View Full Version : Loss of Control In-Flight: Pilot Training Issues

9th Feb 2010, 02:53
I am currently serving as a pilot representative to an International Working Group charged with providing guidance to industry on how to reduce the rate of Loss of Control In-Flight accidents (LOC). Specifically, we are looking to identify the necessary components in pilot training activities that will provide a professional pilot the knowledge and skills to; (in order) avoid upset events, recognize upset events and recover from upset events.

I have been a user of PPRUNE for about two years, and have read many threads which touch on the issues concerning LOC to date. And while the committee I am working on will take a formal academic approach to reviewing the existing literature and research on the topic, I have always found it useful to solicit the "subjective" input of subject matter experts, of which there are many that contribute regularly and positively on this forum. I ask now for you help and input.

Our approach is to first produce a Training Needs Analysis (TNA) which will define what the specific knowledge and skill sets are necessary for a pilot to first AVOID, and then be able to recognize and (if required) recover from an upset. Upsets can be defined as exceeding 25 degrees nose-up, 10 degrees nose down, or exceeding 45 degrees of bank. In addition, an upset can also be defined as an in-appropriate airspeed or aircraft configuration which may require an alternate control strategy from a pilot (aileron hard-over, etc.). Also, of primary importance is a stall condition which is independent of airspeed.

The second chronological approach is to asses the existing knowledge and skill set of the "typical" professional airline pilot vis-a-vis upset recovery training (URT). This is a great challenge as this demographic varies greatly. This part of the effort is a "Gap Analysis" which will identify the deficit between the existing condition and that required to effectively meet the threat of LOC.

The third step is then to identify the variety of training tools available to close the deficit in ensuring pilots retain and maintain the necessary knowledge and skills to confront URT/LOC. These "training tools" include all available platforms (simulators, actual aircraft, etc.), media (academics, part-task trainers, etc.) and methodologies (enhanced URT instructors, MPL, AQP, LOFT-LOE, recurrent training, etc.) that can "prove" their efficacy.

Here is what we have "learned" to date.

1. Lack of wide-spread academic knowledge amongst professional pilots on the aerodynamics of upsets including specifically; high altitude aerodynamics, high-performance, swept wing aerodynamics, the approach to stall, and full-stall regime, use of and aerodynamics of primary and secondary flight controls.

2. Airlines "assume" pilots have this knowledge when "hired" and assume they "retain" this knowledge even though there is little emphasis on it in most airline recurrent training programs.

3. Research indicates that the majority of airline pilots do not posses the skill set necessary to recovery an actual aircraft from an actual upset. Two research organizations have been collecting practical research data for over 10 years to date in URT. This research includes over 1,000 airline pilots to date. Each pilot participated in a three day URT program, and we're given the syllabus, and briefed prior to beginning the course. Although the participants did not know which type of an upset they would encounter, they all knew they would be exposed to URT. The first battery of upsets were evaluated to measure the performance of the participating airline pilots.

Conclusion; over 75% of airline pilots could not recover the aircraft on the first attempt. Now the good news; after the three days of training, over 90% of these same pilots could recover the aircraft from a littany of upset scenarios.

4. There exists a marked decrease in the number of existing airline pilots who have actually been exposed to realistic upset recovery training (in an actual aircraft). The number of former military pilots who have extensive training in this arena is decreasing. Very few civilian trained pilots have exposure to aerobatics or an actual aircraft upset environment. This trend is increasing.

5. Industry has produced an excellent training aid in the "Upset Recovery Training Aid, version 2". However, unfortunately, it has not been widely adapted as a training resource by industry.

6. Full Flight Simulators (FFS) can be a valuable training tool for some URT scenarios. However, many times they are used in-appropriately. The aerodynamic data for a FFS is normally only the "normal flight envelope" of the type aircraft. Therefore, any "extreme" maneuvers out of the normal envelope may be protrayed in-accurately. In addition, and obviously, the capability of replicating the actual "G" environment and acceleration forces is extremely limited. To say nothing of the pshycological aspects of a "real" upset as opposed to a "simulated" upset.

7. Often the instructors utilized by industry in the FFS have no specific or advanced training in the URT regime. This can lead to in-accurate training or a failure to ensure particpants are receiving the appropriate training.

8. We cannot assume that proficient URT performance in a FFS indicates that a professional pilot can perform to that standard in an actual aircraft upset. Indeed, the practical research mentioned above indicates that the very high-stress environment of an actual aircraft upset requires a pilot to actually experience that environment before they can derive the discipline and "experience" necessary to deal with URT.

9. Regulators have contributed to "poor" performance in this regime with negative training. For the FAA, the practical test standards for approach to stall recovery that emphasize "minimum loss of altitude" over regaining aircraft control by reducing AOA and un-loading the aircraft have had negative consequences. In addition, there still exists no mandated URT training (with the exception of the new MPL). We no longer require "spin" training. It is entirely likely that a pilot completing a professional training curricula will never be exposed to an "in-verted" state, or perform an actual approach to stall in a real aircraft.

On top of these issues, there are general trends in industry which may be exacerbating the threat of LOC and the risk of lower URT pilot performance;

1. The over-reliance on Automation (a basic conclusion of Human Factors experts is that humans are "poor" monitors. Indeed, the better the reliability of a system, the less effective the human is at monitoring the performance of the system). 2. A decline in manual flying skills (handling issues). 3. A decline in academic knowledge of aerodynamics. 4. Complexity of automation and flight displays which, when hardware failures occur, mask or increase the complexity for pilots to analyze and recover the aircraft.

Please provide your input on any area mentioned above, or concerns you have regarding how we can best prepare professional pilots to meet the challenge of succesfully preventing LOC.

ps.....I apologize for the spelling and grammatical errors....can't find spell check here....

9th Feb 2010, 03:50
Interesting proposition, but I wonder how many people have tried or done this before...

Yes, you can simulate some of this stuff in full-flight simulators. Yet there are MANY pilots out there who have never gone beyond the required stall demos of a Private or Commercial license, so 3 days of training is some kind of simulator scenario will likely help them considerably. At the other end of the spectrum is the military tactical or "fighter" pilots who have encountered intense, repeated out-of-control training (and, likely, at least a few occasions of unplanned out-of-control flight).

Look to the military schools for the experience in such training...

9th Feb 2010, 03:50

Things would be better if all pilots were required to read and be able to intelligently discuss the following books:

"Stick and Rudder"

"Handling the Big Jets"

"Fly the Wing"

All discuss the concepts you mention to varying degrees...the last two are especially good for swept wing jets.

galaxy flyer
9th Feb 2010, 04:10
Well, Intruder and protectthehornet pretty much said it all!

Being in an out-of-control fighter does impress.


9th Feb 2010, 04:32

A nice approach to the problem and a good analysis.

But, without wanting to belittle the effort, after almost 3 decades in civil aviation, I could have come up with something very close in 30 seconds. Almost any postholder safety in any airline will have also come up with the same conclusion, remedies and will have faced a sympathetic head waggle by his superiors, just to have all the good ideas scrapped due to lack of funds due to the cutthroat competition who has just cut their own new safety departement's good ideas due to lack of .... etc. etc.
It's not the awareness of the situation, it's the willingness to takle it that is the main problem. As long as you can get away with less action and more profit, it will be done.

One huge cost saver has traditionally been the FFS. To a certain extent it has brought in some improvement of training, but it has perverted into the belief, that you can train everything in it. Zero flight time on type on a first commercial flight is not uncommon today, approved SIM time replacing it. Training upsets in a simulator is only half of the truth. The noises and environement are not the same, the G-loads are not the same and all too many times we conclude "oh, that was a sim glitch", "in the real plane you would have done better" ... and the box gets ticked.

Nothing replaces the real thing! Too much hands-on-the-biest-training has been taken away, too much restrictions on hand-flying is imposed.

A second huge cost saver is the more serious threat: CBT. Computer based training can make up to almost an entire technical and aerodynamical training on certain syllabi. The remaining minuscule classroom lessons are given by nice ground instructors. That helps in certain ways, if they are really competent (?), but we are lacking the exchange of knowledge in the good old way the homo sapiens is designed to learn: By beeing tought face to face by someone who has the experience, who has seen it before. Not by a synthetic voice that can be skipped, not by a computer that has only the one sided database, dumbly repeating software or programming errors.

Again, nothing replaces the direct transfer of knowledge, the real thing strait from the horses mouth.

At many airlines we see very discrete fig leaf efforts to regain a bit of flying skills by implementing FFS manual sim sessions (from above you may derive what chance i would give such plans). No wonder mostly old dogs perform better than the more recent pilots. Not necesseraly because they have more experience (that might even be a negative sometimes due complacency or own made up procedures), but mainly because they received a much broader training and aviation education at the beginning of their carreer. The young generation is simply deprived of such and there looms the danger!

bburks, the most stringent part of your efforts must be the convincing of the whole industry (manufacturors, airline managers and the regulators) to go back to REAL training and cut back on synthetic training. It will be more expensive though.

Therefore I may cite Leslie Nielsen: " I wish you all the luck in the world ".

9th Feb 2010, 05:36
I participated as a test subject in the development of the upset recovery program with these people almost a year ago. I thought it was one of the best training experiences in my almost 41 years as an airline pilot.
Welcome to The NASTAR Center (http://www.nastarcenter.com/index2.php)

9th Feb 2010, 05:57
If academics would stop trying to re invent the wheel your questions would be of no consequence.

The very fact that you see a need to ask what is wrong tells me that the training system of today is lacking, I have been ridiculed in the past at work within the European system for questioning the NEW JAR/JAA training regarding single engine training, basic handling, including stalling in the turn, steep turns at 60degrees, soft field and stol take offs and landing, landings with no flap.

These listed are some that come to mind that when I ask someone to my right a Q they get that funny is he serious quizzical look.

I was taught to use the ASI as a angle of attack indicator, ahhh! You say!
I was taught to think heading airspeed time distance,
I was taught to always carry a flash lite, (and have needed it twice)
I was taught to carry a prayer wheel at all times.
I was taught to carry the prayer wheel instructional book at all times.
I was taught to fly scanning the instruments on the other side of the cockpit.
I was taught to fly an approach using the instruments on the other side.
I was taught to time Icing.
I was taught to think of more than one way of checking fuel.
I was taught to listen to the NDB/VOR Ident when using such.
I was taught to understand that the most dangerous approach is the circling.
I was taught to never be afraid to give the landing away if unhappy with it.
I was taught to take an aeroplane back to the apron if I had any doubt at all, and screw the company and the engineers that said other wise.
I was taught to think holding patterns and holding at a limit/ radial, N.S.E.W.
I was taught runway in front, airspeed on the clock, sky beneath me, fuel in the tanks.
I was taught safe circling areas.
I was taught safe altitudes.
I was taught to check Engine driven fuel pumps.
I was taught what would you do if????
I was taught to use other things like engines, trims, doors, etc etc to control/steer an aeroplane
I was taught spinning and recovery.
I was taught Aerobatics.

Now you may laugh, and go ahead.

I my era most became flying instructors at 250-300 hours and moved onto a twin job, then a bigger twin or turbo prop, or agg work or float work.

Today it’s driven by training towards the airlines and the airline way of thinking, it makes no allowance for those that may want to fly other avenues.

The basics used to be drummed into us when I started in 1974, today if I spoke to a Trainee, or F/o as my instructor spoke to me back then there would be hell to play.

The amount of poor handlers came about in my opinion with the reduced training hours and the new corporate image of so called flying schools of today.

In my era it was not unusual for the aero club instructors to fail students (this was before foreign students became the bread & butter)

When I flew in the USA circa 1986, the flying handling standards where as good as what I had in NZ and apart from having to do turns around a pylon, and soft field T/o's & Ldgs, all was pretty same same.

I trained under a then extremely good NZ system with some absolutely wonderful and dear men.

I flew with FAA guys like Glen Veale, & my first CP the late Muzz Brown.

In Europe I flew with some equally good men, all of the same elk, PA, MB, EN, and BF, but by 2004 the new way was even getting rid of these men.

In my opinion it all started to change with the corporate flying schools, passing students due to one being the money and two, as an instructor told me, I am only happy he is not going to fly in this country commercially.

So you ask how can you be assured pilots aren’t going to loose control, you cant! Unless you go back to what was taught before.

The one time the devil outstretched his filthy hand towards us was on a dark dark wet night, and not covered in your normal chain of events/scenarios training, but a unseen little problem that when told to engineering prior to the event, they stated they could find nothing wrong, and said I was the only one complaining, no one else had, (ooopps dont want to get a red mark beside my name and piss off the director engineering, do we now!)

The ability to cope when hugely overloaded came from doing aerobatics and hearing at the precise moment, all those years latter my old instructors (BJT) voice and patter on attitude & stalling:

The reason we never stalled was because I was taught to feel it, I did not blindly rely on a stall warning, the stall warning didn’t work anyway, in this case either aurally, nor by illuminated light due to the aerodynamic attitude.

I make no apology for ranting:

Oh yes, and I inquired into becoming a flight instructor recently in NZ, and was told I could not even do simulator instrument instructing because I don’t have an instructor rating.

Experience its not wanted any more, not valued, and not worth anything.

In saying this the enigma of it all I love flying, love it to bits, and I am aiming at one last gig in Northern Australia, God I hope I get it.

As for the Aviation Industry, I loathe it with all my might.

I would sooner share the company of whores, to that of Liars and thieves, to which the aviation industry use to be full of Flying whores, now it seems the latter occupy the various departments and revenue planning offices about the globe.

And you ask how to change it!


9th Feb 2010, 07:30

I think the question you should be investigating is not why pilots cannot recover from upset situations, but why they get themselves in an upset situation in the first place.

There are a number of threads running on PPRuNe at present which are dancing around this issue given that airlines have found a new favourite way of committing industrial homicide. Gone is the favourite controlled flight into terrain to be replaced with loss of situational or attitudinal awareness.

hoggsnortrupert has, in my opinion, pointed to the root cause of the problem and you have alluded to it in your question, training.

I live and fly in Australia which still has a strong GA environment which is the traditional way of giving pilots experience before they are let loose in the big stuff. In the past a pilot could expect to spend several years in GA before moving on. This gave them extensive hands on flying experience including flight in instrument conditions. They took these skill into the airline environment where their captains had also come up through the ranks with the same background. The result was a cockpit experience level that could cope with automation failures, upset conditions, wind shears and the like.

The current trend is for "fast-tracked" cadets who do not possess that real world experience. Now it is not possible to take a European cadet and give them the GA experience that an Australian pilot will get but it is possible to extend their training to include much more time in a simulator dealing with these types of problems and how to recognise it and deal with it without the need for an autopilot.

It is not happening today and the results are in the crash reports with increasing frequency.

V1... Ooops
9th Feb 2010, 08:55
Captain Burks:

Transport Canada has also become quite concerned about the matter of "training for upset recovery", and has recently published an advisory circular "Training and Checking Practices for Stall Recovery (http://www.tc.gc.ca/civilaviation/commerce/circulars/menu.htm)" that specifically addresses their concern about the level of training provided to air carrier pilots. What is noteworthy about this initiative is that it is directed to the business aviation and air carrier community - people who use FFS for training - and not to the training or ab-initio community.

You might want to have a look at this circular (download the "Word Format" document, the HTML page is a bit distorted) - and, if you think it applies to what you are looking at, feel free to contact the staff at Transport Canada, I am sure they would be interested in supporting your efforts.


Capt Groper
9th Feb 2010, 09:37
Very good topic and one many airlines are now training, but what are the best simulator exercises. There are many good movies, PPP, etc but it's all talk. Nothing beats the practical application.

Conclusion; over 75% of airline pilots could not recover the aircraft on the first attempt. Now the good news; after the three days of training, over 90% of these same pilots could recover the aircraft from a litany of upset scenarios.

bburks - I would be very interested to know what the training entailed. Both classroom and simulator, not just the theory but the actual exercises.

Oh grammar and spell check - just type in word first then check, correct and paste to this medium. :)

9th Feb 2010, 13:32
see the following link (http://www.iasa-intl.com/folders/belfast/stick-pushers.htm) and

this pprune post (http://www.pprune.org/5500988-post1727.html)


11th Feb 2010, 10:33
Unusual attitude training in the simulator takes less than ten minutes on no motion. Firstly turn off the visuals in order to simulate dark night IMC.
Conduct aileron rolls from 20 degrees nose up through 360 degrees of roll with accent on full control wheel against the stops accompanied by slight forward pressure known as un-loading.

Freeze the simulator at the near inverted position to allow the pilot under training to take his time to study instrument indications. Discuss flight instrument indications – sky pointer indication etc. After completion of that part of the training have the instructor demonstrate recovery from nose high.

Gently pitch to extreme 45-50 degree nose high attitude requiring a roll to the nearest horizon in order to regain lost airspeed. Brief that if full aileron ineffective to get nose to drop use rudder carefully to drop the nose. The next sequence is nose low spiral recovery using speed brake and thrust lever closed if excessive speed increase.

Control force fidelity cannot be reproduced of course. But as most loss of control is in IMC and/or night, it suggests the key to successful unusual attitude recovery is basic raw data instrument flying skill. Briefings, power point demonstrations and general discussion on UA training are fine; but there is nothing like hands-on practice in the simulator. Simulator costs for ten minutes are minimal for such a vital flight safety sequence.

The simulator instructor must be competent to personally "patter" his demonstrations. A picture is worth a thousand words. Use the simulator as your picture.

11th Feb 2010, 11:02
Most level D simulators i trained in had preprogrammed UA recovery training situations. With a click of the instructor the simulator moved into an unusual attitude and we as trainees had to recover. It used to be a standard exercise in every simulator training (every 6 months, right after stall recovery training), but lately i haven't done it, mostly due to lack of time. Things seem to have changed, where we were required to do at least one single engine raw data approach on every simulator event we now can use the flight director and autopilot, which is watering down the training effect in my opinion considerably, but alas i'm just a line pilot, not a training manager.

11th Feb 2010, 11:29
The use of "canned" UAs I have always found dreadful and counterproductive (other than for box-ticking purposes).

I echo Centaurus' comments.

From that skill position, as the skillset improves, I have always found it useful to have the pilot close his/her eyes while I put the aircraft into whatever progressively ridiculous position so that, on eyes opening, the pilot is faced with a dynamic, time critical situation requiring very rapid data gathering, assessment, planning and execution.

Invariably, the initial exercises are a bit tatty but almost all progress rapidly. A critical secret to the tale is to make the environment non-threatening so far as the instructor is concerned so that the pilot doesn't waste cognitive effort on "what is the guy in back thinking" and puts all his/her effort into the task at hand.

The comments re sim fidelity and modelling extrapolations are noted and very pertinent. However, providing that the sim presents a "reasonable" picture to the pilot, it still can be useful as a generic training aid in UAs.

galaxy flyer
11th Feb 2010, 14:52
The mob that presently employs me sends us to National Test Pilot School at Mojave every 3 years for upset and formation review. A day of ground school and a couple of flights in an Impala. Aside from making the Tweet look powerful--very instructive. I hadn't done any acro or form since leaving tac fighters, but surprising how quickly recovery skills came back. The profile is stalls, spins and lots of UA recoveries--set up and "self-induced". The non-military types that hadn't done any acro really benefit. No, in one flight no one learns much about formation except some respect


Miles Magister
11th Feb 2010, 15:12
This training is woefully inadequate in the civilian comercial world.

You may wish to look at how the RAF Central FLying School teaches pilots to recover from Unusual Positions both visually and on instruments as it is very good indeed. They are taught right from the begining of training and practice it regularly so it is instinctive.

You may also like to look at the King Air incident out of Glasgow a few years ago for a good example of how not to do it.


11th Feb 2010, 15:22
Captain Burks,

I took the Calspan in-flight upset-recovery course a couple of years ago. What most impressed me was the initial pre-training upset encounter in the Lear. I had practiced the "Pittsburgh scenario" many times to the point that a lateral-directional recovery became automatic although I was never very happy with my performance.

In the Calspan course, we performed all of the recoveries in a fixed base sim the day before the flight. In my pre-training encounters, one was a lateral upset (aileron actuator hardover or a Roselawn departure). I did my well-rehersed "Pittsburgh scenario" recovery -- full aileron/full rudder, when that's not enough unload to make the ailerons more effective. It's amazing how fast the Lear ended up on its back. The IP (with whom I had flown many many years prior) was laughing his --- off.

Later in the training, I had a rudder actuator hardover. The difference from the simulator was striking. One's head is thrown to the side by full rudder at 250 knots. In fact, I called engine failure.

In my opinion, the standard hexapod simulator presents misleading cues in this event. If you couple that with poor aerodynamic models, I think we really need to rethink how we train for Loss of Control.


11th Feb 2010, 17:28
Hi, have returned to the snow country.

Good information.
It is also possible that throttle control may help to lower the nose position.


11th Feb 2010, 17:33
The basics used to be drummed into us when I started in 1974, today if I spoke to a Trainee, or F/o as my instructor spoke to me back then there would be hell to play.

The amount of poor handlers came about in my opinion with the reduced training hours and the new corporate image of so called flying schools of today.


12th Feb 2010, 01:56
Google 'Advanced Aircraft Maneuvering Program' or AAMP.

Googled more and came up with this link. Read the last couple of replies -


12th Feb 2010, 18:36
Simulators are generally very restricted in what can be practiced in terms of disorientation – leading to LOC.
One good demonstration is to simulate IMC and apply a small out of trim condition, preferably with autopilot engaged, before getting the crew to disengage the autos.
The objective is to experience how difficult it might be to disregard some sense inputs (feel) and concentrate on the aircraft instruments. In some LOC events the body senses, particularly acceleration can be very disorienting.
Out-of-trim in roll is very effective (disturbing) – e.g. mishandled lateral fuel balance, but note the number of incidents – stall etc which have involved pitch trim.

bburks – see PM