AF 447 article
 

Unlike 3407, we have a lot more evidence as to the state of mind of the pilots involved. The only silver lining I saw was that once one of the FOs announced he had been holding the stick back the whole time, both other pilots stopped chanting climb and instead chanted descend. At least the other two pilots were able to put together what was happening once they had that last piece of information, but sadly there was not enough time.

Unfreakinbeleivable........it is truly mind boggiling how a three man crew can fail to understand the gravity of the situation. I have seen how people can actually tune out or not hear the sounds such as "too low gear" or "stall, stall". Its frightening.

To me.....IMHO.....another reason to defend Boeing!

Its sad that so many had to lose their lives.

Both Boeing and Douglas have had similar accidents with the crew holding the aircraft into a stall from high altitude to ground impact.

I think there is a fundamental flaw in how we teach crews to recover from a stall especially in big jet airplanes. The whole pull back and power out ATP style stall recovery is killing people.

Maybe......but a crew must know that the aircraft is stalled in order to recover from it. When the airplane is flying you......it's too late!

Agreed, whole-heartedly.

I was shocked by the airline/FAA/civil stall recovery procedures coming from Air Force (and Navy) training. The airline procedure was to get close to---but not actually stall. Altitude control was the top priority, and supposedly, the biggest reason for busted checkrides.

The problem with this type of training is it leaves one with no training for an actual full-blown stall.

I've read articles that suggest the FAA is about to change this to what the military has done for decades. I hope so. It's about AOA control. Once AOA is under control, you can control altitude.

No offense intended but you have some serious misunderstandings about how the Airbus flt controls and warning systems operate.

The airplane was screaming STALL STALL STALL and "cricket" was chirping, which is basically the master warning, all the way to the ground. So I have no idea how they didn't know the airplane was in a stall.

Secondly the airplane was in flight control alternate law which means that the stick was giving direct commands to the control surfaces. It did exactly what the pilot asked it to do. IF the airplane would have been in flight control normal law it would have gone to max thrust and pitched the nose down automatically and overridden the pilots inputs.

Now the whole thing about one guy not knowing what the other guy is doing because both sticks don't move in conjunction with each other I 100% agree that is the most poorly designed and dangerous system I've ever had the misfortune of flying.

However in the case of the 727 and the DC-8 that both stalled and were held into a stall those guys also had full cockpit and aerodynamic indications of a stall and still held the nose up all the way to the ground. Which IMO indicates a serious training issue.

Take a look at these and tell me what happened in both cases.

This one is eerily familiar to the Air France crash.

http://aviation-safety.net/database/...1201-1&lang=en

Here is another also in a Jurassic jet.

http://aviation-safety.net/database/...?id=19961222-0

http://aviation-safety.net/investiga...cvr_abx827.php

I've been flying this plane for 7 months now ... Last sim ride did this same exact scenario , very difficult to recover but we did it and saved the aircraft at 10000 feet from 37000. Airbus says it is mandatory to do this on every sim ride from now on. Great read btw ... Very tragic.

My ignorance ... why did you have to descend so low? Aren't there known pitch and power settings for the bus?

Yep correct, implemented in the training is the "Unreliable Airspeed" scenario. There is a chart in the QRH for the exact numbers for T/O / Cruise / Approach & Landing.

At cruise the golden numbers are 2.5 Degrees of pitch with approximately 79-80% power.

The reason we descended so low was because the TRI put us in the same stall the Air France crew got into. By the time we were able to cut power put the pitch of the nose down, break away from the buffet and recover we lost that amount of altitude.

The article was spot on regarding the sidestick feature and the other pilot not knowing what the other pilot is doing. Coming from a boeing jet it was one of the hardest things to get use to. The artificial feel is a problem during the first few hours on the plane.

Here in San Diego, only few years ago, there was a terrible car accident where an brand new Lexus powered out of control on a freeway. A family of 4 died when the freeway ended at a T intersection and the vehicle collided with terrain.

It appears that the driver, who was a seasoned California Highway Patrol officer, only took one evasive action, he applied full break pressure as the engine was racing.

Because that had such a terrible outcome, no one really questioned the driver's actions. Why didn't he shift the car out of gear, why didn't he shut off the motor… etc.

Don't climb on me if I don't have the facts right with this story, but my point is that sometimes during a panic situation, when there's souls on board, and you don't know what's happening, the body just reacts and common sense doesn't quite kick in.

No offense taken, but I do have some understanding of the Normal vs Alternate Law on the Airbus design. That's why I don't like it.
In regards to the cre not knowing that the airplane was stalled. Have you noticed that they didn't react to the "stall, stall" aural cue? There are times that a crew will get overloaded with other tasks and they mentally tune out the other noise. There have been crews who landed gear up, while the airplane screamed "too low gear".
My comment abou tknowing that they were stalled is not in regard to the plane masking it....."it's about the crew not "hearing" it.

Although.....IMHO....I'm not a fan of the Airbus logic.

No kidding. I was giving spin-training to a new-hire once and when they recovered "way too early" we found ourselves upside down. Simply roll level with full aileron deflection right? Not for this guy, he was "frozen" and I had to "save the day". Relatively simple and we briefed it ahead of time, but once in that "totally new" situation, he could not process the information and react correctly.

A good friend had an aerobat and wanted some aerobatic training, but despite my best efforts and demos, I couldn't get him to use the right control inputs/entry to prevent ending up in a dive during aileron rolls, which I recovered from.

Should these people not be pilots? I don't know. I don't believe that unless these people have the extensive aerobatic experience it's fairly unrealistic to expect them to react correctly in "new" situations (wake turbulence, inadvertent spin, etc). I mean I can have someone recover from a spin, that's all fine an dandy, but what if we didn't experience a REAL tight spin that didn't stop right away where the correct inputs did nothing at first (or caused it to tighten further)? I've had many more "frozen pilot" situations or frantic but inadequate control inputs observed from other pilots whil doing these. I've done these many times over and it's more like "slow motion" for me now, but unless they have, is it realistic to expect a positive result?

Maybe this is part of the issue, being put in a totally new situation with only a few seconds to react correctly. Easy to talk about, discuss ahead of time, but realize that when put into that situation, the pilot or whomever has to proceses the situation, recognize it for what it is or what was talked about, decide upon the course of action, and so on. Easy to say, but much harder to process due to it being a new experience. As much as we train, try to form the correct habits, and so on, unless we can experience all the situations prior, we are just going to have a question mark when presented with a totally new situation.

I know we all think we can, but can we REALLY handle something that's far outside of our norm?

I spent several years in the left seat of the A-320/319 and I never became a fan of it either. I got used to it and understood it just fine but plain and simple it is a poor design. I am not defending airbus. rather pointing out that this is not the only airframe to suffer these types of accidents.

The million dollar question isn't it? If I were king all pilots would have to have some extensive aerobatic training before they were allowed to fly any airplane for hire. And in that training they would have to have a very intense stall training series from every imaginable attitude, power setting, airspeed ETC. I'm talking High nose, low nose, inverted, vertical, spins, inverted spins, tail slides, flat spins, deep stalls,etc etc. Then they would have to go through an instrument emergency maneuver training course.

But we all know that stuff is never going to be required. Aerobatic flying increases a pilots perception of the norm and should be an integral part of every pilots training.

My company went to this well over a year ago. It's much more realistic and practical than the crap I was taught and graded on for the 10 years prior.

I have never liked the basic ideology of the Airbus. Most of my training in the 320/319 fleet consisted of pulling the stick to the aft stop in abnormal situations, (Stall, Windshear, CFIT). Works quite well in normal law. Not so much in alternate or direct law.

We are creatures of habit and are trained by repetition...

I just went through recurrent and we of course did the stall series, wind shear, and CFIT avoidance drills. Our instructor, sim partner, and I were discussing the advantages of an AOA gauge and its' usefulness in such situations during a break. My sim partner use to fly the Falcon 10/100 and we walked by the briefing room for it and stopped inside to have a look at the panel. Right there in the upper left hand corner was an AOA gauge!

USMCFLYR

All modern airliners have an AOA system(usually two), but the output is usually integrated into the FCC, the pilot gets no direct AOA readout.

AF 447 was a case of three pilots flying a perfectly(or at least fully functional) good aircraft into the water.

High altitude stalls in a heavy are an entirely different animal than the low level stalls used in type ratings, much lower available thrust and if the A/P disconnects at a relatively high speed you might be left with limited control authority that requires trimming. It seems that this is finally being addressed in training.

I'm just going off of the fantasy world that is sim land. And I don't know if it applies to all sims or not. But the ones my company has use upsdated data from real flights to best simulate flight characteristics of the aircraft for things below FL250 IIRC. Anything above that just goes by how the programmers put all the 1's and 0's in place, anyway. At least, from what an instructor told me, with that being said......

The high altitude stall is VERY interesting. It's quite the opposite from the low altitude where things can happen quickly for recovery, on a relative scale at least. It's can seem like an eternity as the aircraft is losing thousands of feet to recover airspeed at a slow rate while the aircraft has the slippery feel from the high altitude.

Note in the article that the captain got what was happening and told the PF to descend. The captain said "no" to the climb call out. But the PF pulled back anyway. Could it be that if the captain had visual cues of the yokes being pulled full aft, he would have been more aggressive about getting the PF to descend? We will never know, but it shows that the lack of coordination between the crew and the lack of visual reference for where the flight controls were, added to the situation. Not a Boeing vs Airbus thing, although I can't help but wonder if this was a Boeing, would the captain have known with certainty that the PF was trying to climb and this certainty not created a far more aggressive response by the captain.......

Sad situation :(

And that is what I liked......a direct readout. A gauge or a number on the HUD/EFIS/etc. Just doing the winds ear/CFIT recoveries in the might radar equipped and pressurized King Air would have been better if I had had a reference to pull too!

Not knowing if the PF was still pulling back on the stick would have been easy to tell with some direct feedback of Alpha as FastDEW points out.

USMCFLYR

I agree, but once they got the pitot system thawed out they had everything they needed right in front of them. Including a flight director telling them where to put the nose, based on both AOA and airspeed.

I think the temporary loss of airspeed got them into a mode of thinking that later caused them to disregard all of their instruments. A direct AOA readout would certainly help with a loss of airspeed indication, the 330 apparently had a problem/AD on the pitot heat system. A very rare failure, almost unheard of in aircraft with triple redundant pitot systems.

I have had two complete pitot/static failures, both in military jets, AOA and attitude instruments saved the day with very little ruffle.






July 2009, Airbus issued new advice to A330 and A340 operators to exchange Thales pitot tubes for tubes from Goodrich Sensors and Integrated Systems.[152][153][154]

On 12 August 2009, Airbus issued three Mandatory Service Bulletins, requiring that all A330 and A340 aircraft be fitted with two pitot tubes manufactured by Goodrich Sensors and one Thales model BA pitot (or alternatively three of the Goodrich pitots); Thales model AA pitot tubes were no longer to be used.[155] This requirement was incorporated into Airworthiness Directives issued by the European Aviation Safety Agency (EASA) on 31 August[155] and by the Federal Aviation Administration (FAA) on 3 September.[156] The replacement was to be completed by January 7, 2010. According to the FAA, in its Federal Register publication, use of the Thales model has resulted in "reports of airspeed indication discrepancies while flying at high altitudes in inclement weather conditions", that "could result in reduced control of the airplane." The FAA further stated that the Thales model probe "has not yet demonstrated the same level of robustness to withstand high-altitude ice crystals as Goodrich pitot probes P/N 0851HL,".

On 21 December 2010, Airbus issued a warning to roughly 100 operators of A330, A340-200 and A340-300 aircraft, regarding pitot tubes, advising pilots not to re-engage the autopilot following failure of the airspeed indicators
wiki

Amen!!! At my airline, I'm taught NOT to lose altitude during the staged approach to stall scenarios. Lose more than 100 ft.? I bust!

A bunch of people are having to die before this faulty training is corrected. It's truly sad.

The FAA has suggested a change to the unrealistic method of min altitude loss during stall recovery:

Pilots of transport-category airplanes should be taught to reduce the angle of attack as their first response to a stall warning, the FAA said on Tuesday. In a new proposed advisory circular, the FAA says it aims to provide "best practices and guidance for training, testing, and checking for pilots to ensure correct and consistent response to unexpected stall warnings, and/or stick pusher activations." The AC is an apparent response to the Colgan Air and Air France crashes, in which the pilots' reaction to stall warnings was part of the accident chain. Jet pilots have often been trained to use power as the initial means of recovery, to minimize altitude loss.


The AC also notes that training in stall recovery is most effective when using simulators that are as realistic as possible. If the simulator experience doesn't duplicate reality, instructors should explain the differences, the AC says. For example, certain simulators may not be able to accurately duplicate the motion cues associated with accelerated stalls. The AC also suggests that while it's a good idea to introduce some distractions while practicing the stall encounter, instructors are discouraged from introducing "multiple compounding malfunctions," to avoid confusion. The FAA is accepting comments on the proposed AC until Jan. 12, via Docket No. FAA-2011-1359 at Regulations.gov.

This is just so weird, never in my training, military or civilian, was I taught to minimize altitude loss. It was always about gaining speed to get the wing flying again. So this is a big surprise to me that the FAA was mandating people minimize altitude loss and thus pulling back which would/could prevent you from fixing the problem ... lack of airspeed.

aoa, aoa, aoa. Fly the wing! +1!

Edit....for some reason I can't get all caps here.....

Interesting read put out last year:

http://www.faa.gov/other_visit/aviat.../SAFO10012.pdf

It's nothing but a snapshot of how messed up the FAA is. With the F and the second A being the culprits.

ANYTIME you combine federal and administration, you're bound to get something messed up in the first place, and for it to take a looooonnnggggg time to fix.

It was in the ATP PTS, as well as many many airline training programs to "minimize altitude loss" in not only the recovery, but also the entry. As well as use it as a pass/fail criteria. Sadly, it simply devolved into more of a teaching the student how to enter the stall with minimal loss than it ever was about 1) recognizing the onset of the stall and 2) recovering aircraft control, speed, and altitude. You know, ACTUAL real world life saving actions. As well as how get out of the shaker, try to maintain an AOA (that you have no idea of), etc.

As well as maintaining altitude in the recovery and not going above the entry altitude when the shaker activated, and stabilizing at the speed at which the maneuver was started, etc.

Did you open the link I provided, and read it?

And the plot thickens ...

It looks like it wasn't an FAA problem, but more of an airline interpretation problem. My issue is, why wasn't this caught by the inspectors or check airman certifying these programs?

That's a WHOLE other "cowpie".... don't get me started. (biting tongue)
But I TOTALLY agree!

Of course not.......it is always easier to rant about the gov't in general.
As far as 'minimal altitude loss' - AOA is the answer once again. Gives you the best recovery performance with minimal altitude loss. Hummm.....
35 AOA brothers! (....if you aren't max performing, you might as well be flying an F-4 (the one is for you UAL!))

USMCFLYR

Also on the bust category is not setting up the stall properly - this ranges from Private to ATP PTS. Lemme get this straight... if I don't set up the stall properly- the one I'm never going to let the aircraft get into to begin with - you're going to give me the pink slip?

The training scenarios are flawed. In both AF and Colgan there existed altitude that could be lost - traded - for a flyable wing.

I added a G-IV type in August and the practice was the same.... We'll you lost 50 feet on that recovery....let's see if we can get that down a bit. Really? We're at 10,000 feet!

I was taught to fly by an old timer Army Air Corps pilot. His stall training was right out of primary flight training circa 1942. Steep fast and scary for a newbie pilot but he pounded it in and stalls from any attitude at any power setting and any airspeed were his main focus, he told story of buddies lost making "button hook" turns from base to final in T-6's and such and stressed that none of his "kids" were ever going to die from a stall at low altitude. Before I soloed I was able to do a three turn spin and recover to a point. I'd done half snap rolls to inverted, accelerated stalls from a 60 deg bank to a full break,so on and so forth.

Several years later I was told by a modern style instructor that my stall entries and recoveries were dangerous. The first time I ever witnessed a transport style entry and recovery I was thinking this is a great instrument scan exercise but it's got nothing to do with an actual aerodynamic stall series.

Later when I started flying aerobatics I never found myself in question as to how to get out of a stalled attitude it was simply reflexive.

We'd all be doing ourselves a great service if we went back to Army Air Corp style stall training circa 1941.

The PTS is a guide line of the minimum performance and proficiency required from a student to be proficient for the rating he or she is seeking to attain. There is nothing in the FAR's that says you can't give your students some extra deeper levels of training. I don't instruct much anymore but when I did all of my students got some "extra" non required training on things like stalls and engine failures, fires ETC. The stuff that will kill you.

AirHoss, you and STDeviation's post are so on the mark, it brings me to near tears. I hope that both of you are young enough that you pass this philosophy to many, many newbies. However, I fear that many will look upon you as doddering old fools and, therefore, should be merely tolerated. But then again, there was that old Air Corp instructor that YOU never forgot when you were a young pup?

Yes sir, sure did. I was simply relating my experience, as well as others. From 2000 till 2010, at two separate companies, the "training" was the same. Interestingly enough, the overall training styles at those two companies were different. The first from 2000-2005 was AQP and very big picture. The most recent is just finally staring to transition to AQP. Last year when I took a PC was the FIRST time that there was new stall scenarios given. They also went from a graded item to simply traning. That was in 2010, the same year that's dated on the document of the link you provided.

Good point. The oversight/regulatory agency that signs off on, approves, and observes these training programs has to share the accountability with the airlines themselves. When the POI or the feds sit in on training session, PC, type ride, whatever, they can't stick they're head in the sand and say they didn't know what was going on.

See above. Enough accountability to go around.

EXACTLY what I was getting at. The pure stupidity of having the maneuver graded on how you got into it in the first place. Almost everything you do on a standard PC you have the potential to see on a line flight. EXCEPT purposely entering a stall and doing a steep turn. Seriously, I'm going to go out a do a steep turn? That somehow is a measure of a pilot's ability to control an aircraft and be a safe pilot?

That's about all is. NOTHING more than a scan exercise. With maybe some slow flight practice thrown in there before the break.

I know that my training now is different than before, but having just come from a recent recurrent I was reminded about how mild some of the stall recoveries and unusual attitudes during this training. I like the fact that you mention aerobatic training Hoss because I feel that some basic level of aerobatic training ins highly advantageous. I just feel very comofrtable seeing something no more than 30 degs nose high and within 45 deg AOB (which I believe were the limits the sim instructor gave us for training the unusual attitudes)

I'm new to the civvie world of initial/recurrent training and this stall series/steep turns stuff as part of the checkride - but in my mind it was all part of what we called basic airwork before. Can you make the aircraft do want you want it to do? I'm certainly no fan of the way the stalls are portrayed/graded/handled and I look forward to some more realistic training/checking standards in the future - stuff that will really make a difference.

USMCFLYR

Understand your point. But flying a V1 cut, a single approach especially if it's a non-precision dive and drive style, single engine go around, wind shear recovery, etc is a pretty good display of making an aircraft do what a pilot wants it to do.

:D
Hey! I was impressed with the level of training that I did in both my initial and first recurrent (to include the 50 RVR taxi - great visuals (or lack thereof)
There was more going on at one time in the SimuFlite sim than I ever gave at one time in our previous sims. Enough so that I might not even called the items listed above 'basic' airowrk but maybe even more advanced (or at elast challenging) airwork.

USMCFLYR

AOA vs Airspeed
 

KC10:

I think in the big-jet world (mil and civilian), it is stressed more as a 'lack of airspeed' issue.

But one of the finest aspects of training in the T-38, I think, is the contrast between "high-speed stall" and "low-speed stability demo." (I'm guessing you did these, back in the day, unless you are a post single-track UPT guy).

In the High-speed stall, we go into a level turn at 350-400 kts....and pull until the jet says "I've had enough." It starts rocking and rolling, and is difficult---but not impossible---to control. To recover? Pull just a little less....NOT "Don't Pull At All."

The stab-demo is a 60-degrees nose-up climb until 175 kts. Then, you bunt (unload to 0.5 to 0.0 g) and let the nose fall. Usually see about 80-100 kts as the nose hits the horizon. This is WELL below 1-g level flight speed. However, the jet has no buffet or stall---you just can't ask the wing for more than about 0.5 g. Stay less than that, and you can still maneuver. (Just can't hold level flight).

Both maneuvers teach that stall isn't about airspeed...it's how how hard you are pulling on the stick/wheel. And NONE of the stalls are treated as precision-altitude-control maneuvers.

Plus-1 on the "everyone should learn to fly aerobatics" bandwagon!