The Air Line Pilots Association, International has had quite a bit to say following the crash of Asiana Flight 214 at SFO. Among other things, the association has said: “The NTSB’s release of incomplete, out-of-context information has fueled rampant speculation about the cause of the accident.”
I don’t agree. The fact of the matter is that the airplane crashed on a beautiful day, there was apparently no mechanical failure, and the public feels entitled to all the speculation that anyone cares to offer. That is just the way things work.
From what is known, the crew just turned in a truly lousy job of flying. Some would apparently like for this not to be known for a couple of years, when the accident will be all but forgotten.
In fact, I have been surprised about two areas where there hasn’t been speculation.
Could the aircraft have run out of fuel? The lack of fire in the wings, which hold fuel and which were pretty well ripped when the main gear was knocked off by the contact with the sea wall, certainly raised the question in my mind. They have cleared the wreckage now, but nowhere have I seen an account of the amount of fuel that was recovered from the wreckage. Still, the NTSB reports normal engine operation so the lack of speculation on this is justified.
Setting that aside, could there be a strong similarity between this accident and one involving a Turkish airliner in Europe a few years ago? There, as I recall, a malfunction in the system that controls the autothrottles caused that system to think the airplane was at a height where the flare for landing begins. The autothrottles thus pulled the power back. However, the airplane was about 1,000 feet above the ground and the crew failed to notice the decaying airspeed until it was too late and the airplane crashed well short of the runway.
Here, the NTSB has said that all systems on the Asiana 777 were operating normally. So much for that.
There is an unsigned document that has been making the rounds on the web that calls into question the general ability of the pilots at both Asiana and Korean Air, that country’s other large airline.
The document that I saw was penned by an American who was an expat instructor at both airlines which were forced to hire such instructors because much of the rest of the world said they had to do that because of a poor safety record. The airframe manufacturers help craft appropriate training programs.
The author detailed how the Korean pilots were unable to do relatively simple things with the airplane, like visual approaches. A quote: “So, when I hear that a 10,000 hour Korean Captain was vectored in for a 17-mile final, and cleared for a visual approach in CAVOK weather, it raises the hair on the back of my neck.”
The document did not paint the Koreans as unintelligent. In fact, the author acknowledged that they are bright and smart. But he went on to suggest that the educational system in that country emphasizes memorization and that the kids are taught never to challenge authority. It suggests that you just can’t change 3,000 years of culture.
Over the years, I have seen similar documents originated by Americans teaching people from other countries to fly. One I remember clearly was from a pilot who taught Shah-era pilots in Iran to fly helicopters. His observation was that they had a hard time learning the complexities of a helicopter because the most complex thing they had previously operated was a donkey.
The document about Korean pilots suggested that there would be a lot more accidents like this one.
Automation also gets a poke in the eye here with a corresponding poke at pilot experience. Airline SOP calls for engagement of the autopilot soon after takeoff and disengagement right before landing. That, according to this document, means that on a 12-hour flight the pilot gets only a few minutes of real “flying time.”
I am getting a little tired of people knocking automation any time something happens. Sure, automation can screw up, but it does not do so nearly as often as humans. It also appears blameless in the Asiana 777 crash. The simple fact is that the safety record in big jets is nothing short of wonderful. A passenger does have to acknowledge, though, that when you ride in an airliner from a country not your own, the operation reflects the culture of that country, not yours.
I think that speculation is good because it gives things to think about when the event is fresh in mind and a lot is being said about it. Now it is, and should be, a number one topic of conversation.
There is a lesson here for all pilots. When on final, the important things are altitude, airspeed, sink rate and vertical and horizontal alignment with the touchdown zone. On a clear day, a pilot has only to look out front and glance at the instrument panel to get those essentials.
- From the archives: how valuable are check rides? - July 30, 2019
- From the archives: the 1968 Reading Show - July 2, 2019
- From the archives: Richard Collins goes behind the scenes at Center - June 4, 2019
I was frustrated, early on when CNN said the landing was being made “without instruments.” That was an assumption based on having no idea what the ILS is. Speculation without subject knowledge can be damaging to public perception. CNN has aviation,law enforcement, legal and medical experts. They should be consulted before commentators with little technical knowledge of a given subject begin confusing the public.
Rather.well stated. I am also an expat that taught in the Korean ab initio program and also had served as Chief Flight Instructor at Sierra Academy of Aeronautics and had a lot of exposure to pilots from all over the world. All the previous comments that you make reference to, in terms of culture and education, have been my experience as well. These are not expat egos at play here, but, simple observations of “hand’s on” instructors making the evaluations and judgements that are made of ALL pilots that they work with. No one wants to accept “profiling” these days, it is considered “politically incorrect”, however, if we let the political correctness enter into this discussion because it is one of culture, then we are doing all concerned a disservice. Pilot’s identified with psychological aversion to proper application of CRM, need to be identified and given additional “orientation” and exposure to CRM thought processes.
When CRM was intrduced in the US, the people that needed the focused attention and training were the CAPTAINS. They had to accept that input from a crew member Is not insubordination, or disrespect, and all comments were to taken in the context of Safety and the Ego had to be sent on vacation in the cockpit.
In reference to the actual “event”. When the sea wall was contacted at absolute minimum speed, an even further deceleration should have taken place, and yet, a cartwheel was accomplished, where did that energy come from? I am quite certain that just before hitting the sea wall when they called for a go around, the throttles were increased for a go around. Upon impact, no one gave any thought to the throttles and the fact that the engines were spooling up. This spool up of the engines is what then provided the energy for the cartwheel. The aircraft was stalled at the sea wall and decelerating. Had the power not been applied, this airplane would have simply skidded down the runway to a stop. With that cartwheel, it can only be stated that the folks on board were extremely fortunate that the aircraft landed right side up. Any other type of “touchdown” would most likely have resulted in very many more fatalities, if not all. When watching the cell phone video very closely and visualizing the “energy state” of the aircraft. It should have been out of energy and yet, it comes up and out of the dust to then do a complete cartwheel…a very interesting aspect to this crash.
Blaine, I agree with your analysis, but the one factor you don’t mention is that the left engine was torn off at the seawall so all thrust was from the right engine which caused the counter-clockwise spin.
Perhaps the sheer weight/kinetic energy of the airplane caused the unusual attitude after impact. It was still traveling in excess of 100 knots impacting the seawall.
One of my friends, a retired United pilot suggested the yoke may have been back (last attempt at go-around) causing a nose up attitude, hence the ‘cartwheel’.
As a student pilot I am fascinated and encouraged by all these analyses of 214 by experienced pilots.
Andy I totally agree with you. As a student pilot, one clear learning I take from this is as technology advance we can never ignore the importance of going back to basics. Training on a C172 with no glass cockpit raises the level of aviation acumen for the entire circuit as opposed to a few minutes pre and post landing interspassed with auto pilot on commercial jets. Are we going to allow our ego to block learning from juniors as ascend our career? An example of a crash in Africa where a senior Captain in command was warned by a first officer of plane’s excess turn on take off, ignored, arrogantly, and both never lived to tell save for the black box data. I desire to maintain my humility, learn from both seniors and juniors alike. Thanks for all your contribution towards aviation knowledge and safety.
This is a exactly what I speculated. After the very “low and slow” approach had the pilot flying / pilot in command or the other pilot non flying not applied GO AROUND power the outcome would have been less severe, nevertheless still embarrassing to the crew. Apparently Crew Resource Management totally failed, or I should say did not exist. Visual approaches demand more attention since you are directly feeding the inputs based on outside view and looking at your instruments. Either they were not looking outside or not looking at the six pack or BOTH.
I do not believe we can blame the autopilot or auto throttle or lack of glide slope or or or….The pilot in command must be in command of the aircraft. He or she (the crew) must safely fly the airplane with passengers on board. If any unsafe situation exists, like unstabilized approaches, a decision must be made at the appropriate time to go around, even in ten miles visibility in a B 777 and be embarrased. That is OK. But it is NOT OK to jeopardize the safety of the passengers, people on the ground, the airplane, and public private property on the ground.
I taught Cockpit Crew Management a few decades ago at a young age and I understand culture and egocentric pilot behaviors. We should allow authority of a Captain, an Examiner, a high profile individual to be questioned through CRM techniques. Examples: “Captain, speed is too low, we are unstabilized at 1000 feet AGL, First Officer decent rate is too high Go Around, check airspeed, check altitude, recommend Go Around, etc…..”
Your energy analysis is off. At a moment before impact with the concrete seawall, the still-flying aircraft had two forms of mechanical energy: speed and altitude. The impact would have slowed the aircraft very little. The cartwheel (or nearly so) was initiated by something off-symmetry on the aircraft hitting the ground first, probably the aircraft’s wing (or wing-mounted engine).
The fact the engines were applied to full power moments before the crash is almost irrelevant and are unlikely to have contributed to the cartwheel. They take several seconds to reach full power from idle, it’s highly unlikely they were applied asymmetrically, and contributed hardly any mechanical energy to the aircraft.
This FUBAR crashed would have happened almost exactly as it did if they hadn’t tried to power up at the last few seconds. This flight crew is so incompetent they should be charged with criminal negligence in the US so they can never fly here again. Sum Ting Wong, indeed.
All I can say is that an airplane worth millions of $ has been destroyed, and in the same process, people have lost their lives whilst others have been injured. the cockpit crew of this flight are there to tell what happened. There is no cause to speculate. In fact it has been reported that both the ILS & PAPI were not operational at the time…
Has the status of the Pilots been declared. Documentary about contracts which have ZERO Hours operated by Ryan Air have come in for a critique from unidentified pilots as endangering air safety ie stress related performance!
On another site, I opined that these four ATP-rated pilots collectively made errors for which I’d slap an 8-hour student pilot up the side of his/her head. My post was deleted, so I guess offering such an opinion is politically or journalistically incorrect. Or maybe it was my other comment that “autonomous aircraft are looking better every day.” Then again, I also said that it was nothing more than dumb #$%! luck that the vehicle struck the seawall where it did – 75 feet sooner (at the intersection of the wing and fuselage) and there likely would have been nothing left but fireball debris. That’s three impolite opinions in one post.
Once upon a time, we taught people how to fly. We abandoned that to some extent when we started teaching applicants how to pass the tests (written and practical). Ab-initio training made things much worse. Now we’re at the point where many ATPs apparently can’t deal with a stalled aircraft. Politically incorrect or not, I don’t call that progress.
Control systems that include human pilots in the loop are conceptually flawed. Even the best implementation of a flawed concept will itself be flawed. Autonomous vehicles will improve safety – dramatically, for GA; incrementally for air-carrier ops, where safety already is at a very high level. But as accidents in both arenas show, the weakest link in the safety chain is the human pilot. Eliminate that link, and you’ll have a much safer chain.
It’s ego-offending heresy to say it. But it needs to be said. The military already knows this. FedEx knows it, but won’t talk about it very loudly. One of my pet projects is the Cirrus Vision jet. I’ve offered them lots of advice over the years. Some they’ve heeded (although I’m sure that they came to their conclusions completely without influence from me); some they’ve not. One area of disagreement has been about putting a whole-vehicle parachute-recovery system into a light jet aircraft. They think that it’s been a Godsend with the SR-series; I won’t argue that it’s saved several lives. But particularly in a jet that will require a type rating or its equivalent, substituting an autoland capability for 400 pounds of additional fuel (my weight estimate; not theirs, which is unpublished) would address one significant shortcoming of the design: it’s short on range/payload. It also would permit the elimination of the massive windscreen center post, which is the only flaw in the Vision’s otherwise-spectacular cockpit visibility. Others have pointed out to me that the parachute can be useful in some (not all) instances of structural failure. I ask: what would cause structural failure in a light jet aircraft? Thunderstorm penetration and mid-air collisions come to mind. We already have technology sufficient to render both circumstances adequately avoidable.
Somebody’s gotta be the first. I was hoping that Cirrus would be the ones. They still may. Imagine how many baby jets they could sell if the vehicles required no human pilots. Seriously. The biggest single impediment to the growth of GA personal flying is the barrier-to-entry that is represented by having to become a pilot. At least an autonomous aircraft wouldn’t lose track of airpeed and mush into a stone wall on short final…….
“At least an autonomous aircraft wouldn’t lose track of airpeed and mush into a stone wall on short final…….”
But that’s EXACTLY what happened to Turkish Airlines at Amsterdam & maybe with Asiana at SFO. Certainly, in both cases, the crew was relying on automation which didn’t work as expected.
The aircraft you cited do not have an autonomous control system. An ACS and an autopilot/FMS bear little resemblance to each other. An ACS does not even allow for human intervention, so it can’t be misused by human pilots – because there are no pilots aboard to misuse it.
Not going to happen,for a long,long time,thank God,but keep dreaming.
Sooner then you may think (2015??). See http://www.avweb.com/news/features/fly_by_wire_diamond_da42_dries_207846-1.html
I’m old fashioned – I prefer to do my own flying – including landings. BUT to have autoland as an option – that is something that would be good to have available in the ol’ toolbag – like auto pilots in general.
I’d like to point out that the NTSB chairman did explain the fuel tanks were not compromised, and the main landing gear sheared away from the fuselage as designed. These details are available from the NTSBgov channel on YouTube.
If the fuel tanks were not fractured, I wonder how the cabin was gutted by Fire.
One offered story has been that it was flammable cargo that was ignited as a consequence of crash damage to the cargo hold.
I’ve been saying for years we need to stop duty-free alcohol and burnable material from being allowed on aircraft. I have no evidence that this was a cause in the Asiana fire, but why take on-board the unnecessary accelerant (and weight!) when you could just pick up duty-free items before exiting customs, like they do at Heathrow.
One of the photos after the crash showed the overhead oxygen masks deployed. I do not have type rating on B777, but there probably were Oxygen chemical generators in the overhead area of passengers. That is what was called on the L1011 and other airplanes. They are either manually deployed or automatically deployed. This continuous supply of oxygen right after the crash supplied the fuel. All you need is a spark. That is my take on this. The oxygen system is manually deployed if the auto system does not work during a rapid decompression at high altitude and provides enough oxygen until aircraft is safely down to 10,000 feet. My question to Boeing and NTSB is ” was the oxygen system supposed to deploy automatically in this situation or was it manually deployed, and did it contribute to the cabin fire.”
Uh, O2 is not a fuel, it’s an oxidizer. Are the chemicals in the O2 generators a fuel? Or would pure O2 cause some otherwise fire-resistant materials to burn?
Remember the Falklands War (1982) and the HMS Sheffield? Sank because its aluminum hull ignited and burned a big chunk of the ship to the water line. Add O2 and most anything to start burning and you get the temperatures required to ignite the airframe. JP4 not required.
The person in love with automation failed to note that automation, as well as all other navigation equipment, earthbound or otherwise, is conceived of and constructed by humans. Try taking the humans out of that. I believe in the end this will come down to a tired crew, new to the plane, with failing basic flying skills, who fell for the one big gotcha Boeing flight navigation system has. It can happen to anybody despite claims and it could happen to a plane with the autopilot system on.
Autonomous aircraft control systems bear little resemblance to autopilots. And they never suffer from pilot fatigue.
I can’t and don’t advocate taking humans out of the control-system design loop. I absolutely advocate taking humans out of the cockpit. The biggest “gotcha” in the big Boeing’s flight control system is that it includes real-time human actions in its design.
Perhaps in 20 years. Today, the technology is mature enough for toy quad-copters and smaller UAVs, but not for anthing sizeable and certainly not for anything carrying passengers.
I’ve seen how helicopter UAVs handle turbulence and crosswinds. You can argue that the autoland systems on full-size airliners are better, but why is it that a Cat 2/3 approach has lower crosswind limits than VMC approaches? Or, do you plan on only flying on calm days.
I fully expect to see automated cargo and perhaps passenger flights before I take my final flight west, but not today. I work with several UAV operators who talk about having to “fly them into a hillside” when they “act up.” No thanks.
Control systems are designed to mission specifications. Military UAVs are disposable. Airliners full of people are not. Technology does not stand in the way of autonomous vehicles for human PAX. Human ego, pilots’ unions, and regulations do.
Autonomous cars are coming soon, and they’re much more technically challenging than autonomous aircraft are.
Tom, please explain to me how your autonomous aircraft would have accomplished the “miracle on the Hudson”, or the safe landing of the Qantas A380 at Singapore after the engine exploded, or the hundreds of little equipment/automation failures happening every day that we never hear about because there was a human crew onboard to deal with the problem. You do realize that even the “expendable” military UAVs are flown by a human on the ground? Autonomous cars are more challenging than aircraft? You must be joking!
I would like to see how an autonomous jet would deal with a double bird strike over NYC… Needless to say I think an experienced human is much better at dealing with unexpected problems than a computer.
‘Tom, please explain to me how your autonomous aircraft would have accomplished the “miracle on the Hudson”’
Well, that’s easy, I’m not familiar with the Qantas incident. A fully autonomous system would have done pretty much what the pilots did, only it would be normal, not a “miracle”. It would have tried to restart the engines while simultaneously calculating landing spots. It would have instantly that the airports were unreachable, and that the best surface was the river. It could have easily alerted all relevant authorities at the same time.
We call it a “miracle” because so often people screw up. When they don’t, a miracle has happened.
BTW, see Xavion (http://is.gd/go9MNz) for a $100 iOS app which constantly calculates a best emergency airport and displays a series of hoops so you can fly to the runway threshold. This is far, far more than the simple “nearest airport” aviation GPSs have. The software is incorporated in Vertical Power’s VP-400 unit, coupled so an autopilot, so if you push a button it will do the flying for you. Just a few thousand dollars for experimental aircraft.
And yes, self-driving cars on unmodified roads are far more difficult to design than self-flying aircraft.
I have no time in big jets or anything with that level of automation beyond alt/course hold. But something you said sparked a thought. The auto throttles decrease power at flare altitude? The pilots was obviously making the approach well below glide slope from the videos… is it possible they actually descended to flare altitude and the throttles did what they were suppose to do just not over the runway?
The ILS glide slope was NOTAMed inactive at SFO at the time of the crash (don’t know about the visual aids), so the aircraft had no electronic knowledge, I presume, of its position relative to the proper glide slope.
Apparently autothrottle settings and Flight Management Systems or “autopilot” settings can be complex on the airliners. I have no idea what they would do without an ILS active. But it’s the professional pilots’ job to know that. They clearly didn’t.
From my study of areodynamics, I know that if bad things can happen when airspeed gets 5 knots slow, you know something bad is going to happen if it gets as slow as Asiana flight 214 was, on the hole finnal approach. Has any other captian ever let his aircraft get that slow and low, with a heavy aircraft, and after landing not wanted to tell everybody the co-pilot is the one who did it.
The problem with the “rampant speculation” is that the NTSB was releasing info as they got it without anything to support that evidence. This is no different than a mystery writer throwing out red herrings and led to the news-prognosticators to chase non-domesticated geese. Two examples:
1) The ILS was out of service. My response: So. What? How many thousands (millions?) of landings each day are made without using an ILS? BUT, I’ve already seen comments from the “flying public” about emailing their congressman (little C) to demand that the FAA install a backup ILS so that one is always operating. I’m sure Chuck Schumer is already drafting language…
2) Same issue with “his first time landing at SFO.” Again, so what? As with #1, I’ve already heard calls to require a set amount of landings at an airport before they can carry passengers. Don’t worry, Schumer will save us here too.
3) It was reported the the NTSB said there were “no communications” in the cockpit. Without a CVR transcript as backup, what does that mean? Does it mean that nobody saw an emergency? Does it mean that standard callouts weren’t made? Does it mean that ATC wasn’t on the radio warning the pilots like the LSOs did in the Top Gun movie?
It’s obvious that the pilots got low, slow, and stalled. What’s not obvious is why. Was there a CRM issue? Did they make standard callouts? Was the PNF actually monitoring anything. Let’s have FACTS before we start getting the masses riled up and demanding that do something like “banning di-hydrogen oxide.”
Fully agree Sam.
1) So what if the ILS was out. Surely the WAAS vertical guidance wasn’t out!
2) So what if it was their first landing at SFO. Surely they had flown any number of approaches and landings to SFO in the sim!
3) I haven’t read anything about cockpit communications, except that they apparently did call for a go around 3 sec and then 1.5 sec before impact. Which answers Blaine’s speculation of power being added just before impact.
And Tom Yarsely is totally correct: it’s pure luck that they impacted where they did. Anywhere sooner and there wouldn’t have been anything left of the plane.
What isn’t known and released so far is what was covered in their dispatch packet and how did they configure the avionics?
Wise pilots use as many redundant systems.
FAR 91 requires large aircraft to fly at or above the glide slope. Did the crew configure for an ILS and not monitor for status?
It was a sterile cockpit, too sterile. ILS approaches show altitude at the OM when on the GS. That is an important check, if the plate says 2760 MSL there is a significant, 50-75 feet, difference you want to know if it is a miss-set baro in the altimeter or the GS is out. In any case, pilot and IP/CP failed.
ILS (ground equipment) was out of service.
“the NTSB was releasing info as they got it without anything to support that evidence” Not they were not. Your own examples show the NTSB was releasing info WITH FULL SUPPORTING EVIDENCE. In other words, the NTSB was releasing facts, with no further explanation or speculation as to what those facts might mean. That further explanation will come much later. You and everybody are free, meanwhile, to do whatever you want with the FACTS they supply.
Red Over Red, You’re Dead!!
I learned that in my first week of training.
I would hope that ATP pilots would look at the PAPI lights as a cross reference to their fancy automation equipment.
It’s not a race or cultural issue as (sadly) Dick Collins tried to hint toward. That foolishness belongs to the 50’s era!!
It’s stick and rudder skills that is ignored despite the pilot’s country of origin.
Recall the Air France stall a few years ago? All Caucasian crew I believe!
Dick Collins and the anonymous Internet article author weren’t saying it was racial. They definitely were saying it might well be CULTURAL. I’m an engineer in a computer modeling group with many foreign-born engineers, and each culture definitely has a pattern of thinking, the American culture included. That’s not racist, or 1950s thinking, it’s simple observation. To deny it is to fall into Political Correctness and all that fuzzy thinking.
Given that Westerners generally and Americans in particular designed and built both the ground-based and flight-based flight and navigation tools, it makes some sense to me that of course we are more adept at them: they were designed with our culture as a basis. IMO, recognizing culture was the genius of the late Steve Jobs. He hired excellent engineers and designers, but insisted they design products for non-engineers. Contrast with Microsoft where all the software has an engineering flavor to it and everybody has complaints about its use. Or Linux, where the computer geeks for years didn’t care less about Graphical User Interfaces and were happy with command lines…resulting in Linux being never adopted for the mainstream.
Interesting graph. The July 5th approach looks unstable as well, I wonder if that’s the reason that flight did a go-around. Time to check the LiveATC recordings.
Ohhh!!! Didn’t hear about that one. Pretty interesting find – well done.
Mr. Collins offers, “I think that speculation is good because it gives things to think about when the event is fresh in mind and a lot is being said about it.”
Well, yes and no. When uninitiated folks do it, it’s simply an exhibition of human nature. But, when knowledgeable folks do it, it can create misleading opinions. Now, if people were ready to prioritize those opinions for reliability or how likely they are to be correct, and would adjust those priorities as new evidence comes in, and be ready to abandon an early opinion as it becomes unlikely as more new evidence arrives – then I would not worry about speculation. But, then, it wouldn’t be speculation, it would be more like real scientific exploration.
Problem is, that’s not the way a normal human thinks. Normal humans often (usually?) don’t bother to separate the fly specks from the pepper when rampant speculation is flying about. Normal humans don’t let facts get in the way of a good story.
For a skilled investigator to be postulating some various, possible scenarios early on is a good thing. For newspaper reporters to be doing it isn’t – unless truth isn’t your goal and only selling newspapers is.
Oops! I may have just described our media.
One idea that I haven’t heard discussed is the Design Eye Reference Point. I wonder if the captain adjusted his seat so he could see properly.
“Fly the airplane”, for Pete’s sake! It seems that none of the flight crew was busy with that single, simple idiom. After 20,000 hours of my own PIC experience, it still seems good advice . . . high-time, heavy jet experience and attendant technical conversation notwithstanding.
If the 5 day chart accompanying Dick’s comments means anything, it appears that basic stick & rudder skills had deteriorated significantly. I’m not surprised. In a multi-hour trans-Pacific flight, how much actual “flying the airplane” time does any pilot get? My own stick & rudder skills deteriorate if I don’t use them regularly, and I’m sure that’s true of every other pilot as well. If these guys typically land, even in good VFR weather, letting Otto do the work on a coupled approach with Otto working the throttles along with everything else while they watch, it’s no wonder that they are klutzy at best, hand-flying what should have been a normal visual approach.
So what’s the lesson? Fly the airplane! Do it often enough that the skills remain sharp. That’s true for both airliner pilots and GA pilots. Pilots who rely on automation to the exclusion of actual flying are making a mistake that kills.
As for those who advocate autonomous, pilot-less airplanes, I can’t help but remember the joke, “Welcome to Automatic Airlines”, where “…nothing can go wrong, can go wrong, can go wrong, can go wrong, can go wrong.”
While I am not an ATC airline, jet qualified pilot, I think there has been a whole lot of over intellectualizing concerning this incident. My bottom line is that a cockpit of highly qualified pilots screwed up a visual approach and landing. Would it happen with the same crew again, highly unlikely, but on this particular day they were less attentive than flying (any airplane) required.
As an LSA (CTLS) pilot this incident was a wake up call–PAY ATTENTION TO WHAT YOU ARE DOING–IT IS,INDEED, MATTER OF LIFE AND DEATH!
Hi Ken….with all due respect, it’s interesting that you comment on this as an LSA pilot and indicate that some are over intellectuallizing. You first quote your lack of expertise on the subject and then you proceed to make an over siplification of the situation. Next time you sit next to a pilot who definitively shows a lack of aptitude, and yet, is given the responsibility of other peoples lives, whether it be an LSA or another larger aircraft, it must not be “over simplified”, and does deserve the “over intellectualizing” as you call it. We have a serious problem that needs to be addressed in one fashion or another or more lives will be lost not to “accidents”, but to “events” such as this. In flying these types of aircraft, the pilots MUST “over intellectualize” in order to produce the highest safety factors that we can. CRM, for the most part, is in fact, “over intellectualizing” the psychology that exists between crew members in the cockpit. It is not a simple process to implement good CRM. It takes a lot of intellectualizing to accomplish the rapport and professionalism needed for this type aircraft to be operated safely. A two pilot crew operating an aircraft of this type of performance is a far cry from the fun and enjoyment of your LSA…which we all love to fly, by the way.
I fly a Piper Apache and can only imagine what flying a “heavy” is like, however I believe that all pilots need all the help they can get when trying to land “on the numbers”. I believe that an operating ILS (or even a VASI)would have prevented the accident. Judging altitude visually from a cockpit high up in a heavy jet is bound to be difficult.
From looking at the altitude plot for the flight, I would guess the pilot felt that he was high way out. He most likely retarded throttles to lose altitude. Approaching over “featurless” water gave him no clues about altitude until it was too late to recover. The Tower should have offered him an approach with an operating ILS and in my opinion the accident would never have happened regardless of pilot inabilities. The Controllers and FAA ILS maintenance caused the accident.
Bobby..the flying of your Apache is like a runabout and the flying of the 777 is like the Enterprise (Carrier). The runabout pilot has no idea what the Enterprise Captain is up against. This is absolutey, and unequivocally, NOT an FAA or ATC issue. This issue is soley in the PIC lap where the “PILOT IS THE SOLE ATHORITY AS TO THE SAFE OPERATION” of ANY aircraft.
Bobby, have you ever heard of a RADAR Altimeter? All the big guys have one, specifically so the guesswork about altitude is removed. Also if you fly heavy’s everyday, you will get used to the sight line of a properly executed approach. Likewise, whether it is a 172 or a 777, the VASI/PAPI etc. will give you an immediate, real-time display of your glidepath. I understand your comment about featureless water approach, but again, that is why the pilot(s) has RA and baro-altitude, right in front of them….
If you research the accident, you will see that earlier on in the approach the plane was indeed high and fast (under ATC direction) and most cognizant speculation concludes the plane went from being too high and fast to too low and slow, which is apparently pretty easy to do in a heavy.
I agree with Dick and a lot of other commentators, these pilots displayed a basic lack of piloting skills and were unused and unable to execute a simple visual approach on a clear VMC day. That’s it, no excuses.
There are also a real cultural issue, it is not jingoistic to comment on the fact that 4 ATP rated pilots sat there and did nothing until a few seconds before the accident, when it was far too late.
Finally, so the ILS was OTS, so what? If the pilots had their head in the game, they could have punched the RNAV (GPS) RWY 28L app into the FMS and used the vertical guidance to ‘help’ them keep on a virtual GS, all the way to 200 ft. Not that they should have needed it.
Incorrect. Ignoring their speed is what caused the accident. The ILS being out could be a contributing factor. And the PAPI lights WERE working. They were only NOTAM’d OTS after they slid through them.
By your logic, the FAA would be to blame if you VMC rolled your Apache while making a single-engine landing without a working VASI.
Landing at SFO is anything but “over featureless water”! You can practically reach out and touch the coastal hills, and in between it’s all city and freeway and wetlands.
Both of the RNAV approaches to 28L have vertical guidance. This press hysteria about the ILS being out of operation is totally irrelevant.
The VASI lights were working but more to the point the big jets have radar altimeters that call out the altitude at least 500 feet continuing to the call for ROTATE.
I don’t know how you can lay any blame on the Controllers???
I feel I need to add a few words here in regard to the comments regarding fully automatded systems.
First, no human activity is risk free. Second, it is a chimera in our now risk-averse society to believe that increasing automation will improve safety.
Do you REALLY believe that fully autonomous automobiles will NEVER suffer failures or accidents? That they offer the the prospect of zero accidents and deaths or injuries? Equally, do you SERIOUSLY believe that fully autonomous aircraft will never ever crash, that they will never ever fail, that no-one ever traveling on such an aircraft will ever be injured or die?
As an engineer I consider that modern automobiles are about as safe as they are ever going to get, that modern aircraft are equally as safe as they can be. As a programmer I have to tell you that even the very best supercomputer does not have the intelligence of a cockroach. Yes, we have expert systems, we have computers that can play chess, as well as speak, and do lots of other marvellous things, BUT THEY ARE NOT INTELLIGENT! If you remove the pilot from the cockpit, and fully automate the aircraft, then you will also have to fully automate the ATC, and, do you honestly believe that safety will then improve, that the promised land of zero deaths, injuries or accidents will be within our grasp?
It is already clear that we have all but removed the pilot from the cockpit and now rely far too much on automation, AF447? Unless and until we have machine sentience that is at least as good as human intelligence, and I do not believe that we will reach that state, then in the final analysis it is the pilot in the cockpit that will be the saving grace. I do not wish to live in a world controlled solely by machines. I do not wish to have it mandated that I as a pilot am no longer to be in charge of an aircraft, as a driver that I am no longer to be entrusted driving a car – I wish to use automation only as far as it is necessary and desirable, but not to the point where I as a sentient being am eliminated from the control loop, because I, as a sentient being, can far better decide what needs to be done than any machine following a list of instructions.
The controllers weren’t flying the airplane. The pilots were. Or, as it turns out, weren’t.
This accident is certainly garnering a boatload of attention. I believe that the reason why is that we have almost eliminated most of the mechanical and weather related reasons for accidents. The only ones that occur now are the ones related to pilot performance. The accident rate is so low because on the whole, pilots are actually doing a very good job (FAR 121). So, when a crew turns in a less than stellar job of flying the airplane, we all run around with our hair on fire over the dire state of pilot flying ability. This is just absurd. Look at all of the successful landings that occured during the week leading up to the Asiana flight. They all dealt with the same conditions: ILS OTS and I believe that the PAPI was OTS as well.
Yes, airline pilots can land on just a bare piece of pavement without any electrionic aids. This goes way back to initial training as a pilot. Most of the time, we choose to back ourselves up with whatever is available, ie ILS, RNAV (with vertical guidance). The airplanes we fly today are so capable they can even supply us with a self generated 3 degree vertical path without any other approach selected.
One thing missing from this discussion is automation mode awareness. With the rise of automation in our cockpits, it is imperative to always know what mode your automation is in. Bad things happen when you lose that awareness. You should never have to ask, “What is it doing now”? or “Why is it doing that”?
I flew nights for 12 years and still fly the occaisional red eye. I know how tired I am when I do those flights. When it comes to the approach after flying all night, I want all of the capability I can get. I would have at least programed the RNAV 28L approach to back me up. Do we know what the Asiana pilots did?
It seems as though we are seeing an increasing number of incidents where ATPs have forgotten or have never been properly trained to fly the aircraft. It is inconceivable to me that an experienced pilot and crew could do such a botch job of energy management on final. I suspect poor manual flying skills mixed in with a dash of misunderstanding of how the automation works will be the ultimate cause of the accident. Even a rookie pilot knows to respond quickly to deviations from approach speed on final. The accident is a failure of training in all likelihood, not regs or systems. Are we reaping the seed of ab initio training?
Might I suggest that Dick Collins get above the title credit. I overlooked this fine article and discussion because I didn’t see that it was Collin’s authored.
I got a separate email that Collins had weighed in and then determined that anything that Dick says, Dan must listen to.
So, put Dick’s name above the title
DICK COLLINS SAYS…
My rampant speculation, revolves around an ILS GS OTS, the use of autothrottles when there was no glide slope to follow and fatigue effects on pilots’ vision.
I do not know what cockpit equipment was available, such as a GPS precision approach or just plain using the Outside World Indicator, commonly known as the window. Any 10,000 hour pilot should be able to see a perspective and fly a 3-5° visual slope. The airspeed should have had a bug set, and a then after the crash it was reported the Captain told the cabin crew to not do an emergency evacuation.
The airline is upset because a joke in poor taste made fun of Oriental names. I am more concerned with the dispatchers and the pilot training standards.
BE 300/350, BE400/MU300 ATP CFI ASMEI
Having trained Korean pilots in Korean airlines in Korea, the answer to the ab initio question is, yes. The airlines went to college campuses, recruited people who don’t even have driver’s licenses, GUARANTEE them a job, and away we go. The airlines will look at each pilot with a dollar sign in front of his name. He may get some additional training if needed, but, he was GUARANTEED the job so there is ONLY a Drop On Request. If the DOR is not exercised by the pilot, then he will be “moved along” to the next phase and it is hoped that he will “catch up”. This is why the seniority system is not used since many pilots cannot make an upgrade so they become senior co pilots and will not become Captains at all in some cases, but, are retained as co pilots due to the costs to replace them. An additional factor to look at as well is that most of the pilots simply were recruited and accepted the job since it was lifetime employment. In the States and other Western environments, most people entering aviation are doing so because it has an attraction as something they want to, and if they can display an aptitude for, which is generally required to get through stage checks and check rides, you then have a pilot with more aptitude and desire than simply looking at flying as nothing more than ” a life of employment.
As student pilot with some 63 hours I have problems with flaring a plane on that particular day I was by the Benihana restaurant watching the heavies as the tower called them with my hand held radio I phone with ForeFlight to look at the runways on my iPhone also my fov binoculars saw this Asiana 214 seemed odd to me he seemed to be flaring on the ocean over the lights that extend over the water while the rest of the heavies would come in and straighten the plane on the runway and flare on the runway at least that is what it seemed to me from a distance ?
Low & slow for sure — BUT not just by a little for a short while — they were WAY low … and WAY slow for a VERY LONG period.
It wasn’t even remotely close! This was not a minor, easy to overlook error – they were hundreds of feet low for at least 1/2 mile to perhaps as much as a mile or more out.
I simply can’t imagine how either of the cockpit occupants (pilots? not really) didn’t notice (regardless of a working PAPI) long before they finally did a few seconds before impact (when it was WAY too late) and take corrective action.
Weren’t we all taught as student pilots what a runway sight picture looks like when low, high, or on proper glideslope?
That is what I find inconceivable and inexcusable regardless of autotrottles, culture, etc.
I have over 15,000 hours flying the 747-400 which other than having 2 more engines, is very similar in the cockpit. My lawyer told me to always preface everything I say with ” I might be wrong about this ” but……
So no one mentioned that an LCA (aka. Instructor pilot ) was in the right seat. What the heck was he doing? Does he have to bow down to the captain too? Ok, too low and too slow is not a good combination in the big jet world. How they got there is anybody’s guess. The NTSB will figure it out and they will probably make a video on “Air disasters” about it.
“I might be wrong about this ” but, In all likelihood, the auto throttle was disconnected somewhere and no one noticed. ( Two perfectly timed clicks of the auto throttle disconnect switch and you get no aural or visual indications ). Hmmmm Since they rely heavily on the automation, they probably were very undisciplined when it came to manual approaches. No excuse, just a probability. . So where does it leave us. There will undoubtedly be many contributing factors and will the pilots take the majority of it? Who knows. My guess ” I may be wrong about this, but “. I say pilot error will be the biggest factor.
And yes, had it flipped over on the cartwheel, the outcome would have been considerably worse. Thank God for that.
Fly safe, Bob
I’ll throw my pennies in the well
My guess… He was probably 10 miles from the airport sight picture high and fast, pulled the engines all the way back to get down, got the sight picture he was looking for still at a high rate of descent and engines at idle. Brought the throttles up and waited… and waited… and waited … for them to spool up.
These engines take about 5 seconds to spool. The airplane had decelerated to Vref speed about 1 minute prior to impact, yet the throttles weren’t advanced until 3 seconds before impact. The pilot kept pulling back on the stick to flatten the trajectory, expecting the autothrottle to maintain Vref speed, but nobody was monitoring instruments or throttles to verify. They basically sat on their hands for almost a full minute while airspeed dropped 35 kts below Vref. Unbelievably poor basic airmanship.
WOW!!! And that was four ATP pilots right!?!? Oh man.
I’m for using every bit of the technology that is available to pilots today. Sure beats shooting a four-course range. What I’m also for is having pilots, whether airline or GA, who can manually fly the airplane when the techy stuff goes south. If the airline pilots can’t get enough in-the-air stick time because of company policy, then require more sim time. Oh, I know: it would be expensive. That’s true. But, as Dick Collins points out, we’re gonna see more of these types of accidents if pilots are not on their game when the electrons stop holding hands in whatever conditions that might occur. And I’ll bet you a free hotdog at Sporty’s that it won’t always be in CAVU conditions like it was at SFO. As for GA, hand fly the darn thing every chance you get. After all, that’s what you’re paying all that money for: to fly an airplane. If you just want to manipulate computers, fly a sim on you desk. It’s a lot cheaper, and there’s always that Reset button.
All of the hand-wringing about stick and rudder is valid. But I’d argue for an even more basic factor–pay attention to the automation. Even if you’re automation-dependent, recognize when it’s off. If the guys would have click the auto-throttles back on they probably would have been fine–even with their bad flying skills.
Option 1: know how to fly it without automation.
Option 2: if you really need the automation to fly, then watch it like a hawk.
Most of us who grew up in this here good ole’ USA got the benefit of growing up observing and becoming familiar with the operation of cars, sometimes planes, that were being operated by our parents, their friends, etc. I think this really helped prepare us for not only the operation of said vehicles but also we had the opportunity to see their responses in the myriad of different situations that presented themselves as we moved along to our various destinations. In other words how and what to do in crawling as well as high speed traffic. I remember that by the time I was 13 yrs old I was chafing at the bit coming out of my skin to get behind the wheel and apply what I felt I was sure in my heart and soul I knew how to do properly.
Now if I grew up in some other country where the idea of owning a motor vehicle was beyond my means would I be as well prepared as I found myself growing up here.
Having said all that above, when I was in NAM’ we were sitting in a field in our huey waiting for our next sortie when a huey operated by the Vietnamese Air Force landed and shut down right alongside us. Now in a huey the starter is also the generator. Once you start you flip a switch appropriately called START/GEN to GEN. Part of the start checklist naturally is to flip the switch back to START. Now the VAF ship alongside us is ready to go and we’re all sitting in our ship listening to the ignitors snapping away and black smoke and heat plume coming out of the tailpipe but no whine of starting engine or turning rotor. Once, twice, three I don’t remember how many attempts were made exactly until one of our pilots finally couldn’t take it anymore. He jumped out of our bird shaking his head, opened their door, reached in and made things right.
During the seventies the US government was making friends with the Arab world by selling them military and airline equipment and promising to teach their people how to use and fix them. So they contracted out to various schools here in the US and I just happened to be taking an A&P course at an institution where about 2/3 of the school’s students belonged to foreign air forces or airlines. This was a very good school and the instructors had a reputation of taking no prisoners. I was in a mixed class(Foreign and American) one day and it was either Pressure Carburetors or the inner workings of a mechanical Jet Fuel Control but needless to say these devices are very complicated with action in one part creating reactions in others(changes in press, flow rates, metering, regulating) and you really had to focus to follow, take notes and keep track of everything. Instructor turns around and asks if there are any questions so far One of these foreign students raises his hand and in a frustrated manner asks, “What is inch”. I thought the veins in the instructors head were going to pop.
I think Richard Collins article hit the nail on the head. I have worked as an AMT for a part 121 air carrier on Guam for the past 18 years. In that time I have have handled different foreign carriers which include Korean Air, Japan Airlines, and All Nippon Air. Japan and Korean airlines still have a problem with questioning authority in the cockpit. I was working the night The Korean 747 slammed a mountain ridge about 3 miles short of PGUM runway 6L. Despite the changes the FAA mandated for Korean Air before they could to fly back into the U.S. We still get write-ups or squawks from crews trying to make a CAT 3 / Auto Land 3 approach into Guam. PGUM does not have CAT 3 Auto land 3 capability. The problem is not limited to Korean and Japanese airlines, There have been problems with Philippine and other foreign carriers. A bright spot is because is there are now more Australian, American and European pilots flying with these carriers and the younger generation of pilots who aren’t question authority.
The last sentence should be . The younger pilots aren’t afraid to question authority in the cockpit
Methinks the cultural canard is true when it comes to many technical faux pas today. The world has been copying, adapting and learning US originated technology since WWII, granted many great inventions/inventors were Asian/European as well. Like it or note US know how, do how (operators aka pilots) and culture have proven superior and not translated well to other cultures. Just look at Airbus, a government copycat of Boeing repeatedly getting its control software wrong in the Airbus and EAD Ariane 5 fiasco. The French and Germans are great engineers but copying and re-inventing is not the same as originating. Next add the latest modern ingredient of regulator, managers and pilots having no engineering, flying or operations experience and you can see the future is bright for tort lawyers. When you fly, fly the old school US carriers, if you care to think.
An ATC controller weighs in. A friend sent this to me and it is germane to our coversation to say the least…
Actually I was a bit surprised NorCal gave Asiana a Visual Approach to 28L. Having worked a lot of my career in Chicago and Los Angeles I have seen the consequences of foreign air carriers doing visual approaches. Going back to an Air France 707 landing at Navy Glenview. At LAX, they have prohibited VA’s to LAX by foreign carriers for quite some time (more than 20 years). This actually started after several incidents involving KAL. I recall giving a KAL 747 a visual approach to 24R from 10 east of SMO and advise him of traffic that would be remaining south of him for the 25 complex. Well…you guessed it….there were a couple straight ins to 25L and 25R and here comes KAL on base for 24R flying through 24L, 25R and 25L before aligning for 24R. Needless to say…lesson learned. After a few more incidents with foreign carriers the directive came down. No visuals to foreign carriers.
I was especially surprised at SFO because of the distance between the runways.
Although I have a great deal of time in NASA’s 747-400 sim, I am not a rated transport pilot but am certainly in agreement with all of the professional pilots that have commented on the auto-throttle issues and total lack of attentiveness to speed on final. What the hell were they engaged in? Culture issues, 43 hours in type, fatigue, on and on…..I have never witnessed a professional pilot ignore speed on final. Amazing!!
During my work at NASA over the past 10 years I have had the opportunity to observe hundreds of airline pilots from all domestic carriers with all levels of experience. There is certainly an increasing number of pilots that have difficulty getting to the airport from 20 miles out after reporting it in sight without flying an ILS or at least vectors to final. As an ATC dinosaur heavily involved in Next Gen research, I observe large movements towards automation and away from those skills that both pilots and controllers have honed to a very high degree at some of the major airports in the world. Pilots and controllers in the future will barely resemble the dinosaurs that many of us are in aviation.
I have no doubt that our government in their relentless quest to protect us will be taking a very serious look at visual approaches in the future, once again adjusting the system to accommodate the lowest common denominator. Jim
This has been a really good conversation. We have had a number of highly experienced pilots and instructors weigh in. Some with first hand experience as relates to the issue at hand, and then, a number of pilots offering insights and opinion based upon experience and/ or individual perceptions. It would seem that we have reached a point where it is now an exercise in patience to see what the computer generated video of aircraft profile and configuration looks like and we will see what the real scenario looks like. Thanks Dick for generating an interesting exchange.
I will also speculate that this particular “accident” is going to be a pivotal one in terms of pilot training as pertains to emphasis on real pilot skills versus an automation manager. The bottom line is…The pilot is the final option when automation fails, as it always will either through mechanical failure, electrical failure, software mis programming, or, mis management.
“The bottom line is…The pilot is the final option when automation fails.”
I politely suggest that complete automation is a valid bottom line for addressing when the pilots fail.
Not disagreeing in your theory, machines have the *capability* to perform flawlessly and with reaction times a million time better than a human.
But we’ve had 117M airline flights in the US since 2002 and only 3 of those had fatalities. That is very close to 7 Sigma levels of performance. Given that manufacturers of electronic goods are happy to claim 6 Sigma accuracy (which in reality is only hitting 4.5 Sigma), how much investment would be required in the hardware and the *software* to approach the safety level we already have??
I’m all for safety, but I get the impression that many want to see a “zero tolerance” law for risk of any kind.
Sam..absolutely right, however, this accident shows why we can’t ever reach “0” tolerance since there are pilots that when these excellent automatic systems fail, they are unable to provide the “human backup” of actually hand flying an aircraft. The potential solution, if I am not mistaken, is simply much more focus of initial and ONGOING currency on this ability to be the “buck stops here” human element that is supposed to “save the day” versus the, “ruining the day”. This was an example of “the lights were on, but, nobody was home”.
As noticed, an autonomous UAV just crashed on takeoff in Florida, closing a highway…not quite ready for prime time?
I was very impressed with the fact that the Boeing 777 composite materials fuselage held together so well. I’m guessing a like sized aluminum aircraft would have broken up between the tail section being clipped and then a flop back onto the runway sliding backwards.
Boeing 777 is NOT a composite aircraft.
Never said it was..
sorry wrong article..
utilize a Boeing/supplier
fiber epoxy resin system.
From a passenger view point – when is the aircraft industry going to stop being allowed to use flammable toxic materials in the fuselage – time and time again fire has consumed the fuselage to the detriment of passengers and aircraft. Please don’t tell me there isn’t the technology available or cost would be to much. As for speculation of the cause of the crash, whats the point? Can only be for some people to claim the high ground when the investigations are done. Yes egoistic about covers it!
For my part, as a pilot, whenever something like this happens I wonder just what was going on in the cockpit, and is there anything I would have done differently, or is there anything that I could learn from? Pure human curiosity – not egoism. In my case the curiosity is compounded by the fact that I grew up flying in the SF Bay Area, watching thousands of jets make that approach – and landing runway 28 at SFO as a passenger is simply a thrill for any pilot.
The standards are pretty high now for flammability and toxic gassing. The airline industry has some local control. Every airplane delivered by Boeing, Lockheed, or Douglas has met FAA Part 25 standards. But after a few years, airplanes based over-seas my have interiors replaced with material that meets local practices.
But the cabin fire mostly happened after everybody got out. Despite the Captain saying not to do an emergency evacuation, the beginning fire prompted the cabin crew to ignore the flight crew.
Just as a side note, after years of teaching students how to fly I came up with several teaching techniques and tests for before solo flight. My students made take-offs, patterns and landings without any instruments at all and they were expected to be on the proper altitude and airspeed by sound and visual perspective. After as little as two hours of instruction over a period of days, a student pilot can identify 1,000 foot altitude within 100 feet. Before solo my students would have an unexpected door open just at rotation speed, their react helped me determine whether they were ready to fly solo. If the grabbed for the door, they were not ready. If they said “Did you do that?” while the either aborted or continued the take-off they were ready.
All my multiengine students experienced an actual engine failure 25 feet off the ground, on a long runway and not without a briefing. I wanted my students to be prepared at 2 AM with their wife and kids on the airplane to know what to do. This was in a Beech Duchess, in a King Air or other high performance aircraft, once the gear starts up, you fly the airplane. You don’t retract the gear until you can fly on one engine. In some airplanes you must have both engines at some airports, some density altitude. Knowing what that is is part of a “I’m ready for take-off” radio call.
I don’t think this flight crew was qualified to have me solo them them, they were dependent on the equipment and unable to recognize it wasn’t working the way they expected. They pushed a button, but never monitored.
As I watched the Apollo 11 lunar landing for the Nth time yesterday, I was reminded again that computers don’t always put you where you want to be. If Neil had not been able to hand-fly the final approach (landing with mere seconds of fuel remaining with zero help available if they crashed; talk about pressure)…well, I don’t even want to think about that their fate would have been. Neil had the “Right Stuff”. Shouldn’t every pilot always have the level of “Right Stuff” on every flight commensurate with the aircraft he/she flies in the conditions in which he/she flies? Of course. Please don’t wait for the rules to require you to “brush up”. Use every flight as an opportunity to the extent practical to stay sharp. If that isn’t enough, get an instructor and go knock off the rust. Too much depends on you having the necessary “Right Stuff” every time not to do so.
Flying airplanes should not be about having the “Right Stuff”. That is for test pilots, astronauts and Unicycle riders. Airplanes should be easy to fly and not require that all pilots, all of the time, have the right stuff. We need safe pilots and they need all of the assitance available to aviation when flying airplanes full of passengers. We don’t buy airline tickets to take chances with pilot proficiency. An airplane with an autoland system should always land with the pilot monitoring the landing, not flying it. The approach controller should have given the Asiana plane a runway with an operational ILS.
The “right stuff” is something all pilots should have, must have. It isn’t difficult and I’ve only met a few pilots who should not be flying anything. [One was a USAF “F4 driver” his words] who was unable to adapt to the lower performance airplane. The best pilots I have ever instructed were also military trained, one was a KC 135 pilot, another was US Navy T28 instructor who had also flown the LC 130 to the South Pole. He was getting civilian instructor ratings and he was very good. A few weeks later he was my check pilot when I flew the Beech King Air 200. We got a chance to fly a loose formation with the Space Shuttle while it was on the way from a stop in Wichita to the Paris Air Show. When we were about a mile away, we both said, I forgot my camera.
BTW, that landing and flight management demonstrated by the crew would not have been any better with autoland, which except for a few military airplanes depends on ground based equipment that is followed by the airplane’s radio and autopilot.
Sooo…. is anyone shocked that plane totally crashed, cartwheeled, and only 3 people died? All this negative talk about the obvious shortcomings, and what could be done better, but no-one really pointing out the chances of so many people surviving a crash like that is AMAZING. Is it that the new planes are really built stout? or just dumb luck?
It was a spin not a cartwheel. It apeared to some people to have cartwheeled, but it actually just spun around with the wings in the air and the nose on the ground. It is a well built aircraft, not luck. If that aircraft had flipped on its back or cartwheeled, nobody would have walked away from that flight. The 777 is an awesome airplane if you ask me.
Given these comments on survivability…go look at the videos of the DC10 crash at Sioux City. It was a United DC10 with NO hydraulics. To realize that folks survived that “landing” is beyond imagination. And yes, by definition…it was a “spinning cartwheel”….LOL…
So, let’s add to the mystery and speculation:
This crash happened on July 6, 19 days ago. On July 8th, the NTSB had listened to, and reported on, the CVR recording. As is standard practice, the actual audio will not be released, but they said they *may* release the transcript of the Cockpit Voice Recorder for the crash **at some point in the future.**
This past Monday, a Southwest 737 had a collapsed nose gear. Today, they’ve already blamed the crew (3° nose down pitch at landing) AND announced the CVR transcript will be available **tomorrow**!!!!
Still nothing from Asiana 214… Nope. Can’t possibly be anything to hide here. Move along.
A Canadian perspective
George Jonas: The perils of autopilot
George Jonas | 13/07/24 | Last Updated: 13/07/23 3:56 PM ET
More from George Jonas
For some pilots, it’s 10,000 hours in the air, but only 100 at the controls.
ADEK BERRY/AFP/Getty ImagesFor some pilots, it’s 10,000 hours in the air, but only 100 at the controls.
Asiana Airline Flight 214, a Boeing 777 with 307 souls aboard, crashed on approach to San Francisco International Airport on Saturday, July 6th. The aircraft broke up and caught fire short of the button of Runway 28L. 123 of the passengers and crew staggered away without requiring hospital treatment, but 181 people were injured, 13 critically, and three passengers lost their lives.
It was a nasty landing accident on a nice summer morning. All reports noted that weather wasn’t a factor. It was what pilots call CAVOK, light wind, ceiling and visibility unlimited. It seemed to me, though, that the weather being ideal for flying didn’t necessarily preclude it from being a factor in the accident.
During the next few days, the National Transportation Safety Board (NSTB) recovered the aircraft’s flight data and cockpit voice recorders, analyzed them, then proceeded to do something unusual. At early press briefings, accident investigators tend not to reveal what they know or really suspect. The idea is to discourage “speculation.” (It never does, by the way; it only makes speculation more idle.) Breaking with this tradition, the NSTB’s new head, Deborah Hersman, stopped obfuscating, and shared the salient points with the world media as she received them.
The NSTB’s preliminary thinking reinforced my impression that weather was a big factor in the crash of Flight 214. In bad weather, it wouldn’t have happened. Flying in instrument meteorological conditions, the pilots wouldn’t ask for, or receive, clearance for a visual approach. Instead of hand-flying, they would program the computers to fly one of the standard terminal arrival routes in the murk. The human pilots on the flight deck would act as systems managers, monitoring the multifaceted autopilot that, “in conversation” with the navigational computers, manipulated the controls. They would land on a runway where the instrument landing system was operational (on Runway 28L the glideslope was being serviced), with a human pilot taking control – maybe – for the last 800 feet before touching down, engaging reverse thrust, taxiing to the terminal and Bob’s your uncle.
I wrote a column saying so, underlining that it was speculation of the kind that would probably dismay the Air Line Pilots Association International. (The powerful pilots’ union was furious with the NTSB for being so loose lipped with the press.) I expected to get letters asking how I could sink low enough to succumb to speculation and I wasn’t disappointed. What I did not expect, but was pleased to find in my mailbox, were letters from training-pilots writing in agreement with my speculation.
From Surrey, British Columbia, came this note: “I retired as a B767 captain about five years ago. I think your observations are spot on. In fact I am willing to wager a beer that that is what will be found to be the cause of this crash.”
When we presented situations which involved visual (no electronic) aids, it created shock and fear among all the pilots of both airlines…
Two of the pilots who wrote me worked for the subsidiaries of aircraft manufacturers contracted by foreign airlines to train their pilots in the operation of the equipment they purchased. “We expat instructors were forced upon them after the amount of fatal accidents … began to be noticed by the outside world,” wrote a pilot-instructor whose sojourn in Asia lasted from 2003 to 2008. “I am not kidding when I tell you that requiring them [Korean pilots] to shoot a visual approach struck fear in their hearts … with good reason.”
“Reading your article took me back more than a decade and I heartily agree with your assessment of the weather influence,” wrote a former Director of Flight Training and Standards for a major Canadian carrier, who spent years setting up training programs for both Asiana and Korean Air. “When we presented situations which involved visual (no electronic) aids, it created shock and fear among all the pilots of both airlines…”
Another training-captain thought the dilemma wasn’t just Asian. “This is a worldwide problem involving automation and the auto-flight concept,” he wrote. A first officer calling for the autopilot to be engaged 250 feet after takeoff is standard operating practice at some airlines. “How much actual flight time is that? Hardly one minute.”
This way we may have a 10,000-hour pilot in the left seat, another correspondent writes, with maybe 100 hours of actual flying time. We have taught our machines to perform so well that we forgot how to perform ourselves.
This is only speculation, but so it may have happened that the qualified crew of a major carrier, flying a serviceable aircraft in perfect weather over the flat surface of the ocean, without any known distractions from conflicting traffic, equipment malfunction, or anything else, while hand-flying in visual meteorological conditions allowed the approach to become unstable and made no attempt to go around until seconds before impact.
User-friendly? I don’t think so. We have taught our machines to fly magnificently. Let’s concentrate on pilots.
eMailNews mailing list
Just 2 comments to the Canadian perspective:
– The approach to SFO ain’t over any ocean!
– Why does CAVU weather prevent a pilot from calling up the RNAV approach and letting Otto fly the approach while he enjoys the view of the Bay and Golden Gate? No one told them they had to hand fly the approach.
MANY MANY planes landed on that runway without the ILS (GPS approaches do just as good a job) and I REALLY hope no-one is going into SFO without the missed approach queued up in the GPS ready to GO-AROUND.
Final thought – He had no idea of his airspeed. If pilots aren’t watching their AIRSPEED on approach and relying on automation to do that… YIKES!!! (They should probably start calling airspeed at altitude marks “140Kts 800′ to go” anyone in the cockpit would have caught on instantly and helped out)
Sorry – Getting more details I found out they were 134 Kts 500 to Go when they realized they had a problem. Target speed of 137. Guess I’m jumping the gun a little saying they weren’t watching airspeed.
There is a level of reasonable automation and we have far exceeded it. After 20,000 plus hours in everything from a J-3 to a 747-400, I have seen way too much reliance on autoflight. The airlines are as much at fault as the manufacturers. After my first glass school I realized my stick and rudder skills were in danger. I bought a taildragger and maintained my proficiency in that airplane the rest of my career. Automation, without the practice of basic airmanship, will result in the atrophy of flying skills. We are frequently reminded of that almost every year by the incidents and accidents which can be traced to a lack of basic skills. It isn’t rocket science.
I have been surprised about two areas where there hasn’t been speculation.