Loss of Control In-Flight - Flight Crew training

[n5296s]

It's significant that Air France brought in a group of experienced, well qualified foreign pilots to sort out their typical French arrogant establishment.

Curious to know whether this comes from anything public?

A year or so after the accident, I happened to fly back from France sitting next to one of these "well qualified foreign pilots" We had an interesting chat. His view was that the accident had seriously shaken up AF and that they had done a great deal to make sure nothing like it ever happened again. Would be interesting to know more if there is anything public.

[hans brinker]

Originally Posted by Meester proach

They’ve been banging the Upset recovery gong ever since AF. Every bloody sim now, boom, some crazy attitude.

this whole magenta line children thing is fairly bogus really. BA have taken and trained low hour cadets since the 60s and have they had any crashes attributable to it ?no.

And as for whoever is on about hand flying up to RVSM....yeah great idea....NOT. For a start my pet cat could follow the FD in what will be many staright lines - and don’t suggest raw data - it would be a bad call to try that in the busy TMAs we operate in, not to mention putting your often tired workmates workload through the roof.
You have to balance flying practice with operational safety/ common sense.

I do think every sim should have loss of airspeed and raw data ILS though .

I seriously doubt doing an ILS raw data twice year is going to do wonders for my hand flying skills. I have been doing this for 2 decades, and it is easy to get used to the automation. Thankfully (for me at least) my company gives us a lot of leeway, and lets us do anything from raw data to auto-land, (conditions/equipment permitting). After having the FD and/or AT deferred a few times, I realized I wasn't entirely comfortable going back to basics and have since made it a habit of (again, circumstances permitting) departing/arriving without the use of AP/FD/AT. It certainly has benefited my hand-flying skills, and my confidence in handling unexpected situations. I also have seen quite a few FOs, after a few legs, doing the same, and expressing gratitude at being able to practice skills they felt were eroding after flying fully automated for a few years. If you honestly feel you couldn't hand fly while in a busy TMA, I respectfully wonder if you should do it more, so you would be comfortable...

[Bob Viking]

Meester Proach

Parts of your post scare me to be honest. Have you ever heard the term ‘train hard, fight easy’?

Not practicing something because it is difficult is 180 out from my approach. It is PRECISELY the reason to practice it. As for your crew being tired, can you ever guarantee a real emergency will only ever happen when everyone is completely rested?

As for a TMA being busy, what difference does that make? Traffic or no traffic, the workload is the same.

I realise many of you could point out that my kind of flying is very different and I would agree with you to some extent. Our aircraft are not the same. However, before you rush to ignore my thoughts may I point out that I have flown my little fast jet into and out of San Fran International, Mcarran (night), Chicago Midtown (night) and several other busy international airports in North America, Europe and the Middle East.

That was in a jet with no autopilot, no FMS, nowhere to clip my approach plates (balancing them on my knee or flying one handed are the only options) and being single pilot. Remember, this is also without the benefit of two hours worth of extra fuel.

May I please ask another, unrelated, question of the assembled masses?

We have had discussion of many stall events in civil aviation (obviously AF447 being the main one).

Have any any of these LoC or stall events been computer driven? Or are they a problem of human creation?

I am still not not trying to be contentious despite how my questions may appear.

BV

[blind pew]

Weather radar

In my case I spent 6 years watching the captain use echo (Marconi?) Radar before an aircraft change allowed me to play with the next set up. It wasn't easy and whilst the books described the criteria and cut offs there is also the varying amount of water droplets not only depending on altitude but the updraft velocities.
My next shared set was on the DC9 with one of the world's best operators and training systems of the time. Led by fast jet pilots of various nationalities in an environment that mixed soft centered returns with cumulus granite..
Followed by the death cruiser, another set and the ITCZ which always had a full crew in the cockpit and occasionally the second captain. I thought I had it sussed until one night heading back to Europe from west Africa, after I had diverted 300nm off track to avoid I realised that we were the only one on vhf2 that was diverting and I had misread the returns.
The radar on the fokker 100 was easier.
As an anecdote a BA colleague during training on the VC10 started avoiding a rocket cloud visually..the skipper took over and flew through it..result heavy turbulence check..broken back and a fleet of ambulances waiting on the tarmac at Colombo. It wasn't showing on radar because radar doesn't (or maybe didnt)show velocities nor shear layers.
Having spent 50 years flying and the last 25 gliding and paragliding around the world I am still learning the vagaries of weather..mostly without personal damage.

[pineteam]

Originally Posted by Meester proach

They’ve been banging the Upset recovery gong ever since AF. Every bloody sim now, boom, some crazy attitude.

this whole magenta line children thing is fairly bogus really. BA have taken and trained low hour cadets since the 60s and have they had any crashes attributable to it ?no.

And as for whoever is on about hand flying up to RVSM....yeah great idea....NOT. For a start my pet cat could follow the FD in what will be many staright lines - and don’t suggest raw data - it would be a bad call to try that in the busy TMAs we operate in, not to mention putting your often tired workmates workload through the roof.
You have to balance flying practice with operational safety/ common sense.

I do think every sim should have loss of airspeed and raw data ILS though .

I’m not saying you should fly with AP off until RVSM airspace, I’m just saying it’s legal to do so.
BA might not have a fatal accident yet due to poor flying skills may be cause we are lucky to fly very reliable aircraft. But would the crew be ready to cope with a serious failure in direct law? Flying the FD bar is easy but raw data is another skill you can not maintain if you only practice in the sim. I can tell easily who are the guys who fly raw data in line or not when I go to the sim: Simple the one who never practice in line are tense and inaccurate for most of them. How can someone expect to fly accurately by only praticsing twice a year in a box is beyond me. Oh and by the way, raw data in the sim is easier than the real plane. Just saying. xD

[grizzled]

TheiC

I tried to reply to your PM re the other forum on pprune but the system says you've chosen not to receive PMs. Just FYI
Cheers mate

[fdr]

Originally Posted by StudentInDebt

this amuses me, or saddens me. During a sim involving stall recovery training a few years ago I was asked what signs would indicate approaching the stall by the TC (Checkie). I reeled off the above list of signs and was immediately corrected as the correct answer is apparently - an audible “STALL STALL” warning and then treated to a white board presentation of the speed tape cues to correct my lack of understanding of the question. Finger, pulse etc

Your observations are sadly consistent with observations of multiple training systems and operators. A simple hard fact about the real world vs simulator training is that events in the real world are dynamic and frequently confusing in the development of cues to the crew. The quoted instructor's point on what counts as primary cues belies the fact that in most CVR tapes, there is most often profound confusion and cacophony arising from multiple cues, warnings, alerts and dynamic inputs. The training focusing on exact displayed data is lost in the noise, whereas what does invariably exist is the big picture factors; attitude, air noise, response to controls, vibration etc. Basing the determination of your flight path and energy state to a "STALL, STALL" alert or a flashing red light is akin to planning to avoid obstacles in your car by the cue of the airbag deployment, rather than the cues from looking out the window.

The sad fact is that when a loss of SA occurs, recovery of the split between expectation and reality arises from the outside inwards, from the woods not the trees, from the overview to the detail. Awareness of flight dynamics and the response of the aircraft to flight controls is the absolute first part of flight training for very good reason; detail inputs are beneficial to accuracy but not to global comprehension of flight state.

When we train stalls in the simulator, we train to BS canned, staged conditions, that (and I apologise to all of the TRI/TCE's in advance of the following comment but....) minimises the training value fo the exercise. a stall or loss of control condition in the real world occurs from a loss of SA, and that is necessary to be the case to gain greatest training value. The logical outcome from my heretical position is that recovery training needs to concentrate on big brush items, not detail, "lipstick on the pig" stuff. Every aircraft (other than helicopters) will fly given half a chance. AF447 would have recovered had the pilots taken their hands off the controls, and placed the thrust levers to idle. Same applies to a Pitts, 747, and even most odd ball B/A aircraft like the F4, if the recovery is commenced early enough, at the point where the aircraft stops responding in the expected manner to the pilots control input.

The FAA improved their position on stall recognition and recovery, but the manner in which we do the training today remains less than optimal. Flight crew should not be afraid of stalls, they are embarrassing when inadvertent, but they are just another part of the envelope of the aircraft, and our training acts to increase anxiety in what is just another part of driving the plane. Reliance on warnings and instruments that have shown themselves to be prone to error is a lousy basis to keep the blue side on top reliably.

The French legal system has spoken, and it appears to be a limited scope of causation that has been determined to be causal in 447. The crews competence is in part the consequence of organisational and regulatory training and qualification standards and their efficacy. The decision on the removal and replacement of a known compromised sensor system was also the result of factors beyond the crew who were confronted with the consequence with an adverse outcome.

"Tout commence en mystique et finit en politique." Péguy

[172_driver]

The FAA improved their position on stall recognition and recovery, but the manner in which we do the training today remains less than optimal. Flight crew should not be afraid of stalls, they are embarrassing when inadvertent, but they are just another part of the envelope of the aircraft, and our training acts to increase anxiety in what is just another part of driving the plane. Reliance on warnings and instruments that have shown themselves to be prone to error is a lousy basis to keep the blue side on top reliably.

Nothing to add. Superbly written, that's all

[TriStar_drvr]

This is from my recent training. Although the Sim obviously cannot replicate the G loads of these maneuvers, it does record what loads do occur during recovery. The Sim has been modified to replicate the buffeting as the aircraft approaches a full stall. It will rattle your teeth. Overall a very enlightening training session.

[cwatters]

Seems to me this is an important passage in the SN...

Training should focus on a flight crew’s ability to prevent an undesirable aircraft state by identifying and recovering from any out-of-trim state, particularly during high workload and dynamic situations in all phases of flight. This should include pilot awareness of aircraft trim condition, intervention strategies and techniques when the automated system is not performing correctly...

I agree that hand flying and training in old manual aircraft is beneficial but perhaps the folks that design and run training sessions in sims need to be more creative with the failure modes they can throw at pilots.

[neilki]

UPRT in a 320neo....

[fdr]

I would contend that 3*drivers image (post #89) is a similar setup as neilki's (post#91) insofar as they are upsets arising from an unusual attitude commencement. 3*'s in particular is not likely the consequence of a stall event in the sim, resulting in the attitude. It is correct that some time ago the 737 sim did have tweaking to the aerosim model, and that has resulted in a more realistic stall and post stall behaviour. Buffet in the stall of the B737 specifically is most pronounced at the low altitude stall cases, either clean or configured for landing; the high altitude stalls and accelerated stalls are quite gentle in comparison. High speed buffet is a fairly high frequency vibration that increases in intensity dependent on increasing speed and g load/aoa. High altitude stalls give a mild low frequency buffet that is not as unpleasant as the stall buffet at low to mid altitude and configured for landing. The 737 meets the rolloff requirements of FAR in almost all cases, unless the slats have been poorly rigged, then a rolloff at the break or just before the break can be quite interesting.

Years ago, while evaluating a 3rd party TCI and his students, the NG crew kept on planting the box in the dirt following an encounter into an aggressive microburst model. While they were recuperating and licking their wounds, I ran a quick QTG eval, and then ran the same scenario. The off axis entry into the microburst was causing a pitch up, which is correct, and a yaw towards the center of the microburst, which is also correct. The net result was the crew were adding a pitch up with the initial warning, which was taking the box into a stall while there was a lot of yaw rate, and the box was doing a 180 roll and being planted. If the crew respected the PLI (aoa limit) or kept the ball in the center, the box behaved impeccably. The TCI wanted to fault the box for it's behaviour, until he was debriefed as to the QTG eval and the importance of aoa and yaw management in a stall. Again, the sim even in the out of balance stall was providing reasonable modelling, and was still recoverable, but it needed to be flown to recover from what should be an expected outcome of a stall with a yaw rate on. This sim model was in stark contrast to another 3 engine model that could be fully stalled and would climb in stall at 6000FPM, whereas in the certification testing the plane had demonstrated normal stall break.

The difference between the actual aircraft to the simulator is the continued period of a load experienced at the seat, and the intensity of the buffet that can be encountered. In most cases the real aircraft will leave you in no doubt as to the fact you have stalled the aircraft at the lower levels.

The observations above come from a rather misspent career. Flight crew use SOPs and the FCOM/POH to avoid getting into these areas, which is highly appreciated by the SLF and their families. The planes however behave like a variant of a Cessna 150 out on the edges of the envelope. They certainly behave better than the PT-22 in a stall following Hap Arnolds design input into what had been a beautiful design.

[ironbutt57]

Originally Posted by macdo

Or you could just insist everyone spends a couple years tooling around in a knackered old turbo prob with no Ap! Sorry that bus has already departed, hasn't it!

people onboard that bus learned how to fly, unlike these MPL wonders

[Centaurus]

different

High altitude stalls give a mild low frequency buffet that is not as unpleasant as the stall buffet at low to mid altitude and configured for landing.

Now that is very interesting because one B737 Classic full flight Level D simulator I have observed, has entirely different stall warning characteristics to that you have described. This simulator is about 25 years old and has passed regular fidelity checks by the regulator. As the stall approaches at 37,000 ft and well before stick shaker actuation the buffet in this simulator is very marked indeed and is absolutely un-mistakeable.

On the other hand, when configured in the landing configuration and approaching the stall below 1000 feet on an ILS, there is no pre-stall buffet at all - only the stick shaker shortly before the stall.
The FCTM states a stall warning should be readily identifiable by the pilot, either by initial buffet indication or an artificial indication (stick shaker).

In our operation this landing configuration approach to the stall is designed to replicate as realistically as possible the Turkish Airlines crash on final to Amsterdam where the captain's radio altimeter gave erroneous readings which in turn caused both autothrottles to close to idle without pilot input. In that accident, the crew failed to take corrective action until too late to be effective..

.

the difference between the actual aircraft to the simulator is the continued period of a load experienced at the seat, and the intensity of the buffet that can be encountered. In most cases the real aircraft will leave you in no doubt as to the fact you have stalled the aircraft at the lower levels.

Your point is well taken. But how does one explain that in the case of at least one simulator (737) there is no initial pre-stall buffet warning in the landing configuration (hence the stick shaker requirement). Would buffet be apparent in a more up to date simulator? How should the simulator instructor explain this apparent anomaly to the candidate he is training for a type rating? Is it all about fidelity of "old" simulators getting worse with age?

During annual or scheduled fidelity checks it is apparent that although clean approaches to stall at 15,000 feet may be a fidelity test regulatory requirement in some States, this writer has yet to see a fidelity test on the approach to stall in the landing configuration at sea level. After all it would be extremely unlikely to have a airline fidelity test pilot current on various stall configurations in the actual aircraft. It simply would never happen for all the obvious reasons.

It does make one wonder the how reliable are some simulator stall recovery fidelity checks that in theory require the testing officer to be knowledgeable and current yet he may never have stalled the real aircraft.

[fdr]

Originally Posted by Centaurus

The FCTM states a stall warning should be readily identifiable by the pilot, either by initial buffet indication or an artificial indication (stick shaker).

O wise sleeve valve one, the stall of the 737 is quite compliant with 25.207. The stall warning buffet occurs prior to the break in all cases. The stall warning may occur before or after buffet depending on the setup of the aircraft and the configuration that it is in, however occurs before the break. The aircraft meets 25.203 stall characteristics, however if the slats are not properly rigged, it can give a considerable roll off which is manageable. Accelerated stall at FL370, the buffet occurs before the stall warning. Taking the event to a full break takes some altitude, but is not uncomfortable. level unaccelerated stall at altitude are quite benign. That is for the aircraft, -300,-400 to -500 series. The Classic is not much fun above MMO, the amount of vibration from the ailerons is considerable, enough to give visible vibration of the wingtip from the flight deck, from below MMO dependent on the aircraft condition to above MMO. Up to 0.85 the classic still behaved OK, other than those examples that had vibration from the ailerons below MMO. Above 0.85 there is a noticeable reduction in CL/aoa. The NG fares better at high speed in general, particular without the winglets, the aileron design is better and not as sensitive to shock interaction as the classic is.

One loss took the type to an extreme dive speed, (way above Vdive) at which point there was evidence in the flight data that aileron reversal was occurring. The situation that confronted that crew (and unfortunate pax), was self imposed and was already well beyond recovery at that point. Shortly after that the aircraft started to shed important bits, much beyond what would be reasonably expected. (important safety tip: don't pull breakers airborne or try to determine the limits of the system knowledge that the crew have).

The SLUF is not my personal favourite Boeing by a long shot, but it has been successful for the CFO's, CEO's and shareholders. The later variants are overrepresented in runway excursions, approach speeds are higher than would be desirable IMHO, and technology to improve the lift generation of the flaps exists, but has not been implemented.

[Centaurus]

FDR,
Fascinating and thank you for your erudite explanation of what happens in the real aircraft rather than a simulator.

[Centaurus]

fdr. Check your PM's 15 Sept

[fdr]

Originally Posted by Centaurus

FDR,
Fascinating and thank you for your erudite explanation of what happens in the real aircraft rather than a simulator.

You are too kind, sir. "Verbose" could be a simile for "Erudite", for which I apologise, this is my industry, for better or worse, and discourse is at least provocative to opinions.

The take home from my observations is that the 737 is, in the end, just an aircraft. It handles adequately, I have never been fond of its ailerons, but that is a personal conceit, CEO's/CFO's love them. The stall behaviour is benign, and to a point, operations beyond VMO/MMO are reasonable. Older aircraft, there are some I have not been prepared to take to the normal limit without rework. The NG is a competent cockpit, and the plane does a good job. I don't see that the MAX will be much different, once we have returned to normal, and learn what needs to be learnt. The manufacturer has not done itself any favours in the MCAS saga, they should have opened up from day one, and engaged with stakeholders rather than attempting to limit blowback to their legal standing. When understood and trained, remembering the lessons forgotten from the early years of these aircraft, the manual trim issue can be mitigated.

The events around the MAX have been unfortunate. The industry was quite prepared to blame the crews, the OEM did, the FAA did, elected officials of the US congress did, and many voices on this forum were strident in their accusations on the crews competence. That is disappointing, but in keeping with what occurs routinely in this "enlightened" industry, one that spends so much time with corporate policy statements, and pledges to honesty and integrity, of valuing the employers, and assurance of "Just Culture". IMHO, "Just Culture" is to be found on the label of yoghurt containers, but remains lacking in our industry, from the regulators through the OEM's to the operators. We are diminished accordingly.

It's miller time.

[misd-agin]

Simulators in the past weren't certified to provide accurate post stall AOA performance. So saying anything about a simulator in the past and thinking that is how the airplane would respond doesn't work. Articles in AW&ST have stated that Boeing and Airbus have agreed to a generic narrow body stall simulator performance model. It is not necessarily an accurate depiction of the stall characteristics of any aircraft.

[Mark in CA]

I've seen this theme brought up in many threads on this board. Now the NY Times Magazine attaches it to the Lion Air and Ethiopia 737 Max crashes. I posted this in another thread:

NY Times Magazine: What Really Brought Down the Boeing 737 Max?