Loss of control, loss of nearly 2,000 people in crashes

As the Air Asia Indonesia accident investigators confirm the crash was caused by loss of control following an electrical snag, the tally of people who have died unnecessarily on commercial airliners has taken another step up.

There have now been 18 loss of control accidents since the year 2000, and 1,886 people have died in them because the pilots failed to maintain control of aeroplanes that were completely flyable, and most of which had nothing wrong with them.

The Air Asia accident involved an Airbus A320 at 32,000ft in the cruise over the Java Sea last year on 28 December 2014. The report says an electrical fault – known to the airline and the captain but not resolved – caused an alert to be repeated three times before the captain attempted to resolve the issue by tripping and resetting the circuit breakers for the flight augmentation computers.

The autopilot had been coping with the control effects of the electrical fault, but when the FACs were switched off the autopilot tripped out and left the pilots to fly the aircraft, and they clearly were not ready for that.

The electrical fault was caused by a crack in the solder on a printed circuit board associated with the rudder travel limiter, which prevents the rudder being deflected too far at high speeds. As soon as the autopilot was disconnected, the effect of the fault was to offset the rudder by 2deg, which is not much, but enough to cause the aircraft to roll left to a bank angle of 54deg. Most airliners bank about 20deg (maximum 30deg) for ordinary manoeuvres on commercial flights.

The copilot was flying, and he failed to take action immediately to roll the wings level, so the nose dropped. Some 9sec later when he did roll the wings almost level he also pulled the nose up. Then the bank angle returned to 53deg left, and the pull-up demand on the copilot’s sidestick moved to maximum, actions that suggest the copilot was already seriously disorientated. The aircraft climbed to a maximum height of 38,500ft, stalling on the way.

Once stalled, it descended at a rate of 20,000ft/min into the sea.

The pilots never recovered from the stalled condition. As in the AF447 tragedy the copilot’s nose-up demand – the opposite of what was required to regain control – continued.

There is some evidence that the captain may have left his seat to trip the FAC circuit breakers. At one point in this upset he gave the copilot the confusing instruction to “pull nose-down” (the pilots were different nationalities and neither was a native English speaker), but he then failed to act correctly to take override control with his sidestick.

The industry knows it has this huge weakness in its pilot workforce. The death of 1,886 people since 2000 testifies to it.

There are various components to the problem:

  • highly reliable and accurate automated systems in today’s aircraft mean pilots almost never get the physical or mental exercise of controlling the aircraft and its flight path, so many are not ready when they have to take control;
  • statutory recurrent training requirements are out of date and do not relate to the task of today’s pilots in modern cockpits;
  • most pilots now have no training for recovering aircraft from upsets involving significant attitude deviations from straight and level;
  • most have never handled an aircraft at high level and therefore are not familiar with how small the flight envelope is in thin air, and what to do if the aircraft goes outside the flight envelope (like entering a full stall).

Some airlines, in countries where the aviation authorities allow advanced airlines to vary their training according to evidence of need, the carriers are dealing with this weakness.

But in others where the old recurrent training requirements still dictate training minimums, airlines are still working to the minimums.

And there are even questions about whether, under the stress of aircraft malfunctions or upsets, some pilots’ brains just go all funny and there’s nothing training can do about it. In Toulouse, France, the ISAE is researching this.

Meanwhile, it is a tragedy that, in an industry that is very safe and getting safer, there will inevitably be more of these unnecessary fatal accidents. It only takes the smallest snag to trigger one.

7 thoughts on “Loss of control, loss of nearly 2,000 people in crashes

  1. An AF447 repeat, substituting an A-320 for an A-330 and 300ft Sea depth instead of 13000ft Oceanic depth, a burst belly, crumpled crown, connect both in a watery photo finish, 5.5 years apart. Not to mention, 3.5 years after AF447’s PF/PM “DUAL INPUT” promulgations WERE FULLY KNOWN above the then oceanic abyss. The Airbus latent design defects of the “man-machine” interface are interminable and in political perpetual light. The “air-machine” interface tends to be obscured by it in the fog of the perpetual political-economic war Between Boeing Orthodox and Airbus putative Kathilikos.
    On and in a B-777/787, in Secondary Law, the primary pitch trim switches on the yoke and pedestal alternate pitch trim levers move the stabilizer directly. Unlike Normal Law, there is no autotrimming to maintain a trim reference speed.
    QZ-8501 of Dec 28 2014, crashed with FULL STOP, nose-up travel of the horizontal stabilizer in Alternate Law, like AF447 and its XL Airways Perpignan A-320 entré of Nov 27 2008-AGAIN !!!. Airbus FBW X3, Boeing FBW X0up to and including Dec 28 2014. In short, by substantial contrast, on a B-777/787, upon the same “LAW REVERSION” circumstances, the pilots will have to make a conscious and physically tactile effort to kill themselves. Airbus removed this effort on Feb 22 1987.
    An Airbus Alternate-Direct Achilles -clear and present danger- tail, if ever there was one. Luckily, there is just that one, albeit, replicated from A-318-350 progeny, to all inclusive, so “Hi to Low, look out below!”.
    The Airbus Achilles tail, to wit:
    Boeing NORMAL Control Law prioritizes speed stability, subordinating pitch stability to speed stability in the SECONDARY Law of INTENDED consequences.
    Airbus NORMAL Control Law prioritizes pitch stability, subordinating speed stability to pitch stability in the ALTERNATE Law of UNINTEDED consequences.
    That is the air-machine interface of the Boeing-Airbus divergence deficit upon the abyss of sea and by the ocean in perpetual convergence.
    3-0 to Airbus in FBW premier league, so far (and tragically counting) of an autotrimmed locked-in, artificial deep stall from stick-pulling aft by a panicked pilot, without a nil-all draw. The man-machine interface, in addition to the much debated non-control unison, non-tactile (except below 100FT RAD ALT Flare mode) sidesticks, philosophically contradicting control unison rudder pedals and static autothrottles, is a basic man-handling issue. With non-drone sidesticks and a TOTAL control law philosophical contradiction in unison, disunion aggregate.
    To be an Airbus “cunning linguist”, one needs to be Toulouse-savvy in revealing the ubiquitous French “Mini Manche” term. Because of its size, there is a natural “Maxi” bias to OVERCONTROL. A man-sized yoke has a natural bias to UNDERCONTROL. We should be grateful that Airbus have not diversified into land busses. Seriously, thus far!.
    For illustration of this, the Airbus office ”malicious beast” in anger, I suggest watching the 7 minute video on “Les 4 dernières minutes du vol Air France 447 AF447”. It is the 99% emotive accurate CVR –by attributed final F/O PF right seat, erroneously spoken from the left seat, in the last 30 seconds, on this simulation of AF447’s demise in an A-330 Sim. AF447’s last words were spoken from the right seat on June 1 2009.

    The Airbus control dysfunctional dichotomy is also evident of having non-tactile disconnected sidesticks and tactile connected rudder pedals on this video, as I previously “right rudder, ruddy well said !”.
    The hapless captain on this AF447 flight, could see that desperate-dysfunctional AA587(Nov 12 2001) rudder dance at least, with 1 min 20 secs prior to Oceanic impact, strictly returning to pundit commentary. Airbus’s Bernard Ziegler’s design co-pilots of this control system apparently moved on to greater things in designing the Euro currency at the ECB in Frankfurt, using the same design philosophy. The end result of that was GDP translating as Greek Decimal Point.

    On 21/03/2001 a Lufthansa A-320, experienced a flight control malfunction shortly after take-off from Frankfurt/EDDF RWY18 to Paris/LFPG. Thereafter, loss of control and subsequent near terrain impact was occasioned. The uncontrolled roll, due to the malfunction of the PF’s X-wired sidestick, was recovered by the other pilot and the aircraft safely returned to land. From the BFU report: “Immediately following VR and lift-off the aeroplane assumed a slight bank angle to the left. The commander, who was PF, tried to correct the attitude by a slight input on the left sidestick. However, the bank angle increased continuously up to approx. 22°, within 2 metres of terrain contact. With the commander PF’s call out :“I can’t do anything more!“. The first officer took over the controls with the words “I have control!“ and pressed the sidestick’s TAKE OVER PUSH BUTTON. The First Officer had already beforehand instinctively tried to counteract the rolling movement with his sidestick”. Prior to this flight,the aeroplane had been at Lufthansa Technik AG for two days. QED!.
    As an Airbus pax, after this, I notice 2 control cycles after every push-back/start. In aggregate this is another addition to Climate Change conferences, that cause me jet/place lag by their very time-serving/self-serving mention of acute inanity.
    Airbus Economic Genius. Later seen, allegedly, in the ECB in Frankfurt translating the Euro into Greek. That became GDP as Greek Decimal Point bookkeeping. They remarketed these A-340 lemons as, not their contradictory “4 engines for long haul”, but, “4 engines for the price of 2”, neglecting to state the ETOPS/LROPS bleeding air obvious, “4 engines FOR the price of 2 AT the cost of 4”. The A-343 only climbed at a tangent, due to the earth’s curvature. Aided by 4 hairdryer powerplants.
    http://www.bloomberg.com/bw/articles/2013-12-02/airbus-to-customers-your-200-million-plane-is-not-a-lemon
    Learmount is Boeing/Airbus Catholic, Connolly is Boeing Orthodox. We shall eternally agree to disagree. But, at least in Learmount.com in this instant 24/7 rolling News/Noise, world, reasonable discourse prevails. Religion requires study, humanity requires faith of reason in man’s humanity. And a mind is like a parachute, best used when opened. Boeings are designed by genius for idiocy, like mine. Airbuses are, literally, the inverse of this truism. I’m happy being an idiot keeping the blue side up.
    The F/O on QZ-8501 reverted to his native French language in AF-447-like self-induced stall stress. IDEM to AF447;
    None are so blind, as those that WILL NOT see. And Boeing FBW would be a shadow of itself, if not for Airbus FBW pioneers. After all, imitation is the sincerest form of flattery. QED!

    Like

    • Let me ask you this, David: would you, in that aeroplane following that roll upset, have pulled back on the stick and kept it there after the stall warning went off and the airspeed had disappeared? Like the AF447 crew did too, but also like the crew in an MD-80 did over west Africa last year, and like the crew of the Q400 did over Buffalo a few years ago?

      No, you wouldn’t have. Or at least you think you wouldn’t. I think I wouldn’t have pulled back like that either, but how would I know?

      We should be asking why pilots do illogical things with the stick nowadays, whether the stick is a big one or a small one.

      Like

  2. Having watched and read a lot about the Air France disaster – including reading the report and recently reading a book about it, and even knowing a colleague who was due to board that flight in 2009 but didn’t, when i first heard of the Air Asia crash and some of the early details it looked very worrying, noticing how eerily similar it looked to AF447.

    I never imagined it would be so close to a repeat of it, it’s simply chilling to read this. It’s as if the pressure of life and death itself makes people get this disoriented, in disbelief that it can be happening – when playing flight sim games (much different i know) as a child it was never possible to accidentally do something like this. You just look at the attitude, speed and altitude and sort it out. But looking in from outside the industry it does seem painfully inevitable when pilots are trained from scratch to do this job, rather than building up over years in more manual aircraft. On the other hand, as you mentioned recently perhaps the ‘video game’ generation will be more successful in these situations, if they have had plenty of practice in flight sims with no pressure and the freedom to do what they want in terms of handling.

    It looks more and more like we had a horrible year of preventable major disasters in 2014.

    Like

  3. Has there been an Emergency Airworthiness Directive to check all the in-service RTLUs – or those produced by the specific (but unnamed) manufacturer – to see if they have a similar soldering fault?
    I haven’t seen anything about that, and its absence worries me – the Safety Recommendations [report Section 5, page 125] is completely silent on this.

    The report says “Airbus informed that the installed RTLU on PK-AXC had been improved with both
    Technical Follow-Ups (TFUs)” [page 66, last two lines], so it was an “up to date” revision level, yet both channels failed.

    Like

  4. Indeed David, I agree. The Colgan Air Q400 operating Flt 3407 on Feb 12 2009, was as instinctively illogical as you say, with a man-sized yoke, mishandled by a woman F/O, all puns intended. I must renew my Viz subscription. And Air Algerie 5017 of July 24 2014 mishandled too.
    That was not only Sully’s Airbus Hudson year, but the year-as I call it- of the 3 stalls, following this was the B-738 THY1951 “Poldercrash” of Feb 25 and of course AF447 on June 1. Quite a vintage aviation year it was. Here is Airbus Sully-whom you know- on Airbus V Boeing.
    Here is A-320 Captain Sully of the Hudson of US Airways Flt 1549 of Jan 15 2009, whom you well know.
    What can I say David, I am the Jeremy Corbyn version of Boeing Orthodoxy.
    Sully on AF447:

    On Boeings, proximate to your mention, you omitted US Airways 427, operated by a B-733 of Sept 8 1994. I remember this well, as I attained my PPL in a Cessna the next day in Oklahoma City with a Vietnam helicopter vet examiner. Needless to say, this was a serious topic of conversation with that day’s task in hand.
    https://en.wikipedia.org/wiki/USAir_Flight_427
    In essence, a Parker Hannifin designed rudder, with a lot of previous hardover induced upsets of Boeing assembled responsibility.
    And this previous fatal preamble was on March 3 1991 on United 585, operated by a B-732. So, from B-737 entry of service in April 1967, this “Boeing Vertical Achilles Tail” took 24 years to manifest itself in an initial smoking hole in Colarado. AF447 took 22 years, in A-318-80 aggregate for its “Airbus Horizontal Achilles Tail”. Boeing bit the bullet, Airbus has dodged it. Two different serial killers in effect. Beckett’s “Fail, fail again, fail better” works well for Boeing, absent in Airbus, of blame the pilot.
    Clearly, the Dickensian artful dodger’s descendants migrated to Toulouse.
    https://en.wikipedia.org/wiki/United_Airlines_Flight_585

    “Push, power, rudder, roll” is the mindful mantra antidote to all Airbus and Boeing confusion. Basically a muscle memory for even rote learners. Personal discipline is paramount though through the ambient discomfort.
    Push=Wing unload.
    Power=Speed preservation.
    Rudder=Yaw corrects roll without wing loading. Step on the sky or PFD blue in the black.
    Roll=Correcting LNAV without adverse loading with panic/buffet margin restored. Climb optional, mandatory, if terrain is imminent.

    Coming back to your original point, pilots -be they Boeing or Airbus will do –being human– illogical things with the controls, perhaps more so on Airbus with an “aft-stick, Alpha-Floor Normal Law, expectation bias”, but with the adjacent stick moving visibly in unison, even with 2 muppets flying with an Airbus background soundtrack of “STALL, STALL, DUAL INPUT !”, there is a better chance of one of the them waking up to reality. How to make objects on terra firma appear smaller ?, pull back on the stick. How to make objects on terra firma appear bigger?, keep pulling back on the stick. Heads are killing more than hands in effect.

    Don Wylie -a mentor of mine- died via a migrating wing on Nov 20 2003.
    https://books.google.be/books?id=d_Mh3wwRJSMC&pg=PA64&dq=Don+Wylie+Aviation+Safety+Training+/+Texas+Air+Aces&hl=nl&sa=X&ved=0ahUKEwid4cz1wsjJAhULtBoKHb4qCikQ6AEIMjAA#v=onepage&q=Don%20Wylie%20Aviation%20Safety%20Training%20%2F%20Texas%20Air%20Aces&f=false

    21st November 2003, 08:51

    • Don Wylie, Pioneer Upset Recovery Instructor, Killed in T-34 Crash

    Don Wylie, the 64-year-old founder and president of both Aviation Safety
    Training and Texas Air Aces, and his passenger, 39-year-old Airborne
    Express pilot William Eisenhauer, died yesterday morning when their
    Beech T-34 reportedly lost its right wing during upset recovery training
    in VMC near Conroe, Texas. AST’s Rick Gillenwaters told “AIN Alerts”
    today that, contrary to earlier reports, there was no collision with
    another AST T-34 that was accompanying Wylie’s airplane. Witnesses in
    that airplane, which returned safely to base, said they saw the right
    wing of Wylie’s airplane fail cleanly at the root, according to
    Gillenwaters.

    The airplane, N44KK, was the same T-34 in which “AIN” contributor Rob
    Mark flew with Wylie in the course of preparing the article on upset
    training that appears in the current issue of “AIN”. A 1999 crash in
    Georgia in which a wing separated from a T-34 during simulated combat
    was attributed to stress-related cracks in the wing spar, according to
    the NTSB, a weakness that led to the grounding of T-34s used for such
    flying pending modifications. Gillenwaters said that N44KK was one of
    the AST airplanes that had not yet been modified with strengthened wing
    spars, but it had been inspected and deemed fit for upset recovery
    training pending modifications, which had been scheduled to be
    incorporated in 60 days or so. Texas Air Aces conducts simulated combat
    flights. Aviation Safety Training teaches experienced private,
    corporate, and airline pilots how to recover from aircraft upsets.
    While on the line or Wylie on the line, the pilots of Alaska Airlines 261, operated by an MD-83 on Jan 31 2000, had this innate muscle memory prescriptive right stuff with a remarkably calm demeanor throughout the descent which had the previous side wind view appear out front and followed this muscle memory. Unfortunately, their plane did not have the right stuff, being stabilizer jackscrew threadbare.
    https://en.wikipedia.org/wiki/Alaska_Airlines_Flight_261

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s