not supposed to return it to earth unscathed if everything went southward so not much point in having backup FCS modes. No ACES in passenger jets.

Well... might be idea but it doesn't work out this way. I'd estimate the display to be 2 - 2.5 times wider than PFD and display being monochromatic aggravate the complete scan. It excels at visual approaches where any change in flightpath is immediately recognized as conformal flight path vector jumps around and that's the only thing you need to know besides speed. It is excellent for CAT3a approaches where there is not much scan anyway - you just make perfect doughnut with aeroplane symbol and guidance cue, shorten the speed error tape as much as possible, keep energy caret aligned with wing symbol using throttles (all of the symbols are concentrated in the middle of the HUD so you just stare straight ahead) and rely on F/O scanning classic PFD and HGS annunciator panel to shout "GUIDANCE" if things go pear-shaped.

It's not kneejerkiness I don't agree with, it's idea that pull-up-no-matter-what came initially from attempt to return couple of hundred feet lost on altimeters when speed correction went missing.

Exactly! He did not understand what happened despite uttering "We have no good display of speeds" so had no idea the best thing to do under the circumstances was...to do nothing. Pull was so consistent that it just couldn't be unintentional. Problem was within cognition.

No. He didn't know anything. He wanted to pull-pull-pull for reasons he never verbalized. If he was afraid of overspeed and tried to trade speed for altitude, why going to TOGA? As soon as the AP dropped out, all the thinking on the FD stopped for good.

Done.

Yes but this is all very well when you either know what happened or maintain presence of mind not to do harm when you are not sure what exactly you should be doing.

You suggest that A330 is sensitive beast that has to be treated gently. I say that it is exaggeration and you can throw her around hamfistedly at cruise altitude, in alternate law for 4 minutes with normal acceleration varying between +1.96 and -0.26G with end result being no worse than just utter mess in cabin and galleys.

How are we going to resolve this?

Why are we not discussing some other LOCs like Birgenair? 20 000 hrs could not prepare the skipper for simple recognition that rejecting the takeoff when airspeed is not working is a must or just to crosscheck the three ASIs. Or close call at EHAM where the guy who taught others how to perform V1 cut just couldn't make it properly when it happened for real? That's what AF447 is about: the crew that just couldn't perform when it needed to. It's very ugly picture and I understand the need to try to overpaint it with unrealistic theories about FBW or instrumentation.

Depends on what local CAA will allow, operator demand and training organization provide. In my case it was all spiced up by flying in direct law. It was piece of cake but then it was sim, not a real aeroplane over Atlantic at 4AM.

OK I'll give up quoting for a bit but when making sweeping statements about general flying capabilities of modern pilots, bear in mind that more than 30 other crews passed through same ordeal without a scratch and at least four of them were Air France.

Do that in roll direct and I suspect you will learn gentleness, and there might even be an additional mess in the cockpit.:}

Alternate1 and Alternate2 laws might work similarly in pitch, but they bear no resemblance in roll.

In response to Cland, the three points where I was called out .....

a) Seems to be confusion here, even amongst the 'bus drivers about the various "alpha" limits or protections or whatever. And when one is the actual FCS limit. So if the FCS doesn't have an absolute AoA limit, then at least show the pilot the stall AoA.

My system had AoA limits depending upon A2A loadout, heavy iron on the wings or in landing configuration. We could only select the A2A or A2G limits, not the gear down limits.

b) Looking at my HUD video you will see what most considered the minimum displayed data for a safe and precise approach. The ILS cross bars are easy to understand. AoA bracket shows where you are and when aligned with the flight path marker( FPM), where you should be. Heading, altitude ( AGL or baro) and speed.

If you want some kinda "magenta line" to helpsteering , fine, but that can be a symbol that you keep aligned with the FPM. Had that on my A-7D, but no autopilot connection to anything. Viper was pure manual, as seen in the video.

c) We had one important backup mode - Standby Gains, which is extremely relevant to our discussions here the last three years. If air data was deemed unreliable the FCS used a fixed value for "gains". One with gear up, and the other gear down. Sure, if way off of the Q that the system used, it got touchy or sloppy. but very easy to fly, much like the "old days" when we didn't have all that stuff.

We did not worry about a myriad of alpha stuff, as we had one with gear up and the other gear down. Both limited the jet to stall AoA ( roughly). You could not command above those limits as the AF447 folks did. Could still run outta energy in a steep climb and momentum would allow the jet to exceed the limts because the flight controls could not react quickly enough ( very hard for our jet, and even more so for the heavies, IMHO).

Having an ACES II seat was very comfortable, and several of my friends are true believers. One, and main reason I never wanted to fly the heavies after getting out was the responsibility of many SLF behind me. If I screwed up, or the jet screwed up, so it was just me. Ain't gonna talk about that anymore.

.....and I assume Cland and maybe one or two others have bothered to look at my HUD tape of the emergency landing. Otherwise, much of my views could be misunderstood.

Chris Scott - I have just found your post #800 buried in the grass at the bottom of a page (missed it previously). Your point about 'trigonometry' is irrelevant when we were talking AoA signals and is only relevant in terms of IAS which only enters the fray due to some software code. Therefore it would not matter if the jet was flying backwards or upside down - an airflow of that magnitude would, I suggest, generate a 'useable' AoA signal. Whether the designers choose to ignore that signal is another matter. That was one of the errors of design.

Clandestino said - not sure what 'vent' means here, but it does appear that the designers forgot this very important 'trigonometry' and that a jet in a fully developed stall can and did, in fact ?'vent'? below 60kts (IAS).

@BOAC

Firstly let me say that I am NOT arguing that the A330 design does not need any change - on the contrary I think the stall warning logic should be altered. However the discussion relating to validity or otherwise of the AoA signal does, I think, ignore some relevant facts:

I doubt very much whether usable AoA signals would be generated if the aircraft were flying backwards or upside down, but I imagine you wrote that tongue in cheek ...
More to the point, if an equipment manufacturer declares in his formal paperwork that his equipment will work to specification only inside a given range of ambient parameters then the aircraft designer has no option but to conform to those limits. It is not a case of design error but one of a fully auditable certification trail.
The problem comes of course when the system sees input conditions which say that the operating limits of the equipment have been transgressed, but the measurement of those ambient parameters is itself false. To that extent Chris Scott's comment on trigonometry is relevant - the input data, in this case airspeed, was rendered false by the angle at which the airflow entered the pitots. Was it then a design error to fail to consider this case?

I doubt very much that the designers "forgot" that an AoA in excess of 40 deg would give stupid speed signals, but they did of course fail to believe that a trained pilot would allow the aircraft to get into that state. Be fair, would you have believed it before publication of the AF447 final report? In which regard I would refer you an earlier posting, from PJ2 I think, that designers work on an assumption that the aircraft will be flown to a given set of rules, allowing for some sensible variations thereto, but not to any extreme one cares to dream up.

So what of their treatment of the stall warning which is, when you go back to basics, the reason for all this discussion?

With hindsight the logic is faulty, but I don't see any need to alter the logic governing validity of AoA signals which might well bring on other problems we know nothing of. Instead I would propose a change of stall warning logic so that if triggered by a valid AoA signal it should stay latched on until it received another valid AoA signal that stall conditions were a thing of the past. That would have avoided the confusion found in the AF447 cockpit for a minimal change to the aircraft systems. It would also, incidentally, bring the aircraft into line with the latest EASA requirements on the subject.

Indeed "tongue in cheek".

I don't think I have ever argued that AB was partly 'responsible' for the accident due to the Stall warning 'logic', but just that it was significantly contributory. I agree people would not 'expect' the crew to take an airliner to these realms. However, why WoW was 'abandoned' I do not know - I thought it was a well-proven input? Surely the point is, Owain, that as far as we know there was a 'useable' AoA input available, probably throughout the episode, and indicating a stalled wing. The fact that 'trigonometry' rendered the USE of this signal impossible due to the IAS constraint is indeed relevant, but I cannot see why posters are going on about 'non-swivelling pitot tubes' etc etc. Surely we all understand why the IAS reading was invalid?

I agree with your last paragraph and if that is also the EASA thinking, then good, although as with all ever-increasing complexity of systems there may well be a hidden pitfall in that approach.

BOAC,

The post of mine that you refer to was merely an attempt to explain to you and rudderrudderrat that you were misunderstanding Clandestino's cryptic remark about trigonometry, and taking seriously his reference to "no swivelling pitots". It was not a statement of my own opinion, merely an attempt to help you understand his point in his absence (for which, predictably, he has offered me not a word of thanks).

It's well known that Clandestino's style of response sometimes invites irritable misunderstanding by the passing objects of his contempt, myself included. Did my polite, but non-didactic, explanation go over your head? Should I have tried the long version?

(Ah, I see OG has done a far better job!)

....and I hope you understood my t-i-c remark about there always being a reading from an AoA probe (given enough wind to 'swivel' - that's for real) regardless of 'trig'?

Affirm, BOAC :ok:

BOAC

I don't think I implied that you were arguing about responsibility - if I gave that impression then my apologies. Nobody is going to contradict a statement about the S/W anomalies being a contributory cause though.

So far as WOW is concerned I don't think I have ever seen anything from Airbus on the subject - mostly the argument about it being 'abandoned' seems to emanate from PPRuNers speculating - but I stand to be corrected. What I would say is that leaving aside the AF447 affair, the existing logic functioned without complaint for several decades; why take the head off if it is working?

Now if your objective is to supply an AoA signal pure and simple and which is not going to be used as an input into some more flight critical system then I would agree there may be no need to complicate life by making that signal contingent on some other input - it being tacitly assumed that the aircraft will not be flying at a real 60 kts indicated or otherwise. Implicitly here I am taking the view that since airline pilots have been flying without AoA information since the year dot such information cannot really be classed as flight critical no matter how useful it might be in certain circumstances. [pace gums!]

I entirely agree your point about non-swivelling pitots etc.

To be clear, my remark about EASA requirements simply reflects the latest stall warning requirement CS25 207 (c) which, as I have said before, differs from the requirement applicable at the time the A330 requirements were frozen.

Good thots, OG.

To be honest, I don't know when the Viper FCS ignored AoA or didn't. Best I have in documents is that it was used as long as we had WoW switch decompressed. When I did a tail slide one day all the warning lights came on ( AoA sensors at limits, calibrated speed below 120 knots) and without resetting them, I had complete control once the nose fell thru. Reset the FCS panel and fight was still on, heh heh.

OTOH, the A-7D and other jets displayed the AoA whether the plane had a FBW system or not. Viper didn't, as some folks the braket was a nuisance symbol in the HUD.

I learned in planes that had very clear warnings you were approaching a stall. In later planes, we had buffet and wing-rocking and such to let us know we were reaching the edge of the envelope. It was a different feel than when nearing critical mach, which usually exhibited stuff like aileron reversal, nose tuck and a pronounced "buzz" in the airframe as the shock waves formed.

So my thot is that a display should be available and right in front of the pilot whether the FCS or Otto is using it.

I'm wondering if that speculation stems from the 1993 LH A320 accident at Warsaw, where the WoW timing was not sufficient to allow deployment of spoilers:

Lufthansa Flight 2904 - Wikipedia, the free encyclopedia

In response to that, Airbus reduced the required pressure on the WoW sensors, but they never abandoned it as a systems input.

OG, I'm almost certain you're correct in that the <60kts AoA NCD flag has nothing to do with determining air/ground mode - it's purely an attempt to enforce the AoA vane manufacturer's specified limitation of minimum 60kts airspeed for valid data, and prevent junk data from being used in FCS calculations.

Just a reminder - the Final Report para. 1.6.11 says it thus:

...and should read and hopefully re-thought, having seen vanes moving on walk-rounds in a stiff breeze! I'm sure the intention was well-founded, but hindsight tells us the application was not, and WoW would have been far better.

With all due respect, there's a difference between "moving" and "providing valid data".

See above - I'd be surprised if WoW was not an input *in addition* to the <60kts IAS check, as opposed to "instead of". This is purely a way of preventing duff data from getting into the flight control systems calculations - entirely separate from air/ground mode determination.

Dozy
That of course is a totally different application. For the record (I don't want to introduce another thread drift) it wasn't that WOW timing was adrift.Rather it was that the aircraft touched down so fast that even when derotated there was so much residual lift with full flaps that there was not enough weight on the wheels to overcome the breakout friction and allow the WOW microswitches to 'make' AI did introduce a software modification as a result but I cannot remember the details.

@OG - check the Wiki link.

Followed up with:

I remember the timing issue because the left gear touched down 9s after the right - it was the delay in the left gear getting the required compression value that caused the delay in spoiler deployment.

[EDIT : I'm aware that it's a totally different application, I was simply trying to distinguish the air/ground mode criteria, which are not relevant to AF447, from the AoA vane maker's specified NCD threshold of 60kts, which is. ]

Would anyone like to know how much ADR/CAS independent AOA the digital portion of the display seen in Turbine Ds HUD U-tube link will indicate?

Probably not. :}

I, for one, would!

In fact I'd be exceedingly interested to see how the interoperations between flight control systems (both traditional and FBW), warning/alert systems and AoA sensing equipment pan out on every type in passenger service when exposed to values as extreme as AF447.

I don't think it's a shortage of people wanting to know - I think it's more a shortage of willing volunteer pilots* to fly the profile!

* - Because most pilots don't have a death wish...

@Dozy

Nearly but not quite:

Checking back on my source data another AI mod was to apply a small amount of dumper deflection immediately at touchdown, which killed enough lift to trigger braking but which was not enough to hamper a go around.

Thanks for the quote, OG. Yes, one mainwheel has to suffice. Landing with residual bank (e.g., deliberately on the upwind main gear in a strong crosswind) provides routine ground-spoiler deployment, encouraging the downwind wheels to touch down in short order, even if the pilot leaves a lot of into-wind aileron applied (in the transition to ground law). The sudden loss of roll-spoiler effect is partly responsible, of course. With the nosewheels down, IIRC, nosewheel steering is not available above the 72 kt ground-speed, whether rudder-fine or tiller.

(Excuse the off-topic memoir.)

OK465,

I would like to know. I believe the green band is indicative of 3º at nominalized conditions on a B-737. The tick line near the top represents stick shaker AoA, approach to stall.

TD & Dozy,

it's sort of a trick question in that....

ALL AOA data, both HUD (BAE) & PFDs is independent of the 2 ADR modules. (However there are other HUDs that have that annoying ADR/AOA dependence for AOA display.) Same with FPVs. This aircraft PFDs always display it and have FPV capability regardless of type HUD.

The green band is the non weight based AOA for the weight based approach speed with appropriate flaps out. The tick is as TD says the onset of the shaker. The PFD also has a PLI (pitch limit indicator) that corresponds to the 'tick' AOA.

The maximum digital readout one can see in an FFS is around 30.5 degrees, this with BOTH ADR modules failed, no air data. I think it's limited by the fact the aircraft just won't sustain more than that, unlike the 330. However that could just be the point it pegs for any AOA above 30.5, until AOA is reduced, a system somewhat I believe like RetF4 envisions. I wouldn't trust the sim much beyond this point however....yet.

IMO, contrary to some opinion, it is emminently easy to utilize as a performance instrument, not a control instrument, the data is already in the box and it just takes some OPC pin activation and CDS enabling to display it.

And you don't have to ever look at it if it's otherwise objectionable to you. We know who you are.:}

Some good points, Okie.

My aero understanding and actual college courses and actual flying with AoA indicators tells me that the "stall" or max L/D is pretty much the same as long as you don't have all the flaps and leading edge devices and such taken into account. Weight is not a factor, as the wing doesn't care according to the configuration. It's the lift curve for the configuration, not the weight.

IMHO, there's no reason to display the "bracket" as we had in the A-7D displayed all the time in the HUD. Wanna put it someplace on the crowded MFD, then fine. So in the Viper, we only got the AoA bracket with gear down for approach. And since we didn't have a lot of extra aero devices than flaps, the bracket gave a super reference for approach AoA that was fully compensated for weight, as we should realize that AoA for stall or such doesn't care about weight or gee.

Not to say that the AoA indication is a panacea, as it has to take into account of aero configuration. The display can be raw or it can be biased depending upon configuration. But it sure as hell tells you where you are and where you should be. Simple cross check with actual speed that you or your right seater calculated/looked up will help. Bit me in the butt one night on a slick runway when my trailing edge flaps were not fully deployed, but leading edge flaps were fine. So optimum AoA was higher speed than I should have been flying and I almost ran off the other end of the runway. BFD, lesson learned, as I had placed my trust in the "optimum" AoA for the approach.

The thing I keep harping on is that neat HUD that shows actual flight path vector based upon inertial data ( could be actual or a blend of pitch and some gyros). If I pulled and the FPM didn't move "up", then I was pulling too hard or was close to a stall. The pitch lines were also very comforting, especially for a night takeoff or for a carrier shot, because if that FPM was not above the "horizon line", then you were descending regardless of your basic attitude indicator display. I was taught, and flew many hours in bad weather and had to use the basic attitude and power specified/recommended. Then cross check for altitude and vertical velocity. Anyone here that didn't fly that way?

In almost 2,000 hours of HUD flying, I only had the FPM invalid once ( translate to 10,000 hours for you heavies). I knew it instantly and used the steam gauges to continue the takeoff. That was in Viper, as the A-7D used TAS and gyro attitude to provide a poor man's FPM.

I also experienced a static port freeze one day upon descending for a TACAN approach. Airspeed went to hell as I changed altitude, and altitude was frozen. BFD. I had that neat FPM and those pitch lines to tell me I was doing fine. Comforting. Set power per normal and come on down. I would have done the same thing without the HUD and inertial FPM, just as most here would have done. Basic airmanship and flying skills, ya think?

Nope, it didn't work.
 

What relevance it has to AF447? For Finnegan's sake it didn't crash in spiral dive or rapid roll to ocean but in upright stall! That's what happens when you try to pull the aeroplane where she just can't go perfomancewise.

Herein lies the rub; on 330 it is very much mach and configuration dependent. No airspeed - no mach - no meaningful alpha limit.

Who gives a rat's stern about my wishes? I have commented on the way I performed visual (no steering cue) and CAT3a (cue required) approaches using head-up guidance system I.A.W. what my employer expected of me.

Thanks, I stand corrected: Viper had degradation modes, different in details but principally similar to Airbus FCS laws.

Neither need Airbus pilots as long as they don't bust the lift limit.

...on aeroplane supported by cropped delta of relatively low aspect ratio i.e. with far larger useful AoA span (per design and intended use) than moderately swept, high aspect ratio, supercritical, cruise optimized wing of 330.

It's hard to climb heavy steeply but if combat aircraft controls can not react quickly enough, what chances have the civil transport ones?

As someone who was using HUD in airliner for a while I agree it is quite useful but I don't think it would have saved the AF447.

Typo for "went".

20/20 hindsight. No one has proven before that it can.

Spot on. Instead of introducing WoW signal into stall warning logic and creating another possible point of failure, not completely dissimilar to one affecting Spanair 5022, this one could be done by simple software code change. However...

...if there was confusion about everything else, at least actions on the stick show that CM2 firmly believed that stall warning can be resolved by pulling.

It was never there. If you already must have speed and mach input to calculate critical angle of attack, it makes sense to simplify it by just putting in a couple of lines of code that cut alpha readouts at speeds that are completely unsustainable.

I am just mocking the inability of some to understand pitot tubes reliably measure only total pressure resulting from airstream approximately parallel to aeroplane's longitudinal axis. Such a design has been with us since they have first been attached to aeroplane and everyone except a couple of PPRUNers finds it fine.

Those who think IAS reading below 60 kt was really invalid - not.

I don't recall anything about this. What I remember is that post-Warsaw spoiler logic on 320 was changed inasmuch selecting reverse cracked open the spoilers thereby killing a bit of lift and compressing both struts. The rest remained unchanged and full deployment was still only available when both MLG WOWs detected landing. Way the Airbus spoilers designed, WOW signal is a must to prevent airborne activation of ground spoilers.

Interesting... I'd like to hear more about about junk data when Airbus flies slower than 60 kt, I mean why is this problem not solved yet by introducing the alpha probe that can work reliably below 60 kt?

Yes.

WoW is probably not part of "stall warning logic" outside of air vs. ground mode operation. The argument seems to be based on the misapprehension that the <60kts NCD flag was implemented instead of WoW to determine AoA vane validity, which is a fallacy. It is far more likely that both inputs are used.

I was going on the Wiki page, which stated:

Me thinks that Cland and Gums are on the same page for most of the philosophy here.

I can also appreciate the enginnering inputs from Doze, but saw many well-intentioned "engineers" suggest inputs to our FBW system and other stuff that we operators threw out. They were mainly "protections" and such, including a FCS "pull up" if Hal thot we were gonna auger if we exceeded pitch and altitude and such trying to hit our target. (99.9% of we pilots threw the notion out. There comes a time when "pressing" just a little bit makes the difference, and ya take your chances.) So we opted for a big flashing "X" in the HUD when HAL thot we were pressing too hard. Ignore it at your peril, and I have HUD film of a friend that ignored and paid the the price.

The mentality ( attn Doze) that the system acted like it was designed has some problems if the aircrews don't fully understand the "system". And as I have oft-repeated, the 'bus reversions modes are complicated and should be more straightforward. The basic FCS system should be basic enough in the reversion modes to HELP the pilot and not confuse the pilot about what protection is still here and what isn't. And as Cland and I seem to agree, there's a huge role for basic airmanship.

@ Cland:

- AoA is, indeed, a factor at mach approaching "critical" mach and such. The airfoils nowadays are much smoother than the old ones I flew. Only flew three that had significant warnings you were gonna get into trouble. And I admit that I flew airfoils that could exceed the mach ( M=1.0) without losing control of the jet.

- Aspect ratio and the airfoil shape of four of the jets I flew definitely made thos jets more tolerant of doing strange things when flying at "the limit: as the AF447 were doing when things went south.

- Running outta "E" is still a player, and the Russian crash might show that when the crew did a go-around to an extreme pitch attitude. But that's another thread.

It took us some very extreme pitch attitude/power combinations to run outta "E", and then exceed the aero surfaces' capabilities to keep us under control. I just can't see such with the heavies without serious airmanship problems.

I appreciate the oppo to add to this discussion, and have learned a lot from the heavy pilots here.

The "system" makes the A330 behave like any other aircraft. If you keep pulling up and let the speed decay, you will stall. F/O Bonin was a sailplane pilot of considerable skill - at some fundamental level, he understood this. If some folk want to make an argument that he was pulling up in the hope that Alpha Protection would keep him safe, they're welcome to it. But that argument chafes against the fact that as a sailplane pilot, he had to understand that if the aircraft was stalling and the increased thrust didn't help, then he had to get that lift from somewhere.

But he never acknowledged that the aircraft was stalled, despite the warning blaring in his ear for more than a minute. The horror afflicting the piloting fraternity is that he had all these indications and yet continued to pull up - a response for which there is no logical explanation. This accident is as inexplicable as it is horrific - something inside him kept telling him to pull up in the face of all the data he was presented with. I suspect that psychologists will be tasked with trying to provide an explanation until long after I'm gone, but it's not a case of being confused by the systems, it's a case of a pilot - for reasons that we'll never understand - pulling up consistently in the face of a situation where the precise opposite action was required.

Only problem I have with your "analysis", Doze is that the jet flies like all the rest.

That is true if all you want to do is have auto-trim corrected for pitch attitude, roll limiting to 30 deg or so, and so forth. Mission requirements, I guess, as you have pointed out to me and others about my personal mission/operational requirements. All I can say was the thing flew like everything I ever flew until at the edge of the envelope, and then it excelled.

I shall never understand the continued back stick, so I guess I have to wait until we meet in that hootch bar in the sky one day where all we pilots wind up. Still makes me cry.

I also understand that many heavy pilots have never flown their jets close to the performance envelope WRT mach, AoA and so forth. So not interpreting mach/overspeed indications versus incipient stall is not a surprise to me.

I also flew sail planes and enjoyed the feeling of silent flight but it,as Sully and I feel, it doesn't help you much flying airliners. He landed in the river because he knew how to fly and saved everybody and was a hero.

I would hope all of us have the piloting ability to do what he did. Button pushers would not have a chance in his condition.

Kind of sad, isn't it? This is the new generation of pilots sadly.

The new pilots should force themselves to make them as good as the old guys and not rely on automation if it all fails. We could do it, why can't you?

Originally Posted by Clandestino in response to Chris Scott
Originally Posted by Machinbird
This the relevance.
Bonin applied excessive control inputs that the aircraft would have tolerated had it been in Alt1 law, but since it was in Alt2b, a roll direct law, the sensitivity of the roll channel was far higher than he expected and he set up a roll oscillation that lasted 30 seconds. During this same 30 seconds, he inexplicably pulled up into a climb that used up virtually all the aircraft's energy reserve above stall speed. In my opinion, this roll oscillation fits the general definition of a PIO.

What most readers here do not seem to appreciate is that a PIO event will grab 100% of your attention while it is in progress. My only experience with PIO was in an early visual simulator and it was brief and without consequence. Nevertheless, it made a deep impression. The best analogy to it was like unexpectedly tripping and beginning to fall and looking around desperately for a handhold.

In effect, while piloting an aircraft, the aircraft is your body. Would you be able to execute a checklist or maintain a scan while your body is falling out of control? This type of 'emergency' will cause a tremendous flow of adrenalin in those without prior experience. Will you be able to pilot rationally immediately after such an event? This is my take on the cause of AF447's loss.

'WoW' is involved in the FWC stall warning logic but here it is called flight phase [IN FLIGHT = flight phases 5,6 & 7).

!But that is not the issue!

All 3 ADIRUs did send an AoA SSM NCD (No Computed Data) signal, for the FWC this means there was no valid AoA to compare against a threshold which depends on Mach, S/F and SB config.
(Max AoA threshold 16.3º in S/F Config FULL (ALTERNATE & DIRECT ) and a fixed 23º in NORMAL)

Actually this philosophy is identical to Boeing T7 only the values are different.

Now if you want the system to be changed the solution has to be found in the ADIRUs NCD state within ARINC specification.
(in AF447 the measured air speed was sometimes 0kt - and even negative q -, just setting a lower value or 0kt would not have solved the issue here.)

For accuracy of the vane:
The AoA vane accuracy is shopchecked in a laminar air flow windtunnel at 100kt.
At an offset of 3º(+/- 0.2º) from original value the vane should return to original position within 0.25º with a max. overshoot of 0.5º to pass this accuracy check.

@ Clandestino, Dozy

That agrees with my understanding of information obtained from the horse's mouth.
The 'wiki' comment probably arises because the A320 oleo design was changed to reduce the breakout 'stiction', but that was done to improve the harsh taxy ride that occurred on some surfaces. As a byproduct it reduced the vertical force necessary to start the torque links moving from their fully extended position (used to signal WOW) but that was not its primary purpose.

A classic case of 2+2=4.5!

BEA Recommendations Follow-Up AF447
 

The forgotten report:

Suite Données Aux Recommandations De Sécurité

Hardly any of the BEA's recommandations have been implemented.
The report makes a scandalous read with the DGAC and EASA still passing the buck.

Why not a redundancy-lost message?
 

While considering system updates, how about the generation of a message something like:
pitot redundancy lost, review UAS procedures
before the autopilot dropped out.

I don't know how long before autopilot drop-out it could/would have been generated, but it would surely have reduced the startle factor.

My guess would be 10s of seconds from the rejection of the unfrozen pitot to the divergence in readings from the other two pitots leading to the declaration of UAS.

Nope, it's pretty much instantaneous in human terms. The software is checking and cross-checking several times a second - you'd be talking about maybe 2 or 3 seconds from initial icing-over to the systems detecting an extended discrepancy and disconnecting AP.

In the case of AF447 the point is somewhat moot, as the PNF is supposed to monitor ECAM and action things accordingly. The PF started pulling up almost immediately the AP disconnected without first assessing the situation. The cause of AP disconnect was right there in the ECAM - but it seems that the PF's instincts overruled his reason and training, and neither the PNF or the Captain could break through his panic.