Flashair Airlines Boeing 737 accident - a lesson from the past.

[Centaurus]

It is said; “Those who forget the past are doomed to repeat it”

In 2004, many of today's new airline pilots would still have been at school. They probably would not have known about the horrific crash in that year, of Egyptian Flash Airlines Flight 604, a Boeing 737, that went out of control after getting into an unusual attitude in IMC soon after take off.

It is worth reading what happened and how, with the right training, that type of accident can be avoided in the future. New airline pilots especially, need to be aware of past accidents and learn from them. In this particular accident, it is clear that automation dependency leading to disorientation and the captain's apparent inability to competently hand fly on instruments, were major factors leading to the accident.

Added to that combination, was the cultural aspect of the first officer's reluctance, or lack of flying ability, to take over control when it became clear the captain had lost the plot and was incapable of recovering from the ensuing unusual attitude.

Although this was an overseas accident, it is all too easy for today's pilots to be complacent, knowing that airlines in Australia have an exemplary flight safety record; primarily due to good training

Unusual attitude training in a simulator is regarded by some as just another regulatory box ticking exercise. It is not helped when simulator instructors and check pilots are reluctant to take a control seat and demonstrate the various unusual attitudes and their correct method of recovery. A good demonstration by a competent simulator instructor is worth a thousand words but unfortunately it doesn't happen very often. The pilot is left to read all about it in the flight crew training manual and hope to replicate the manoeuvre in the simulator.

Read the following links to see what happened to the Flashair 737. Start with the first link which depicts the youtube.com animation. After viewing all three links, new pilots will perhaps then understand the importance of effective unusual attitude training in the simulator.

https://www.youtube.com/watch?v=GCLD6bwGq1I

Flash Airlines Accident - 3 Jan 2004 (http://www.b737.org.uk/flashair.htm)

CVR Database (http://www.tailstrike.com/030104.htm)

[50 50]

I wonder how many "simulator pilots" have done a significant amount of unusual attitude recovery in an actual aircraft? Interpretation of instruments is one thing, but is it the same when your @rse is floating off the seat and your stomach is in your mouth?

Unusual attitudes are felt as well as seen, and G tolerance takes time to build up. I think unusual attitude recovery is a fundamental flying skill that should be an integral part of basic training, not an add on extra. I don't mean a single lesson either,but a continual incorporation with the standard syllabus.

It is well documented that pilots with an exposure to aerobatics and advanced handling revert back to basic stick and rudder skills, built into muscle memory, in the event of an unusual occourence. No matter what the aircraft.

Simulators just don't cut it, burn me at the stake, but they don't actually fly.

[Oakape]

Kenya Airways flight 507 out of Douala & Ethiopian Airlines flight 409 out of Beirut come to mind.

[PLovett]

I believe that, according to ICAO, loss of spatial orientation is now the favoured way of committing industrial level homicide in the airline transportation world having taken over from the former favoured way of CFIT.

Says something about the method we are teaching people to fly in IMC these days. In both the Kenyan and Ethiopian Airline crashes the CVR records the pilot flying (in both cases the captain) asking for the autopilot to be engaged. In the Kenyan case, and I believe also the Ethiopian one, it failed to engage due to control forces being too high at the time. With the Kenyan crash neither pilot had the wit to observe the MAP was still showing FD rather than AUTO as a near new 738 rolled over and into the mud. No-one was monitoring the flight instruments.

ATSB paper (http://www.atsb.gov.au/media/29971/b20070063.pdf)

[A37575]

Simulators just don't cut it, burn me at the stake, but they don't actually fly.

From what I read in most of these unusual attitude crashes, they occurred at night or in IMC and in each case the problem was the pilot's lack of basic instrument flying skill as he attempted recovery to level flight.

Granted that airline simulators are not designed to replicate high G forces that can occur in attempted recovery from such events as a steep spiral dive or extreme pitch-up . However the flight instrument indications are the required fidelity. This is where simulator training in unusual attitude recoveries come into their own. It teaches recovery on instruments

[50 50]

A-lots of numbers.
I understand the purpose of a simulator. I don't understand how one can be expected to perform an unusual attitude recovery, at night, in IMC, when they have never been exposed to it in day VMC, with a clear horizon.

You could teach a non pilot to fly an instrument approach in a simulator, but surely you would not let them have a crack in a real aircraft, especially if they can't fly a day VMC circuit. I can't think of a more demanding task that a pilot could have to perform, and I just don't see how simulator training alone is sufficient.

It removes the physical sensations, and the "oh Sh!t I'm going to die" panic factor. The same recovery techniques can be taught in a real aircraft, that is fit for the purpose. They are few and far between however, and there is the basic problem. 90% of training aircraft are not suitable for unusual attitude and mishandling training, therefore it is not covered, or only covered in one or two lessons.

There are innumerable pilots and instructors that have never been beyond 60 degrees angle of bank, and then they are expected to perform a recovery, at night, in IMC after some simulator training.

[waren9]

is it really that hard? blue side goes at the top.

[kingRB]

actually alarmed at how many references were made to "god" and "god willing" by the crew

[Capn Bloggs]

It removes the physical sensations, and the "oh Sh!t I'm going to die" panic factor. The same recovery techniques can be taught in a real aircraft, that is fit for the purpose. They are few and far between however, and there is the basic problem. 90% of training aircraft are not suitable for unusual attitude and mishandling training, therefore it is not covered, or only covered in one or two lessons.

There are innumerable pilots and instructors that have never been beyond 60 degrees angle of bank, and then they are expected to perform a recovery, at night, in IMC after some simulator training.
I think it is a bit unrealistic to expect airline pilots to do U/A training in a fit-for-purpose aircraft. However, I think it is definitely doable in the sim. Start off with wingovers, simple but getting harder cross-ref exercises, then go on to barrel rolls and if your sim will do it, loops. The critical issue is not the seat of the pants feeling (which you have to ignore anyway) but getting your brain to reliably interpret rapidly changing instruments.

Then move to closing your eyes, get put into a (at first gentle) UA then recover. Then ramp them up as proficiency increases. Some people don't like closing their eyes. IMO that's what has to happen to get any real value out of it. Somehow we have to train against the startle factor and sitting there, eyes wide open, while checkie Bloggs sets up a UA is just a waste of time.

The other point is that you have to do enough of it, and recurrently, to maintain the skill. Time for an extra day in the sim...

[50 50]

I'm not suggesting that you go out and barrel roll a 747, though it has been done. I'm talking about during initial pilot training in a light aircraft, that is fit for the purpose.

If you're doing a command upgrade in a turboprop aircraft, then you must have considerable experience, the question now becomes "why does stalling make you nervous"? Stall recovery is a fundamental flying skill that should be as easy as breathing. It also has the advantage of being able to sustain your life.

Blogs is 100% right, it has to be done often enough to become muscle memory.
How many pilots out there haven't stalled an aircraft since their GFPT? Lots I'm guessing, yet people expect to be able to draw on a long forgotten skill right when your life depends on it.

Having said that, provided there is air underneath you, one can keep an aircraft stalled as long as you want without hurting it, or yourself. Stalls and unusual attitude recovery should be taught from the very beginning, but they just aren't. 1 or 2 lessons is insufficient, and give the illusion of competency.

If half the problem is the crew let it get there in the first place, then surely the other half was an inability to recover.

[Jabawocky]

I don't fly jets….point 1.

I have to agree with bloggs, but don't tell him his head might swell if he feels a fan club starting up. :E

Add to that some real flying in a suitable GA machine just to keep it real from time to time.

I have spent a fair bit of time learning all I can from Chimbu Chuckles and if you read his stories here on pprune, raw grass roots flying and attempting to make jaba sick (so far failed but come close) keeps his stick and rudder skills at their peak.

I am sure if presented with a real UA in his transport jet he would have no issue in recovery and not even raise a sweat. Why? because it is a non event due to lots of sim and real world practise. (please do not take 'non event' too literally).

My next renewal will be a full day booked as I want a heap of UA and all manner of things to do in addition to just flying approaches and ticking the boxes. Why? Because I subscribe to the Bloggs theory above. And it is a good challenge/fun :) .

[Arm out the window]

I don't understand how one can be expected to perform an unusual attitude recovery, at night, in IMC, when they have never been exposed to it in day VMC, with a clear horizon.


IFR flying lesson 1, don't believe your senses, get on the AI, set power and attitude and start scanning.

I agree it's good to have done 'robust' UA recoveries in the real thing, and the points you raise about how it feels are valid, but the crux is not to rely on the senses, as they may be telling you lies. However, if you can fight whatever your body's telling you and just go through the steps to recover methodically, you'll be OK, or at least as OK as it's possible to be under the particular circumstances.

[BEACH KING]

Usual attitude?
The bloke that does our BFR's makes you put on the hood, and then tells you to put your head as close to your arse as you can get, and then proceeds to makes the world appear to be inside out, back the front and upside down. If you stuff up the recovery..he makes you do it again until you get it right. Then when you do the night check, he does it all over again.
The most unusual attitude, is the trip to the pub afterwards to point out (over a couple of beers) where you need to lift your game.

With the flashair accident, I seem to remember that the Captain's AH went US, and he took no notice of the FO and standby AH's and the advice of the FO and FE.

[Delta_Foxtrot]

Ladies and gentlemen, I owe the last twenty years of my life to the simple UA recovery techniques I was taught, and later taught to others, at Point Cook in the CT4A.

Rolled upside down at about 150 ft AGL on take-off at Wagga due to wake turbulence, instinct kicked in and I survived. Unfortunately the aircraft didn't, but you can always build a new one. I still haven't figured out how to build a new me.

Btw, it wasn't my fault - controlled aerodrome at the time; not enough separation applied and I'm not entering arguments.

[A37575]

With the flashair accident, I seem to remember that the Captain's AH went US

There was no evidence of that either in the official report or the CVR. Maybe you are thinking of an another accident?

[Wizofoz]

He's probably thinking about the Korean Air 747 freighter out of Stansted.

[Jabawocky]

Beachie, I think Wiz is right, wrong prang :ok:

So how do you go when the ato hands you a perfectly S&L plane after tossing you around? One local bloke used to live doing that. He reckons a large number then upset the plane!

:ok:

[BEACH KING]

No, Flashair prang was the one I was thinking about, but A37575 is correct, there was nothing wrong with the Captains AH in the official report.
My incorrect recollection was from the "air crash investigation" program where the Egyptian investigators were initially pushing the crook AH theory.
The sad part is that the other two crew knew what was happening and could not, or did not do anything about it.
As Centaurus says, a lesson for all. I highly rate his book too!

[Anthill]

Many years ago, Ansett determined that the 2 greatest risks to its operation were 1) CFIT and 2) Jet upset. This resulted in Ansett implementing a genuine jet upset training program.

This program consisted of each pilot in the company being given a take home VHS video and workbook. Both provided a comprehensive discussion of aspects of jet upset, CRM/HF considerations, aerodynamics and case studies. High and low altitude upsets were considered.

On completion of the workbook (which included a set of discussion questions), pilots were rostered for a 1 day class room session and also a 2 hour simulator detail.

The simulator sessions were the same across all fleets. It commenced with a wake encounter at the OM SY RWY 16R ILS where the aircraft rolled inverted. Virtually all pilots crashed. At the end of the session, the same scenario was presented and it was so confidence building to effect a recovery and avoid hitting the ground. This Jet upset training and Ansett's excellent cold wx operations training (developed by Finnair) wer two of the most informative and useful pieces of training that I have received during my career.

Upset training needs to be pitched at a level beyond "read the FCTM and remember to roll towards the sky pointer". Low-time pilots, a poor scan, reliance on automation(which has its place) all conspire to pose a bona-ride threat to contemporary (as well as historical) operations.

Centaurus has provided a fitting and timely reminder of the threat posed by jet upset. My opinion is that few organizations have sought to implement a training program that fully addressed a long identified operational threat. What has been learned from AF447? There has been attempts to discuss jet upset and recovery and this is useful. However, in the absence of a formalized training program, the industry is relying on the skill of individual instructors to impart the requisite information. This is simply not good enough. Companies need to implement a formalized training program and ensure all crews are trained to a common (high) standard.

[Lookleft]

I remember the Ansett training Anthill and it was an eye opener. Every time I depart off 34R it reminds me of the final engine failure exercise. The downside is of course the incorrect application of jet upset techniques such as the American Airlines A300 prang in New York. Thats when I found out about the limitations of the certification of full rudder application.

A lot of the jet upset training came into being because of the 737 accidents in the US. Since then there doesn't appear to have been rudder hardovers in 737s. Either the training has kicked in or the rudder PTUs have not "misbehaved".