What’s really happening in the real world?
Over on AOPA.org recently, another in the very long line of drearily apocalyptic articles (rants, really) on the dangers of over-automation in light aircraft cockpits, was posted in a point-counterpoint format by Rod Machado, accomplished author and professional flight instructor, and George Perry, the Director of AOPA Foundation’s Air Safety Institute. I say “point-counterpoint,” but their actual presentation was more “point-point” style, as in, both authors agreed that there is a VERY BIG DANGER posed by over-reliance on automation in the cockpit of light aircraft these days.
As I said above, this is yet another in a very long line of such Cassandra-style warnings to us lowly private pilots flying light aircraft. These warnings, besides being largely off-base (as I’ll describe in this essay), are getting to be really annoying to read in aviation journals that purport to speak to the private pilot population.
Does this really have lessons for pilots of small general aviation airplanes?
What’s rather silly about the Machado-Perry duet of Cassandras was that, to make their points about the dangers of private pilots over-relying on cockpit automation and technology, they cited as their scientific examples a dumb automobile driver who couldn’t figure out how to open her car door without a functioning remote keyless entry (A car driver? As a good example of today’s airplane pilots? C’mon, Rod… really?), while the other writer cited the dreary oft-told tale of Asiana Flight 214 and the three professional airline pilots on the flight deck who, between them, were too incompetent to land an airliner, filled with human passengers, no less, on a CAVU VFR day on a long runway with a functioning lighted visual glideslope system.
I immediately sent an email to George Perry with my comments on their “point-point” storyline, essentially taking issue on the following main points (which in response to me, he disagreed), including:
- You’re writing for a membership that is composed mostly of private pilots flying light aircraft, most of which are not nearly as automated as turbine-powered commercial airliners, with most of us flying mostly single pilot. So why are airline accidents constantly and almost exclusively referenced as the proof that over-automation is killing private pilots in light aircraft? As Robby the Robot used to say on those old Lost in Space TV episodes, “That does not compute, Will Robinson!”
- The actual accident data show that general aviation safety has improved steadily and substantially over the approximate 15-year timeframe over which the limited amount of automation and technology has become available in a significant proportion of the light aircraft fleet in the USA. The dawn of light aircraft advanced cockpit technology really took off (pardon the pun) with the introduction at the end of the 1990s of the Garmin panel-mounted 430/530 series IFR certified moving map GPS navigators coupled with modern digital autopilots. All-glass production aircraft like the popular Cirrus SR 22 and others accelerated the implementation of advanced technology in light aircraft. Most older legacy aircraft are still limited in such technology today, but partial updates are gradually spreading throughout the fleet. According to the latest Nall Report, we have experienced some rather impressive results in terms of steadily decreasing GA accidents and fatalities, both in raw numbers and in estimated accident rates (based upon estimated flight hours). The bottom line is that GA accident rates and fatality rates decreased by somewhere between 10-40% over that 15-year timeframe, depending on which stats you’re citing. The most recent fatal accident rate computed by Nall finally broke through the 1.0 barrier (0.99) in 2013.
- There are no published accident data that report or demonstrate that use of automation or technology by private pilots systematically causes lots of us to misuse or over-use technology, since there are no cockpit voice recorders or flight data recorders installed in most light aircraft to collect such data. So when writers and ranters talk about private pilots over- or mis-using cockpit technology in light aircraft, they are only repeating their own biases and opinions, because they have no data to back up their rants. Hence they default to citing a handful of airline accidents. In fact, given the steadily and substantially decreasing accidents in GA during the same period in which advanced technology and automation began to enter the light aircraft fleet in large numbers, the data seems to suggest that just the opposite phenomenon may be going on.
Have advanced avionics reduced fatal accidents?
If anything, review of individual GA accidents in the NTSB database and the Nall Reports seem to indicate that rather frequently, pilots who don’t use the technology that is available to them tend to die needlessly. A famous example of that scenario was John F. Kennedy Jr. who flew himself, his (at the time) advanced technology, autopilot-equipped Piper PA 32R, and his unfortunate wife and sister-in-law into the sea during a hazy summer evening, when simply engaging his autopilot almost certainly would have saved the day, and three lives.
Many more examples of under-use of technology come from the accident records – and the response to same – of Cirrus SR 22 owners who declined to avail themselves of both the advanced automation features of their glass panel aircraft and/or of the airframe CAPS parachutes. Both the airframe manufacturer and the Cirrus type owners’ association worked together to study the problem, and as a result created and promoted Cirrus type training to encourage familiarity with the aircraft’s systems and capabilities, and the correct use of the CAPS parachute (“Pull early, and pull often!”). As a result, the fatal accident rate in that type dropped dramatically in only a few years. It turns out that advanced technology really DOES save lives – if the pilot is prepared to use it.
Now, for those who would argue counter to this point that some pilots actually do crash their airplanes because they were playing with the knobs and buttons instead of flying the airplane, or through mental atrophy somehow forgot how to fly the airplane, then I would not argue that such accidents never happen. Of course such accidents almost certainly do happen. But the key question here is: How many such accidents happen each year due to excessive reliance on advanced technology, as compared to the number of accidents that are avoided (most of which are never reported to the government) because the pilot had and used advanced aviation technology?
That is, unfortunately, an unknowable answer, given that almost no light aircraft have cockpit voice recorders and flight data recorders. So the only real data that we have to go with are the bottom line accident frequency data. And those numbers suggest a pretty clear correlation, if not causality, between advancing technology (and training to use it) and contracting accident rates.
Where do you draw the line? After all, “Human flight is advanced technology.”
Back to the notion of using the technology we have, and not just relying on it being in or on the aircraft, the same principle also applies to basic aircraft technology too, not just the advanced stuff. For example, your airplane (with one or two exceptions) has rudder pedals as part of its flight control system, but it’s amazing how many pilots forget they’re there. All too often, pilots either don’t use the rudder at all, or they try to use the rudder for a purpose that is incorrect, such as attempting to apply excessive rudder to make the infamous crosswind, skidding base to final turn, instead of using coordinated ailerons and rudder… the result of which is all too often an unfortunate spin into the dirt from low altitude. Or when a pilot, in the middle of a stall at a high angle of attack, tries to use aggressive aileron deflection to control roll instead of the rudder, and inadvertently puts their aircraft into a spin.
Actually, when it comes to technology, there isn’t any part of an airplane that doesn’t constitute “technology,” since none of us were born with wings, feathers, and tails. So those pilot writers ranting and warning against “over-reliance on technology” are actually arguing counter to the very notion of human flight itself. Human flight is advanced technology.
The “tech is a trap” crowd and their claims that “too much technology is making us pilots into useless blobs of unthinking flesh“ virtually always cite the same three or four commercial airline accidents to make their points. But of course we all know that commercial airline flying is very little like private aviation flying in light aircraft.
For one thing, by rule commercial airliners today require two pilots in the cockpit, thus providing at least redundant “human technology” to aid the pilot in command. Also, the airlines employ and enforce detailed standard operating procedures that must be submitted to the FAA that dictate the use of autopilots and other technology most of the time on most flights. For instance, most pilots of the “big iron” are required to be on autopilot for all operations in the RVSM flight levels (FL300 and above), which is where most airline flights spend most of their time. Also, airline SOPs often require that autopilots and autothrottles be used for most approaches to landing. And of course, by both FAR and airline SOPs, only autopilot-guided approaches are allowed on Cat III ILS landings. So the natural result of these rules is that most airline pilots get relatively little hand-flying experience compared to most light aircraft pilots.
Additionally, it’s also difficult to get in much real stick and rudder flying on the real transport aircraft since most of the time the airlines want paying passengers on board who for obvious reasons would not want to be subjected to steep turns, the stall series, touch and goes, short field landings, etc. that we private pilots in light aircraft are free to do as we choose (and which the smart ones regularly practice in order to keep up our piloting skills).
Was the Colgan crash about automation or bad piloting?
Yet the tech-is-a-trap Cassandras generally ignore the actual facts and circumstances of the airline accidents which do not support their theories of “over automation” as applicable to private pilots in light aircraft. A frequently mentioned accident for this topic of automation is Colgan Flight 3407 (cited by Machado in his AOPA piece), in the context of the pilot in command making an error with respect to how he controlled the aircraft as it approached a stall during an approach to landing. However, it also was revealed during the accident investigation that the accident pilot had a rather blemished record as a professional pilot, including multiple failed check flights, which indicates that there was likely an issue with his piloting competency.
I don’t really care how much automation is on a plane I’m riding as passenger, but if I were aware that a pilot like him was at the controls… then I’m not boarding! The NTSB cited several principal contributing factors for the Colgan accident, but contrary to common sense, the bureaucrats seemed to pin the blame mostly on pilot fatigue and lack of pilot training, which is a way of pointing the finger at the airline rather than at the individual pilot in command. It may sound harsh to say this, but though the PIC may be dead now, so are his passengers and first officer… his own death doesn’t absolve him of being a bad pilot.
Another prototypical accident cited by the “over-automation/over technology” ranters – again, this is a commercial flight – is Asiana 214 (cited by Perry in his AOPA piece), whose flight deck crew, including a check pilot, between them could not land a perfectly functional airplane on a CAVU VFR day using a perfectly functioning lighted glideslope system when the ILS was temporarily out of service. The NTSB pointed the finger at over-automation, but I think they have it bass-ackwards. If you can’t fly the airplane at all – because, if you can’t land in CAVU daytime VFR, then you’re not relying too much on automation, you’re relying entirely on automation – then you have no business being allowed anywhere near the cockpit, in my humble opinion.
Another prototypical commercial air accident often cited in the anti-tech screeds is Air France 447, wherein an apparently incompetent flight deck crew could not handle a temporary loss of airspeed indication, and promptly stalled their perfectly airworthy aircraft all the way down from FL370 to the sea, with the pilot flying holding his sidestick all the way aft to the stops throughout nearly the entire accident sequence. What kind of over-reliance on tech causes any trained and certified pilot to believe that continuously holding the stick all the way aft to the stops is EVER a good flight technique, except in a full-stall landing from a couple feet above the runway?
What the tech-is-a-trap writers ignore is that the relatively inexperienced Air France first officer and pilot flying, prior to the accident sequence, had attempted repeatedly to obtain the permission of his captain to climb the aircraft to get above a line of thunderstorms painted ahead on the cockpit radar display as the flight entered the tropical convergence zone. The captain repeatedly denied his co-pilot’s requests to climb until he finally retired to the sleeping cabin, while an underqualified (non-captain) relief pilot took his left seat.
The Air France crew seemed unable to control the Airbus after the autopilot quit.
Then, moments later, when icing in the clouds temporarily disabled the airspeed indicators, which in turn disengaged the autopilot, the pilot flying in the right seat almost immediately began his desired escape to higher altitude by yanking on his stick. In short order, he inevitably stalled the aircraft, and kept it there in a steep nose-up attitude nearly the entire trip down to the ocean surface. If the PF had simply kept his hand off the sidestick, the accident sequence would never have begun.
That was not an example of over-reliance on automation – that was, rather, an apparent case of fear-inspired temporary insanity, and a case of very bad piloting, done contrary to any and all pilot instruction ever given. It appears that the PF’s actions were driven by one man’s extreme fear of thunderstorms. Being afraid of thunderstorms is, of course, a healthy fear for any pilot… but not so to the extent that it causes a pilot to completely ignore the laws of aerodynamics, with fatal consequences for all souls on board.
So basically, this business of confusing apparent instances of piloting incompetence with over-reliance on technology, as a practical warning to us light aircraft pilots, is simply not useful at all. This line of thinking and argument combines scenarios that are not alike and not at all relevant to our flying, and misses the important lessons for all. The end game of this argument actually creates a detrimental result in terms of attitudes toward flight safety.
The effect of these constant warnings against over-reliance on aviation advanced technology is to perversely discourage pilots from taking advantage of the real advances in aircraft cockpit technology that are proven to improve our safety of flight. That is, if they do not discourage private flying altogether. The logical inference of all these “technology is a trap” warnings is, well…
“What the heck? If no matter what we do to try to make private flying safer, it is not accomplishing anything useful, then to heck with it! If I care about saving my own skin, maybe I should stop flying altogether, or at least, maybe I need to avoid spending any more money to upgrade my aircraft.”
Advanced technology means more than just autopilots.
That is NOT the message that we private pilots need to take from the actual real world data. The real lessons for us light aircraft pilots are as follows:
- Advanced technology, both preflight and in the cockpit, is making flying significantly safer, especially for us private pilots who mostly fly single pilot, with no competent co-pilot to help us collect and analyze flight information, help us make better decisions, and help take the mental load off the PIC.
- Advanced aviation technology is not just autopilots or autothrottles connected to GPS navigators. Advanced technology is the full gamut of technology including pre-flight planning software and hardware; autopilots; GPS nav units, both panel mount and portable; ADS-B in and out with weather and traffic; satellite cockpit weather; radar; fuel, terrain, altitude, and traffic alerters; satellite position beacons; engine monitor systems; synthetic terrain vision systems; fail-safe “sunny side up” attitude recovery systems and “auto land” and “auto descend” systems for incapacitated pilots; airframe parachutes; backup attitude instruments; and a whole host of other safety related and flight management stuff, some of which we haven’t even imagined just yet. Most of these technology-based pilot aids are designed to help make up for the fact that most of us don’t have co-pilots and their extra set of eyes, ears, brains, hands and feet to supplement our “single point failure” human systems. Some of these gizmos also greatly improve our situational awareness. And the portable electronics are so cheap compared to certified avionics that redundancy of key flight systems is practically available to virtually all pilots today. In the air, redundancy is our friend.
- Some of these techno gizmos are very pricey additions to light aircraft, while others are relatively inexpensive and easy to implement. So not all of these technologies will be equally distributed through the fleet. We aircraft owners and pilots will need to make our own personal calculations of value added vs. cost for each gizmo, and decide accordingly which to employ. The decisions will often be based upon what kind of flying we do, and our available budget, and also the type of aircraft we fly. However, just about anyone who can afford to fly an airplane can also afford to buy a tablet computer and the apps necessary for moving map GPS displays, and most recently, Bluetooth-connected ADS-B in traffic and weather displays, and even backup AHRS and synthetic vision displays. A single axis digital autopilot is not cheap, but at around $10-15K installed and coupled with a GPS and/or navcom, it can literally be a lifesaver and a routine convenience to those of us flying single pilot. Having a portable GPS at hand is a popular addition even to old Stinsons and Stearmans if you ever leave the traffic pattern at the local airport. A 406 Mhz or other satellite personal locator beacon can be a real lifesaver in the event of an off-airport forced landing, because surviving the landing itself doesn’t necessarily save your life in challenging environmental conditions or when you urgently need medical care.
- The more techno aids we single pilots have available to help out in pre-flight planning and on the flight deck, the more likely we are to avoid an accident. That’s because we pilots are single point failure human systems, and as such we are entirely fallible and subject to making dozens of errors per hour on the ground and in the air. Making errors is part of being a human. Pretending that we are all Super Men and Super Women who never need a hand in challenging times is mere puffery and the ego talking, not common sense. If we always had a qualified co-pilot to help out, and if we diligently used well-defined crew resource management techniques, then the need for high tech in light aircraft flying would be less (but not zero), but alas, most of us don’t. Even if we have a qualified co-pilot, he or she is equally subject to making errors. There is no such thing as having too many checks, reminders, and backup systems once we’re in the air.
- Despite all of the advantages that advanced flight technology offers, there is still no substitute for having and continuously maintaining basic pilot skills, no matter how much technology we have available to assist us. If you can’t safely take off or land the airplane, or if you can’t perform basic flight maneuvers, then you better get some flight instruction and start practicing ASAP. If you don’t want to do that, then stop pretending you’re a competent pilot.
Is it a distraction, or a safety enhancement?
In closing, let me say that aviation writers like Machado and Perry and others who’ve written on this topic are only trying to help pilots avoid letting their piloting skills lapse, and those intentions are just fine as far as they go. But their arguments are theoretical rather than data-based, and misdirected to private pilots in light aircraft. Additionally, their arguments seem to avoid the logical conclusion, which is, “Learn to fly well the aircraft you fly, however it is equipped,” and instead seem to suggest in the main that technology is a trap.
Effectively discouraging the adoption and use of safety-related advanced technology by light aircraft pilots is, unfortunately, an artifact of their truncated line of argument. And it also tends to feed the meme that we see over and over again in our culture these days, which is to always blame others – be they companies, the government, anyone with deep pockets, including technology itself – for our own failings. “Hey, the technology made me lazy, so it’s not really my fault.” As Machado wrote, “Generally speaking, technology either liberates us, or enslaves us by increasing our dependence on its use.”
Sorry, the buck stops here with us pilots. Technology doesn’t make us do anything, or enslave us. When we sign up to be aircraft pilots in command, we elect to use technology. We become solely responsible for flying the airplane, and for the safety of the flights that we conduct.
If we have a lot of automation available in our airplane – while acknowledging that most of us flying light aircraft don’t – then it is our responsibility to maintain our ability to fly the aircraft both with and without the advanced technology. It is our fault alone if we allow our flying skills to atrophy through lack of practice. Just as it is our fault alone if we don’t bother to learn how to use the technology available in our cockpits to make our flying safer.
- Automation, technology and flying light aircraft - November 11, 2015
- Tragedy on the beach at Venice – what would you do? - August 1, 2014
Great piece! I was taught that each aircraft was designed for its ‘mission’. It may be fine to fly with a six pack and steam guages if you wish to stay local and do some sight seeing on a CAVU day. But flying long distance cross country flights IFR with passengers on board… I feel that the ‘advanced technology’ in my SR22 reduces the risk and adds an extra margin of safety for myself and passengers.
I would point to the recent chute pull by the ex-CEO of Walmart… He and his passengers walked away and are alive because of that ‘advanced technology’.
The arguement between the flat earthers and the world is round folk has gone on long enough… I’ve given up trying to bring those to the ‘light’.
On advice… Ben Franklin wrote
‘Wise men don’t need it and fools won’t heed it.’
Again thanks for the great read!
This reminds a lot of the twin vs. single safety debate. As Dick Collins pointed out years ago, the fatal accident rate following an engine failure in light twins was (still is?) about twice that of singles. He laid this on the doorsteps of the poor performance of light twins on one engine and (and this is the important part) the lack of proficiency on the part of GA pilots while flying those aircraft on one engine.The point here is that if you don’t maintain proficiency on both your plane and your avionics, sooner or later something is going to turn around and bite you somewhere uncomfortable.
But let’s look at it from a different point of view. If many pilots aren’t willing to put in the effort (or can’t afford the time) to keep up to snuff on their automation, maybe they shouldn’t be flying an airplane with all those bells and whistles, just like they shouldn’t be flying a twin if they don’t have the resources to invest in keeping their skills up to par so that losing an engine doesn’t make turn them into a statistic.
Thank you Thank you Thank you!
It’s time people realize that the airline safty record is heavily dependent on automation. I’m willing to bet that we can decrease accidents by 30% or more if auto pilots were cheaper and used more often, if VFR pilots can use thier coupled A/P to auto shoot an ILS approach as a get out of jail card, and if envelop protection saved people from doing stupid things. The list is long, and I refuse to be amazed that in freaking 2015 I have a moving map in the cockpit … Come on!!
As a student pilot, yes, still! (#GX-0940253,) I want to close the gap on the gaining the knowledge and experience from many flight hours from those that have been there. Steam gauges are part of the coordination needed to know what the numbers look like, with the other part being the memory of how it looks and feels combining the entire essence of flying. Glass, and all that goes with it, is the now and the future of aviation, in my view. I want anything and everything available to me to use for safe flight operation, as it won’t interrupt the pleasure of being aloft and making those decisions. We need the constant problem solving and achievements as part of our growth potential and as a necessity to survive as pilots.
Kindest Regards,
Chris Gabriel
Arlington, VA.
wow- using data to make your point – a novel and welcome approach (are you listening AOPA writers…..).
Duane:
Thanks for authoring a thoughtful “counter-point” piece. Nice job, and a lot of work.
A prominent factor in many/most accidents is a dearth of situational awareness. For all of its virtue and for all of its faults, there can be little doubt that modern avionics provide an order-of-magnitude increase in at least positional awareness (“Where the Hell am I?” being an important component of situational awareness). Time was, pilots needed to devote a fair amount of attention and brainpower to determining and tracking their position in the universe. No more. While one man’s expectation might be that freeing up that brainpower would provide a reserve of same, to devote to “higher-level thinking,” the reality seems to be that the resultant lack of required effort and concomitant diligence causes a more widespread reduction in alertness and increase in cognitive reaction time.
Which brings us to the widespread presumption that enhanced situational awareness will improve safety. It will… IF… the pilot knows what to do and how to do it, in the situation about which s/he now is so well-informed. Duh.
Over on AvWeb, I recently suggested a 2-question Reader Poll:
1. Have you ever unintentionally stalled an airplane?
2. Have you ever unintentionally spun an airplane?
Unfortunately, their chosen wording was “Have you ever unintentionally lost control of an airplane?” With apologies to their fine staff, I’m very unclear how one can intentionally “lose control.” The point of my suggested poll was this: I really don’t know how common unintentional stalls and spins are. I do know this: absent structural failure or flight through a tornado, an unintentional stall or spin is prima facie evidence of a lack of situational awareness.
The FAA’s recent love affair with AngleOf Attack instrumentation is a well-intentioned effort to improve situational awareness. I fear the presence of a six-lane highway to Hell…
Situational awareness requires a diversity of focus. To any extent that technology as implemented compromises that diversity, it’s being implemented poorly.
For generations, steam gauge pilots were taught (well, implored at least) to “Scan, scan, scan. Do NOT stare at ANY one instrument.” In VMC, that scan is supposed to include the full-color, hi-res display that’s available in the windscreen. And yet, that little TV screen is so damned fascinating – it’s hard to take your eyes off of it.
Anybody here know what “target fixation” means?
I praise glass for all of the benefits it can bestow – chief among them being amazing positional awareness.
I criticize glass for its many counterproductive “cute tricks” like translating-tape displays of airspeed – a VERY bad man-machine-interface format. Now comes AoA, and I have yet to see a properly-done glass presentation of that information. Not even one. And it would be easy to do. But it would be “non-standard,” so don’t hold your breath…
For those of you who are about to drop off of the front edge of your seats, here it is:
Abandon the translating-tape airspeed display. Restore the rotating-pointer airspeed display, which already provides three types of information at a glance: absolute airspeed; relative airspeed (value within the entire range); and rate-of-change. Add a tri-color AoA arc, positioned circumferentially and “outside” of the normal airspeed arcs. Here’s the MMI magic: Use the existing airspeed rotating needle to point at a MOVING AoA arc display. The three AoA arc segments would expand and contract as necessary, in real time ( can you say “coffin corner?”). The one display now would provide the three types of airspeed informanton PLUS AoA and load factor! The relationship between airspeed and AoA would be immediately apparent under ALL flight conditions. While we’re at it, we can move the redlines, bluellines, white and yellow arcs, and adjustable bugs in real time, too. Again, Duh!
Technology can be our friend. But until we become passengers in fully-autonomous airplanes (no flight controls for us!), it’s up to us to employ technology productively. Some of that burden rightly falls on the manufacturers. But the life-or-death element of that burden is borne by US.
It’s often said that babies and money don’t come with instructions…
Yars – thanks much for your comments. You are always one of the most thoughtful and insightful commenters on this site.
Regarding your question of what the pilot actually does with the reduced brain load provided by the newer techno gizmos in the cockpit, that’s a good point. A smart pilot uses the freed-up brain cells to focus on things like flying the airplane, looking for traffic, and thinking ahead of the aircraft instead of falling behind. Not so smart pilots, well, who knows? You can lead a horse to water …
I completely agree with you on the brain-killing airspeed tape display on current glass panels. And for the very same reason that I still use a round dial quartz wristwatch with rotating hands today, despite the fact that the early digital quartz LED watches of 40 years ago supposedly rendered obsolete such obviously medieval analog data representations. I suppose that the human factors folks have an explanation for that. I just know how my own brain responds.
As for AOA indicators, the FAA and some of its proponents are wrong if they think that another data display in the cockpit will prevent pilots from losing situational awareness … it’s just another display that’s easy enough to ignore, just as pilots of stalled aircraft tend to ignore the airspeed indicator.
My own opinion on AOA indicators is that their principal value is in calibrating the seats of our pants much more directly and precisely than airspeed indicators can possibly do. In other words, its principal value is as a training and practice aid, but in real time it’s your butt that ought to tell you when the stall is approaching. Pay attention to your butt!
The problems with airspeed indicators are two fold: (1) the supposed “stall airspeed” varies greatly with changes in bank angle, gross weight, and elevator back pressure; and (2) at low airspeeds the pitot-sensed airspeed is at its least accurate and least stable readout, so therefore when you need it most, the airspeed indicator is the least useful as a performance indicator. AOA indicators, on the other hand, are accurate and precise throughout their calibrated range.
Like your article Mr. Collins, but I almost always do. I retired from a major after 30+ years off B777s. In case of the 777 I don’t think there is one person on earth that knows all the FMS can do, so learn what you need like the back of your hand. Then learn as much of the nice to know as you like.
I cannot fore the life of me understand the SFO accident. I would think most student pilots would figure out something is wrong when the deck angle is so high you cannot see forward. As for Air France…. (I have 7 transport type ratings, that and 5 bucks might get you a small coffe at SB’s) I am not aware of a transport airplane that when all the chips are down, all the air data is missing just go to 5° up deck angle and max power, wind your watch, scratch your a___ ask the other pilot to look up power setting for you weight. That will save your hinny. Now your are going to be a very busy crew but at least everyone is still alive. Over my life time my thinking is 9 accidents out of 10 I say, “but for the grace of God there go I” the tenth is what in the world were they thinking?
Nice article but unfortunately all the comments about the airline accidents Asiana, Colgan and particularly Air France are not correct, I recommend you to read the Air France accident official report is available online.
Leonardo – I can assure you that I read the BEA report on the AF447 accident report, and the relevant NTSB reports, as well as numerous other analyses by private parties in the media. The comments I make above are derived directly from my study of the facts as presented by BEA, much of which were derived from analysis of the cockpit voice recorder and flight data recorder.
Different parties with different interests can review the same data and information and come to their own conclusions, often based upon their own parochial interests. Some prefer to blame the airline for training deficiencies, some prefer to blame the airframe manufacturer for the human factors design decisions made, some prefer to blame “technology” in general, and some prefer to blame the pilots. I come from the “blame the pilots” sector, because it was pilot error that caused the crash, just as I described above. The accident was initiated not by the aircraft or by the weather, but by the incompetent reactions to same by the pilot flying. The PF had no effective senior supervision in the cockpit until near the end of their plunge to the sea, when the captain finally reentered the cockpit. By that point in the accident sequence, it was nearly too late for the captain to assess the situation and intervene, although he appears from the DVR data to have finally realized near the end that the aircraft was in a deep stall.
You are welcome to submit your own detailed comments or a complete essay to the editors of Air Facts Journal – they always welcome reader submissions and they like a good debate!
Thank you.
Duane, excellent discussion; as always, we need to remember the term, “pilot in command.”
I earned my instrument rating in a 1966 C-172 with a non-standard panel and no autopilot. After flying IFR for a few years, some busy experiences in turbulence convinced me to agree with Richard Collins that single-pilot IFR without an autopilot is a losing game. If I’d had today’s panel technology, I believe I would have continued flying IFR, just more safely.
Dear Duane
Thanks for your response but you oversimplify and in my opinion in the wrong way what you say about the AF accident the pilots had a double ADR failure not just a simple erroneous speed indication error with that failure I can tell you the A330 becomes a very confusing airplane you get multiple alarms and confusing indications the PF pulled the stick because he thought was going to overspeed not to go over the clouds and when the PM on the left seat was not a Captain was a pilot with 9.000 hours and good experience on the type and anyways he doesn’t understands either what is going on.
This accident happens because they didn’t understand the indications and in my opinion because the trainning on the Airbus for this kind of situations is not the best, actually there was no stall trainning in Airbus until this accident as it is supposed you can’t stall an Airbus (in normal condition) and a situation like the one of the AF is not usually trained on the simulator.
Colgan Air is in some aspects is similar to AF the pilot didn’t understand the change in mode on his speed indication and react incorrectly in both cases Colgan Air and AF pilots are not trying to correct a stall as in their minds they were not stalling they were reacting to indications they didn’t understand and in consequence their commands were wrong.
After the AF accident unreliable airspeed, 2 and 3 ADR failures have become a favourite of the instructors in every simulator recurrent and even when you know you are in a sim I can tell you are confusing failures that you have to be very careful in first be able to identify what is going and second apply the proper procedure, AF pilots did a mistake that costed their lives but is not as simple as they have a speed indicator failure and pulled the stick.
I love technology but without the proper training can become very quick your main enemy, and of course we should use it as much as possible in general aviation.
Good weekend!
Leonardo
Leonardo – thanks for your detailed comment.
I believe it is well established in the accident record that the pilot flying (PF) expressly stated that he desired to climb the aircraft, as it is in the CVR transcripts that he made repeated requests to his captain to do just that as the aircraft approached the line of thunderstorms visible on the radar display, as the flight entered the notorious tropical convergence zone.
It was only after the captain retired to the sleeping cabin, and when the aircraft entered the area of thunderstorms, that the pitot tubes iced up causing the loss of airspeed indication and as a result caused the autopilot to disengage. The relief pilot in the left seat, apparently not aware of the obvious growing anxiety of the PF concerning the thunderstorms, was likely not clued into what his PF actually did with the stick, which was to yank it aft to the stops and hold it there for most of the accident sequence.
I’m sorry, but you appear to be grasping at straws with your explanation of “worried about overspeed”.
It is entirely implausible that the PF was trying to avoid an overspeed, which of course never actually occurred. There never was any cockpit indication of overspeed. In fact, the PF didn’t need to do anything but monitor the aircraft performance via the artificial horizon, and the GPS altitude and groundspeed indicators on the MFD, and leave the damned stick alone.
But no, he yanked the sidestick back, climbed the aircraft to about FL370 where it stalled and fell back down nose up, and throughout the entire accident sequence, the actual condition of flight was gross underspeed, the aircraft fully and deeply stalled, with a steep nose up deck angle, all the way to the ocean surface where it pancaked in.
The deck angle was recorded throughout the accident on the FDR, and for most of the accident sequence, the deck angle ranged between 20 degrees nose up and 40 degrees nose up! How could these pilots miss that fact, other than they allowed sheer panic to override every bit of their pilot training and actual flight experience to date?
Certainly every pilot on that aircraft was aware that, operating at FL350 it was very near its maximum cruising altitude for the early phase of the flight with little fuel having burned off. Which is why the captain refused the PF’s multiple requests to climb the aircraft. The pilots would also be aware that any drastic and sustained yank back on the stick at that altitude would quickly put it into a deep stall as it entered the “coffin corner” of its flight envelope. Yet that’s exactly what the PF did.
It’s also true from the data recordings that the left seat pilot at one point in the sequence recognized that the PF had stalled the aircraft, and he tried to lower the nose with his sidestick, but the control priority of the Airbus did not allow the left stick to override the right stick.
Among the human factors design factors in this accident, the Airbus model they flew does not provide a mechanical linkage between the two side sticks as they operate by wire and completely independently. Still, just looking at his own artificial horizon, and also looking to his right at what the PF was doing with the stick, should have alerted the relief pilot that the PF was stalling the aircraft. A left seat pilot in that situation would have been wise to have cold-cocked the right seat pilot and taken over sole control of the aircraft, but unfortunately, by that point in the sequence and altitude it was virtually too late to recover even if he had.
Yes, I imagine that the cockpit of the Airbus, or any other modern airliner, would be a potentially confusing and noisy environment with stall warnings and autopilot disengage warnings sounding … but that is what recurrent training in the simulator is supposed to prepare the flight deck crew to handle with ease.
I never flew an Airbus or any other airliner, but I used to be a nuclear reactor operator on a US Navy fast attack submarine. We drilled constantly as various ship and power plant emergencies were simulated with alarms going off, fellow operators keeping each other informed of their response actions, and the engineering officer of the deck quietly issuing commands to do this or do that, and yes, it can seem hectic with alarms and lights going off … especially when it was the real deal and not a drill. But that is precisely why we drilled incessantly, to prepare us for chaos and unexpected instrument readings and multiple alarms going off at the same time. So when the real emergencies happened, we just did what we knew was necessary and the results were always good – which is I’m still here!
Despite any and all other contributing factors, there was never any logical justification for the PF to do what he did with the stick. As another experienced 777 pilot commented elsewhere in this thread, in the event of a loss of airspeed indication and automatic autopilot disengagement, there is a logical and safe way to fly the aircraft: add power and establish a safe deck angle (five degrees was suggested) while the flight deck crew sorts out what is going on. It’s called, as all pilots should know, “first thing in any emergency, is you fly the airplane!”
It’s most likely that 999 out of 1,000 flight deck crews, of 9,999 out of 10,000 flight deck crews, would do exactly that. But in the peculiar situation of an obviously panicked right seat pilot flying, and a captain who, if he’d been on the flight deck and who well understood both his aircraft and the mentality of his co-pilot, wasn’t there … and a left seat pilot who was confused about what the PF was doing with the flight controls, this crew wasn’t competent to handle what was in fact NOT AN EMERGENCY but merely an abnormal flight instrumentation and controls situation, with ample backup indications available by which the aircraft could be flown safely by a competent flight deck crew.
Good article, and good follow up
Every airline I have flown for drilled it into us that there is a proper procedure for taking command.
Taking over, handing over and I have control.
Its true that there is no annunciator on the Airbuse to show which side stick has control, so this action is vital.
A330 pilot mates tell me that the problem also relates to the computer downgrading to alternate law, which removes the anti stall protection, a situation which is not adequately explained nor followed up in the sim.
Further more the younger new breed of pilots train without spin and stall practice in suitable aircraft and most pilots I flew with were paranoid with the thought odf spins and stalls.
Thanks Ted.
It’s interesting that in recent years it seems that many of the ATP pilots particularly on non-American airlines, seem to come up mainly through professional flight schools … wherein they graduate very early to right seat jobs in the airlines after only minimal experience in light aircraft. Historically, most US airline pilots seemed to come up through the ranks either via extended experience as CFIs, or flying light commercial aircraft (package transport, business charters, cropdusting, etc.), or from the ranks of former military pilots – all of whom had substantial experience with maneuvering flight before joining the airlines.
The recent US change in law requiring 1,500 hours before an ATP can serve as a right seat pilot on a US airline seems designed to at least somewhat address that concern, although a lot of pilots seemed to object to that change as a “knee jerk reaction” that was actually not relevant to the Colgan accident (both accident pilots had logged much more than the new minimum 1,500 hour logged flight requirement).
It does not seem readily explainable that the flight deck crew on AF447 seemed so ill prepared to practice effective CRM, and to clearly establish control of the aircraft, given that, as you wrote, virtually all airline pilots are schooled and drilled on that concept. Again, this may be more a reflection of a general panic on that particular flight deck, exacerbated by the violence of the thunderstorms through which the aircraft was flying, and the loud audible cockpit alarms, and quite likely a general state of fear by the PF which seems well established by the CVR transcripts.
As for some airline pilots being afraid of maneuvering flight including stalls, spins, and recovery from abnormal attitudes, this seems to be an issue for both pilots of light aircraft as well as professional airline pilots. I’ve known a number of private pilots who simply don’t/won’t ever practice maneuvering flight or stalls, steep turns, etc. except during required BFRs. If all one ever does is fly straight and level from point A to point B, then it’s hard to see how one can be properly prepared for unusual circumstances or recovery from abnormal flight attitudes.
Dear Duane
I have 6.000 hours as Captain on Airbus types and 6.000 more in other airliners so I guess I am more or less qualified to talk about why the pilots reacted so wrong in the AF accident.
The reason to disconnect the autopilot is not due to the speed is due to the double failure of the ADRs to explain you how all the indications and protections on the Airbus works will be too long plus how many times pilots react quick to avoid QAR or FOQA records (an overspeed for example), any pilot on the Airbus can override the side stick of the opposite side you only need to press a button which is the only button you have on the side stick and you immediately get the priority there is nothing wrong on the Airbus design about this (there are other dangerous things on the Airbus design but not this).
As in most accidents involving unreliable speed indications the pilots react wrong because they don’t realize they have an unreliable indication they still trust what they see, the PF in the AF at that altitude and after the disconnection and initial climb the VMAX red barber poll will increase and the natural reaction will be to pull even more to avoid go to over speed, why? because he doesn’t understands that his speed indication is wrong he is not trying to climb above the clouds in this particular moment.
There are many famous accidents related to unreliable airspeed indications one not so famous was a DC9 from Austral airlines that crashed in Uruguay again pitots with ice and they were descending so the speed indication started to show less than the real one, the highly experienced copilot on the DC9 (analog no glass) decided the speed was too low and extended the slats…when his speed was like 320 KTS…they couldn’t recover even a body part intact…again the pilots reaction is incorrect because they don’t recognize the erroneous indication…
Colgan Air is similar there is a change in mode on the speed indication that the pilot was not aware so again he corrects wrong and stalls without understand he is in a stall….
Asiana well the main problem in that accident is cultural and is a well known problem in the Korean airlines about CRM and how the Captain and in this case the instructor is god and they will crash before even open their mouths about something is wrong and is what happened, the pilot in the jump seat saw and realized what was going on but he kept his mouth close due to the cultural aspects of the Korean society (I fly in Asia).
I agree with your article in relation to your comments on General Aviation (I still fly a Cessna 170 that I keep in California to enjoy on holidays) but as many other good writers on General Aviation topics you also use as example airline accidents without have the experience on the field to get the right facts is like if I read a report about a nuclear submarine accident and just based on that then I start to state facts about safety in nuclear submarines….have a good day.
Leonardo – the debate between us is interesting as is your professional background, but you seem to be avoiding the issue of the pilot’s actual manipulation of the flight controls in this accident, which are documented in great detail by the FDR. Regardless of what one or two malfunctioning performance indicators were telling him, that does not explain why he did what he did with the stick.
You clearly are in a position to exert personal authority in this debate as an Airbus pilot with 6,000 hours in type. Yet I don’t believe that our respective piloting experiences are relevant whatsoever to assessing the grievous errors made by the AF447 flight deck crew.
After all, any primary flight student in ground school is taught what a stall is, and how to avoid it, and how to recover from it (push the stick forward to lower the nose and add power!). Every qualified pilot in the world understands that there is NEVER any approved manipulation of flight controls that calls for pulling the stick all the way aft to the stops and holding it there continuously for several minutes. It’s literally unbelievable that any competent pilot would ever think that that is the right thing to do in any aircraft, no matter what some of the flight instruments are telling him or her, or even in the face of temporary confusion.
“Think” is the key word in this instance.
It doesn’t take three and a half minutes for qualified pilots to determine that the airspeed indicator was temporarily malfunctioning, and to determine that the aircraft could be flown by the remaining panel performance indicators, including the artificial horizon, the GPS ground speed, and the GPS altitude readout. This is standard instrument pilot scan and partial panel stuff.
Actually, per the FDR data, the pitot airspeed indication eventually returned to normal as the aircraft plunged below the icing altitudes. But alas, the airspeed indication was still ignored by the crew as the PF continued to yank back on the stick.
It just makes no sense to do what he did.
However, the evident fearful panic of the pilot flying seems possibly explanitory of his actions.
Prior to the loss of airspeed, the PF was openly advocating to get his aircraft up above the thunderstorms, and was repeatedly rebuffed by his captain. That could fairly easily explain – from a psychological perspective, if not an aerodynamic perspective – his growing panic, and subsequent failure to follow his training and fly the aircraft. When people panic, we effectively lose control of our higher level conscious logical thought processes, and we just act out of instinct, regardless of logic or training. Fear is a powerful emotion, possibly the most intense emotion that we experience as human beings. Fear can easily cause us to act without thinking.
None of this had anything to do with “over-reliance on automation”. What the pilot flying did was not due to a lack of refined airman skills – what he actually did with the flight controls was far beyond the pale of what airmen do.
Dear Duane
I am enjoying very much this discussion but I still disagree on your analysis of the AF accident, and yes this accident has everything to do with your article, again I agree with the use of technology in General Aviation as much as possible, when last year I bought my Cessna 170 first thing I did was to get the highest level of subscription in Foreflight and bought a Stratus, a Bad Elf as backup and also a Spot emergency locator, plus other gadgets. I love my “Classic” 52 170 but also I embrace technology to increase the safety of my flights and I train my self in the proper use of it. I also agree with some of the comments by Rod Machado and if we are not trained properly to use the “latest” technology (basically we are talking about glass cockpit and autopilots) this technology can go from be our best friend to be our worst enemy in a minute and kill us and that is why the Air France accident is so important and NOBODY is paying attention to the lessons we are getting thanks to the death of 228 people on June 1st 2009, also you are not paying attention to their lessons and those lessons are as valid for a pilot with an Ipad with Foreflight as for a pilot in an Airbus 330, and those lessons can save your life.
The official report of the Air France accident has 223 pages I tried to resume that in 4 pages, not easy, and I hope my work is clear enough to understand why this accident really happened and no this accident didn’t happen because the pilot pull the side stick, this accident happened because the pilot didn’t understood the indications in front of him and couldn’t recognize that those indications were wrong, he was also confused by several of the systems that are designed to help him but that suddenly were working against him, he needs to ignore them to take the right decisions and he didn’t have the proper training to do that, and that also applies for General Aviation and that is how 228 people died in less than 4 minutes…no they didn’t died because he was pulling the side stick…and you would be surprised that in an Airbus and in that particular context in combination with the training we get is not a surprise for me what happened and can happen again unless we really learn the lesson, and yes has all to do with the interaction between pilot and automation.
The accident resulted from the following succession of events: Temporary inconsistency between the measured airspeeds, likely following the obstruction of the Pitot probes by ice crystals that led in particular to autopilot disconnection and a reconfiguration to alternate law.
“There are several conclusions about why this accident happen but in my opinion the main reason is (from the report):”
-The crew not identifying the approach to stall, the lack of an immediate reaction on its part and exit from the flight envelope, The crew’s failure to diagnose the stall situation and, consequently, the lack of any actions that would have made recovery possible.
-The crew not making the connection between the loss of indicated airspeeds and the appropriate procedure
“It is fundamental to understand that we don’t know to which indications the PF was reacting as the official report states that there are no records of those indications.”
Note: Only the speeds displayed on the left side (PNF) and on the ISIS (Standby instruments) are recorded on the FDR; the speed displayed on the right side is not recorded (PF).
“The following paragraphs show how confusing was the general situation in the cockpit and this kind of confusion has been common in every single accident related to unreliable speed indications in every case the pilots fail to understand the real problem and then in consequence react incorrectly”
-At 2 h 10 min 16, the PNF said “we’ve lost the speeds ” then “alternate law protections”. The PF made rapid and high amplitude roll control inputs, more or less from stop to stop. He also made a nose-up input that increased the airplane’s pitch attitude up to 11° in ten seconds. Between 2 h 10 min 18 and 2 h 10 min 25, the PNF read out the ECAM messages in a disorganized manner. The PNF said that the airplane was climbing and asked the PF several times to descend. The latter then made several nose-down inputs that resulted in a reduction in the pitch attitude and the vertical speed. The airplane was then at about 37,000 ft and continued to climb. At about 2 h 10 min 36, the speed displayed on the left side became valid again and was then 223 kt; the ISIS speed was still erroneous (means 2 out of three indications still wrong). The airplane had lost about 50 kt since the autopilot disconnection and the beginning of the climb. The speed displayed on the left side was incorrect for 29 seconds. (“Also at the beginning they got stall warnings when they were not actually in stall condition but in normal flight”)
-At 2 h 12 min 02, the PF said, “I have no more displays”, and the PNF “we have no valid indications around fifteen seconds later, the PF made pitch-down inputs. In the following moments, the angle of attack decreased, the speeds became valid again and the stall warning triggered again.
“Another confusing fact is that the ECAM (computer that shows the emergency procedures was showing 7 different procedures to deal with being the real problem that was creating all the failures the last of them…”
“Dear Duane you state that most pilots will react in the right way to this kind of situation, well as you like facts here we go also in the official report:”
-Analysis of 13 recent incidents related to unreliable airspeed indications showed:
The variations in altitude were contained within about one thousand feet. There were five cases of deliberate descent, including one of 3,500 feet. These descents followed a stall warning; Four crews did not identify the unreliable airspeed situation (30%): in two cases, the crews concluded that there was an inconsistency between the angles of attack; in the two other cases, the crew considered that the speeds were erroneous rather than unreliable.
-Case of TAM flight on 12 November 2003 This case, which happened to an A330-200, was not one of the thirteen events studied above because no crew report was available. However, in the light of the data from flight AF 447, it seems useful to mention it. In fact, following icing of at least two Pitot probes at FL360, the crew made some high amplitude flight control inputs (to the stop), sometimes simultaneously. When the AP disengaged, both pilots made pitch-up inputs (one went to the stop) that resulted in an increase in pitch of 8°. On several occasions, the stall warning was triggered due to the nose-up inputs, and the crew reacted with strong pitch-down inputs. During the 4 minutes that the sequence lasted, the load factor varied between 1.96 g and -0.26 g, the pitch attitude reached 13° nose-up and the angle of attack reached 10°. Altitude variations, however, were less than 600 ft.
-1.16.8.4 Statements by other crews faced with similar situations A comparative analysis of reports and statements by other crews based on seventeen events that occurred in similar conditions to those of AF447, two of which are studies in 1.16.2, brought to light the following trends: Analysis of the situation by crews appears difficult; Calling on the « unreliable airspeed » procedure was rare; Some crews mentioned the difficulty of choosing a procedure bearing in mind the situation (numerous warnings); Others did not see the usefulness of applying this procedure given that in the absence of doubt about the unreliability of the airspeeds, their interpretation of the title of the “unreliable airspeed “ procedure did not lead them to apply it; Some gave priority to controlling the pitch attitude and thrust before doing anything else; The triggering of the STALL warning was noticed. It was surprising and many crews tended to consider it as inconsistent.
“As we discussed before in your opinion the PF tried to climb above the clouds and the report states that the probably reason was that the PF was trying to avoid an over speed situation and also following the Flight Director (with erroneous indications), also there is a full page in the report talking about how airline pilots consider over speed a serious risk due to several reason and that they will have a fast reaction trying to avoid it”
-In the absence of airspeed information known to be reliable, it is possible that the PF thought that the airplane was in an over speed situation, notably due to his interpretations of several clues:
-The aerodynamic noise, The buffeting, that he might have interpreted as being due to high speed, The speed trend arrow on the PFD, which at that time indicated acceleration. Some of the PF’s actions may be interpreted as indicative of a perception of a risk or of a diagnosis of over speed. Firstly, the PF reduced the thrust during the seconds preceding the activation of the STALL 2 warning and the onset of buffet. Secondly, 51 s after the triggering of this warning, the PF said “I have the impression we have speed” then moved the thrust levers to the IDLE detent. He reformulated his impression a few seconds later, combined with an attempt to extend the speed brakes. Other factors which may have prompted the PF to fear an over speed situation were: The display on the ECAM (max speed 330/.82) combined with the reconfiguration to alternate law which may have been read; The fact that, in cruise, the upper red strip on the speed tape (MMO) is about ten knots above the current speed, whereas VLS is barely visible at the bottom of the tape (thirty knots less); The dangers associated with over speed situations embedded in the collective consciousness of pilots.
Finally, although the PNF had called out the reconfiguration to alternate law when reading the ECAM, and even though the indicators of the loss of protection should have been displayed on the PFD (SPD LIM and an amber cross in roll and yaw), it is possible that the PF was not fully aware of this reconfiguration and of what it implied.
-He may therefore have embraced the common belief that the airplane could not stall, and in this context a stall warning was inconsistent (“Airbus is proud of that you cannot stall their airplanes and also that double failures are almost impossible….”)
-When the Captain returns to the cockpit the two co-pilots informed him that they had lost control. The PNF stated that he did not understand the situation and that they had “tried everything”
“They were not trying to recover from a stall as they never realized they were stalling so any discussion about stall procedures or hand flying are useless is not what killed them…let’s learn the real lessons here there is a very high price tag on this lessons.”
Good sunday,
Leonardo
Leonardo – despite the numerous citations from the BEA accident report for AF447, you still have not explained two things that I have already asked you about above in this thread, re, (1) why the pilots didn’t consider the artificial horizon and the GPS groundspeed and altitude – all three of which clearly showed the aircraft was stalling, and (2) why did the PF incompetently and nearly continuously yank the stick all the way aft to the stops throughout most of the accident sequence?
I believe that you decline to respond to these points and instead shower us with other info from the accident report because you know that acknowledging these facts blow up your entire defense of the pilots.
You are grasping at straws here, in defending the actions of the pilots. I get your sympathetic position with respect to your fellow ATPs. It seems pretty common in these airline accident discussion threads for some of the ATPs commenting to defend their fellows, or to chalk it up to “unavoidable confusion” or whatever. That’s poppycock – ATPs are paid to do their jobs which is to fly the airplane safely. There is no credible defense for the failure of any of them – thankfully a very small minority of ATPs – to do their number one job in life.
You can cite complicating factors from the accident report ad infinitem, but it does not explain away the fact that a flight deck crew failed to fly a perfectly airworthy and stable aircraft, and thus drove it into the sea, killing hundreds.
All because of a temporary conflict in the flight instrumentation, easily diagnosed and overcome … and they did it by incompetent manipulation of the flight controls. Your defenses of the flight crew are direct evidence of one of my major points in this essay, i.e., that our culture and some people are far more focused on shifting blame and refusing accountability than they are in doing the jobs that they’re supposed to do.
Dear Duane
Looks like you are not reading what I wrote or the report, first there is no angle of attack indicator in the Airbus, the standby instrument is the ISIS which also showed erroneous indications is on my previous comment and is on the report, to access the GPS indication to cross check the real speed you need to check the MCDU go to two different pages to finally see the ground speed so you need to mentally calculate the TAS and consider the wind which may be also be wrong due to the ADR failures of course they could do that if they really understood what was going on which they didn’t…and as I said before even go there they had seven abnormal procedures on the ECAM…
Why he keep pulling I already answered that too as the report says is “supposed” that he was trying to avoid an overspeed indication not a stall situation, for long time they were 17 seconds missing from the CVR recently those 17 seconds become public and the PF says “Sh… I have been pulling all the time” yes he realises his mistake was too late of course.
I took the work to resume more than 200 pages and share the best of my experience of an incident I know very well if you don’t want to listen well your problem I hope some other reader will pay attention, I am not defending the pilots I am just analizing facts and due to your lack of knowledge of the Airbus you keep saying this was a simple failure and it was not, you said you read the official report I doubt it because the report is very clear and detailed…you should avoid comment about airliners as you have zero experience on them when I am also not only Captain but also TRI TRE but looks like I know nothing and you know more than me based on your experience so I recommend you to stick to General Aviation or Subsmarines (and I don’t talk about Subsmarines because I know nothing about them)….
Have a good day!
You seem determined to defend the indefensible. This discussion has nothing to do with operating submarines vs airplanes (I cited the submarine experience only to demonstrate that I know something of the value of training and drills and effective crew resource management). Sorry, you cannot assert such expertise as yours in order to shut down discussion amongst the community of pilots of a clear case of grievous pilot error and incompetence … or that absolves professional airline transport pilots of their responsibility to not crash perfectly airworthy airplanes into the ocean.
Dear Duane
You should be working for the NTSB…what to say another desktop expert.
Have a good day !
Dear Ted
The Airbus does have an indication of who is in control visual and aural you will get a “Priority Left/Right” depending who took control and a “Dual Input” in case both pilots are moving the side stick at the same time.
Have a good day
Leonardo
Great article – spot on… up until the comments back and forth regarding AF447. It seems very hypocritical to cite others’ parochial interests, yet be completely unwilling to entertain the possibility that it may have been something other than pure incompetence that caused the PF to apply full aft stick. The sheer fact that it is so inexplicable that he would make such an ‘obvious’ error is a good indication that we are missing something – or there is something more to learn here. Was it pilot error? Absolutely. Can competent pilots make errors? You betcha. But hubris is far more dangerous than incompetence. Instead of discounting this mishap as another ‘couldn’t happen to me’ scenario, we should learn from the mistakes made. Only when you accept that you could also make these sames mistakes – presented the same situation – can you truly learn from them.
I was involved in a mishap investigation involving a helicopter apparently losing tail rotor thrust approaching a landing with a significant tail wind. It turns out that the PF ‘inexplicably’ applied the opposite (and full) pedal input in response to what may have been an unexpected and uncomfortable yaw rate caused by the tail weathervaning into the wind as he turned crosswind to land. The resultant yaw rate was perceived to be non-commanded and therefore they executed the loss of tail rotor thrust emergency procedure which includes securing the engines to reduce torque. Unfortunately they did this much higher than they realized, and they crashed a perfectly good helicopter. Anecdotally, I have heard of similar incidents where the pilot inexplicable applied the opposite correct input. This is something we fear and practice in the simulator a lot, to the point where it has to be instinctive since there is really no time to think, and things get ugly quickly. But it seems the brain can fall victim to stress and fear, and sometimes we subconsciously react poorly, not incompetently.
So for the AF447 scenario – I question whether the PF was even aware that he made the input, or that he intended another input. And as Leonardo explains, the right seat may have been presented different information than what the flight recorders captured, so second guessing his response without knowing what he was seeing (and feeling) is short sighted. I’d expect there was significant spatial disorientation going on, coupled with potential previous anxiety of the storm, and they were effectively incapacitated.
I am glad to hear the Airbus community has not discounted the AF accident and now practice this scenario in the simulators – hopefully so it won’t happen again.
I haven’t read the full report, and haven’t reviewed it recently, but wasn’t there a known issue with the pitot probes being overly susceptible to icing – to the point that there was an airworthiness directive to replace them?
Actually, Dave, if you read carefully my original post and subsequent exchange with Leonardo, I made it quite clear that the AF447 accident was indeed about something much more than just garden variety piloting incompetence as we think of it, i.e., “errors in stick and rudder skills”. Indeed, my post was about how characterizing accidents like AF447 as examples of “over-reliance on automation” are gross misinterpretations of the evidence we have available.
Rather, I posited that fear and panic was the most likely explanation of the pilot flying’s otherwise completely illogical response to the conditions of that particular flight.
We all make errors, constantly, It has been scientifically proven that humans engaged in practically any kind of work or mental activity, from knitting to reading to typing to flying airplanes, make something on the order of 20+ to 40+ errors per hour. But those of us who are competent at whatever it is we are doing are able to recognize those temporary inadvertent errors and correct them as we go.
That is, when we are thinking logically and according to our training.
When people are panicked by abject fear, however, it has been well known for thousands of years that people in a fearful and panicked frame of mind quite often (but not always) don’t react logically. Many a battle has been won when one opponent did something that the other side did not expect, and as a result of the conflict between expectation and reality, caused one side to panic and run or even just give up.
In fact, in the American Revolutionary War it became “standard practice” for the unskilled revolutionary militias to drop their weapons and run after the first volley was fired in battle. The British became so accustomed to that behavior that they depended upon it. Later on in the war, in several famous battles such as Cowpens, the American field generals used that British Army expectation to fool the red coats into attacking the fleeing militias while seasoned regular Continental army units swooped around the flanks and overwhelmed the pursuing Brits.
Panic often defeats logical thinking and even ingrained training.
What is important is that pilot training, and pilot knowledge, skill and judgement, cannot be substituted by technology. I think the author Duane Truitt and the AOPA writers Rod Machado and George Perry agree on that point. In addition, more technology requires more training. We should use technology to our advantage, but not surrender the responsibility to the machines. I am being taught that our equipment can fail, and if that happens, I will have to understand the situation and use the resources available to me to fly my plane down safely.
Great article and discussion. Thanks!
Came here when searching for evidence that automation makes plane/flying safer. Could not find it. What I would really like to see is real evidence. It is all very well to say that
” In fact, given the steadily and substantially decreasing accidents in GA during the same period in which advanced technology and automation began to enter the light aircraft fleet in large numbers, the data seems to suggest that just the opposite phenomenon may be going on.”
But that as we all know “Correlation does not imply causation” so I would like to seem some actual evidence. So far I’ve drawn blank…will keep searching.