A Boeing 777 Has Crash-Landed At San Francisco International

[PilotDAR]

Stall warning...what is that noise?

'Might be the gear horn instead, equally worth ignoring on short final!

Oh yeah, and the pilot could not figure out the VASIs? Red over red, you're dead! Would you even have to have that explained? Don't a bunch of red lights near a runway mean don't land on it?

[righthandman]

Looks like the NTSB is assessing whether and to what extent to cite Boeing's design as a factor in the Asiana accident at SFO last July. Unsurprisingly, Boeing and Asiana have slightly different views on what the findings should be. (Copied below from Bloomberg.com for those without high-speed connections).

Asiana Crash Debate Goes Beyond Pilots to Automation
By Alan Levin Jun 22, 2014 11:00 AM ET

U.S. investigators are debating whether to blame a Boeing Co. (BA) jetliner’s design for helping cause a cascade of pilot mistakes in last year’s Asiana Airlines Inc. (020560) crash that killed three Chinese teenagers.
h to destroy a 777. A fourth 777 presumed destroyed is the missing Malaysian Air 777-200ER model, the same as the Asiana jet.

Boeing designed the auto-throttle to aid crews while not replacing them, John Cashman, the company’s chief pilot on the 777 program, told an NTSB board of inquiry on Dec. 11. The crew is “ultimately responsible for the safe conduct of the flight,” Cashman said.

Flying is a very dynamic environment meaning things can sometimes go very wrong very quickly. A good example I suppose can be when you are not fully competent with the automation. It CAN be very disconcerting when you get unexpected results I am sure. But I mean come on, the reason you are sitting there as a human pilot is to DO SOMETHING if the automation is working against you rather than for you.

If you don’t understand what it’s doing I assume the pilot of this model of Boeing can disconnect all flight controls and thrust settings from the automation and take over complete manual control with a push of one button, no? I ask this because I really am not that familiar with the plane. I flew a 737-200 for a while and we hand flew the approaches and required one push of the auto-pilot disconnect at the appropriate time.

Before you were allowed to solo you had to demonstrate you could notice and correct a too slow and/or too low approach (or too high and/or fast approach for that matter). In this flight the whole approach was not stabilized the entire time from what I read and the 3 other pilots just went along for the ride while the guy flying flew a perfectly good plane into the ground well short of the runway.

To me it’s the driving equivalent of failing to disconnect a car’s cruise control going at 100 km/hr into a corner that can’t be done below 40 km/hr and on top of that not turning the steering wheel at all when the road is no longer straight. It just shows a complete lack of competence. This airline should be sued out of existence.

[pdw]

... things can sometimes go wrong very quickly, ...

"Things" actually did "go wrong very quickly", and just seconds from that famous "lateral deviation" ... occurring while apparently descending-into/entering a stronger left-crosswind air around 500-1000' agl.

Higher up at the beginning of the approach very dry desert air (desert source region) is moving from SE way above at altitudes beyond 5-10K. IMO therefore little visible cloud can exist to highlight any layers of differring air, just all nice-looking VFR/CAVU, but also no shear if there are no real definite layers noticeable when transitioning into them gradually. Over the cold bay at 100-200agl for example, it is also a cooler/shallow air layer moving east, ... there in a similar direction as the landing.

At 500ft agl the speed was right-on/137kts and glideslope is almost (finally) perfect; then seconds later the serious decay transpires in short seconds (the early photos of accident smoke/dust plumes show there is in fact tailwind at/nearing zero AGL). IMO the actually negative headwind now at the surface was unlike the strong positive one at/above 2-3000agl, nor quite-surprisingly the change to brisk crosswind that is gradually cut-into approaching in between these two altitudes of zero & 2-3K (the significant difference-in-air is crossed / descended-into in what looks like over a 3-4mile distance) ... ie the unexpected / brisk "lateral" drift noticed probably around 1000' or so.

After arresting the 1200fpm at 500ft, which is partly the headwind loss increasing-the-descentrate there while under the low power ... yet this was allowed IMO because of being above glideslope from being too close (ie being "high and fast" at the start of the long final / the 180kts requested by ATC). The airplane IMO must be more nose high (so the Vasi probably still looked OK) after 500'agl. It is at that point "things" set up to tank in 10 seconds, where the Vasi is lost fast too. This makes it a very-small / much-smaller-than-normal time frame where the auto-throttle misunderstanding adds its influence to cause this disaster "very quickly".

Also, what happened to "was blinded" ie 'something blinded the pilot' ...printed in the uncensored/early news quotes during that week ?

I see that fatigue is also mentioned in the 2nd/3rd paragraph in that executive summary of the link provided above.

[BTD]

Also, what happened to "was blinded" ie 'something blinded the pilot' ...printed in the uncensored/early news quotes during that week ?

That was ruled out as a factor early on. During the interviews with the crew even the pf said he didn't believe it was a factor. Just that it happened.

[pdw]

Yeah, there wasn't any more mention of it.

I see mainly the amount of time they're fighting the glideslope "high" is fairly lengthy and quite a while already with the power set low. But after 500agl the airspeed bled off real fast esp when the nose comes up there to reduce descent-rate. This gives me the impression the PIC simply might not expect the need for power to get so great at that point (at/after 500 agl) as the nose comes up there in the flattened-out descent angle fro glidesloping at "700fpm" again, right where the negative component grew as well.

IMO the airspeed decay is missed (or not looked-for if reliant 100% on auto-throttle) where it counted ... by just 10 seconds or so, it seems. That's the only way the mention of "blinding" could have possibly been an influence as well, in my opinion (ie what masked their view of the airspeed indications ?).

[goingmach_1]

I have read this entire thread and there is one thing I can tell you all. You were not there, not in the cockpit. As anyone can attest, flying any airplane is very dynamic. Can you imagine a complex machine like this? A almost too competent crew. Each one respecting the other and expecting them to do the right thing. Nobody speaks up because they expect that whoever is flying the machine will will correct the situation. They all have seen worse and turned a bad situation into a positive outcome. Surely on a beautiful screaming VFR day, whoever is flying will put it on no problem? A lot of assumptions going on.

For those of us who come off the "classics" i.e.: no auto throttles or minimal automation, can attest to, automation is another layer of management. Perhaps even too much management. Back in the day we flew stick and rudder every day. We had our hands wrapped around that machine completely. No button pushing going on. Well a little help, like a bare bones flight director. Point is you were totally involved in the situation from the start to the finish. I mean you drove the thing from start to finish and were attuned to any deviation along the way. If it did not feel right, thats because it was not right, and you corrected. Not confused by, what button to push? If it started to slow, you added throttle. If you were too high, you corrected. Too low you did something about it, immediately. Stick and rudder stuff.

Recently I have transitioned to a very modern push button kind of airplane. Point is after 20,000hrs of jet PIC, I feel left out of the loop. "Must push buttons!" was the message I got at the training centre. Pushing buttons will fix it.

Well what happens when you push the wrong button? Like a FMS sequence key to enable a very complicated VNAV arrival, or departure in Europe. One little F'up there and the automation aint going to work. Back in the day, we would manage that just fine, thank you very much. Like I said. Flying is very dynamic.

This accident was very dynamic.

[Diadem]

I have read this entire thread and there is one thing I can tell you all. You were not there, not in the cockpit. As anyone can attest, flying any airplane is very dynamic. Can you imagine a complex machine like this? A almost too competent crew. Each one respecting the other and expecting them to do the right thing. Nobody speaks up because they expect that whoever is flying the machine will will correct the situation. They all have seen worse and turned a bad situation into a positive outcome. Surely on a beautiful screaming VFR day, whoever is flying will put it on no problem? A lot of assumptions going on.

For those of us who come off the "classics" i.e.: no auto throttles or minimal automation, can attest to, automation is another layer of management. Perhaps even too much management. Back in the day we flew stick and rudder every day. We had our hands wrapped around that machine completely. No button pushing going on. Well a little help, like a bare bones flight director. Point is you were totally involved in the situation from the start to the finish. I mean you drove the thing from start to finish and were attuned to any deviation along the way. If it did not feel right, thats because it was not right, and you corrected. Not confused by, what button to push? If it started to slow, you added throttle. If you were too high, you corrected. Too low you did something about it, immediately. Stick and rudder stuff.

Recently I have transitioned to a very modern push button kind of airplane. Point is after 20,000hrs of jet PIC, I feel left out of the loop. "Must push buttons!" was the message I got at the training centre. Pushing buttons will fix it.

Well what happens when you push the wrong button? Like a FMS sequence key to enable a very complicated VNAV arrival, or departure in Europe. One little F'up there and the automation aint going to work. Back in the day, we would manage that just fine, thank you very much. Like I said. Flying is very dynamic.

This accident was very dynamic.

I would say that a crew member who doesn't speak up when the PF is obviously not flying correctly because of the assumption that the deviation will be corrected is hardly a competent crew member, let alone "too competent". If you can't even tell the person next to you that they're doing something wrong, what business do you have acting as PNF? Do you even have a role at that point, or have you just relegated yourself to the position of ballast? What's more, if you don't know the aircraft well enough to operate it safely, you're not competent; if I overtemped an engine because I didn't understand the system, then the argument that it was too complex for me would get me no relief from blame.

[Eric Janson]

I have read this entire thread and there is one thing I can tell you all. You were not there, not in the cockpit. As anyone can attest, flying any airplane is very dynamic. Can you imagine a complex machine like this? A almost too competent crew. Each one respecting the other and expecting them to do the right thing. Nobody speaks up because they expect that whoever is flying the machine will will correct the situation. They all have seen worse and turned a bad situation into a positive outcome. Surely on a beautiful screaming VFR day, whoever is flying will put it on no problem? A lot of assumptions going on.

Due to the unique nature of the operation all Captains at my Airline are qualified to sit in the right seat as PNF. I frequently do flights where all 3 of us are Captain and the designated Commander may well be the relief Pilot for the sector. It is far from an ideal situation and may lead to a situation mentioned above. I make it a point in my briefing that I expect everyone to speak up - safety first.

Just out of interest we added up our experience on a recent flight - 85 years of flying experience between the three of us!

For those of us who come off the "classics" i.e.: no auto throttles or minimal automation, can attest to, automation is another layer of management. Perhaps even too much management. Back in the day we flew stick and rudder every day. We had our hands wrapped around that machine completely. No button pushing going on. Well a little help, like a bare bones flight director. Point is you were totally involved in the situation from the start to the finish. I mean you drove the thing from start to finish and were attuned to any deviation along the way. If it did not feel right, thats because it was not right, and you corrected. Not confused by, what button to push? If it started to slow, you added throttle. If you were too high, you corrected. Too low you did something about it, immediately. Stick and rudder stuff.

Recently I have transitioned to a very modern push button kind of airplane. Point is after 20,000hrs of jet PIC, I feel left out of the loop. "Must push buttons!" was the message I got at the training centre. Pushing buttons will fix it.

Well what happens when you push the wrong button? Like a FMS sequence key to enable a very complicated VNAV arrival, or departure in Europe. One little F'up there and the automation aint going to work. Back in the day, we would manage that just fine, thank you very much. Like I said. Flying is very dynamic.

This accident was very dynamic.

The basics of flight remain the same no matter what you are flying. Where training falls short is teaching the appropriate level of automation to use in any given situation. Training is increasingly the minimum amount to meet the regulations. Far too little attention attention is paid to basic flying imho.

No matter how advanced the aircraft you need to be ready to intervene if things do not go as expected. This includes disconnecting the automatics and flying manually. This is clearly stated in both the airbus and Boeing Flight Crew Training Manuals.

It is disgraceful that Asiana are trying to blame Boeing when there are clearly cultural issues and a lack of basic flying skills involved.

It will be interesting to see what appears in the final report.

[PilotDAR]

You were not there, not in the cockpit. As anyone can attest, flying any airplane is very dynamic. Can you imagine a complex machine like this? A almost too competent crew. Each one respecting the other and expecting them to do the right thing. Nobody speaks up because they expect that whoever is flying the machine will will correct the situation. They all have seen worse and turned a bad situation into a positive outcome.

This might have elements of an excuse, if the aircraft had suffered a massive failure (both engines quitting short final - LHR). I do not accept that even one pilot, let alone four? could let things get this far out of hand, and then blame the automation.

I can't imagine that complex a machine, no, I have never flown anything like it. But, aside from a few sailplanes, everything I have flown goes up when you push the power up, as long as your good judgement gets it in time. Good judgement failed here, I don't care how complex or simple the plane. Many times as PNF, I have spoken (and to much more senior pilots) to assure awareness of something I thought I was beginning to observe. So far, so good.

Pay attention all the time, and catch it early, speak slowly and respectfully, and be prepared to have it explained why you were over reacting, and it's all okay = good. Speak late in a panic, and just catch it in time, everyone rattled = adequate. Don't speak and crash = inadequate. Fail to use the most basic of flying skills, no matter what aircraft type, and crash = woefully inadequate.

The basics of flight remain the same no matter what you are flying. Where training falls short is teaching the appropriate level of automation to use in any given situation. Training is increasingly the minimum amount to meet the regulations. Far too little attention attention is paid to basic flying imho.

No matter how advanced the aircraft you need to be ready to intervene if things do not go as expected. This includes disconnecting the automatics and flying manually. This is clearly stated in both the airbus and Boeing Flight Crew Training Manuals.

It is disgraceful that Asiana are trying to blame Boeing when there are clearly cultural issues and a lack of basic flying skills involved.

+1

[pdw]

... this far out of hand,

How far ?

At 400ft it is still looking OK, except that the speed decrease has suddenly gone too far by 350ft agl. At that point the speed deficiency needed recognition. There were only two pilots ... right ?

Recognition of a rapidly-approaching minimum speed missed by five seconds (here the miss-point arrives doubly fast as positive headwind lost during 1200fpm descent actually goes 'negative') already means that the spool up time has become much less affordable inside a rapid airspeed decrease timeframe. Throw in some confusion of the misunderstood auto-throttle setting (ie more seconds lost) and it's no surprise it too joins the blame game. If these two pilots are freshly woken up for the landing I can see the brightness of flying it into sun could have affected things and also worth mention.

Also, regarding a shorter decay period experienced here. In my neighbourhood recently a policeman observes a local driver go through yellow that turns red as she was in the middle of the intersection entering traffic onto the main highway that goes through our town. When it finally got to court, the lady explained she then spent 2 days observing the length of the yellow and that there was a big difference in the expectation of its duration from one light to the next . The judge was so impressed with the data provided ... a nearly two second difference on an average 4 second yellow that the case was dismissed. Here in the San Fran accident in VFR, just saying, perhaps a similar seriously shortened/unexpected timing (short/aggressive decay period) along with other accident sequence factors presented in/by the various posts/posters ... training etc.

[cdnpilot77]

What?? Perhaps they just fu$&ed up a visual approach. They didn't have the hand flying skills despite how many hours are in their log book. They failed at the very fundamental task in aviation of taking off and landing an airplane safely. That's it, no traffic lights, no mystery voodoo whatever...they at that moment, were incompetent.

[Illya Kuryakin]

So, after 17 pages, ya'll figure they pooched it? NSS!
Illya

[cdnpilot77]

So, after 17 pages, ya'll figure they pooched it? NSS!
Illya

What an insane theory!!! There MUST Have been other intergalactic forces playing tricks on the hearts and minds of the pilots.

[Illya Kuryakin]

So, after 17 pages, ya'll figure they pooched it? NSS!
Illya

What an insane theory!!! There MUST Have been other intergalactic forces playing tricks on the hearts and minds of the pilots.

Indeed. pdw, an ill wind perhaps?
Illya

[trey kule]

At 400ft it is still looking OK, except that the speed decrease has suddenly gone too far by

Ah...PDW....things were not still looking OK if your speed is decreasing...and it was not instant or suddenly...there typically actually is a nifty little trend thingie that tells you if your speed is decreasing or increasing...not sure about this particular aircraft. Speed (AOA). Is kind of one of those pesky little things that should get your attention. Even at landing weights if the engines are spooled or spooling down, things could get interesting if you suddenly need power.

[pdw]

Framed it real smartly there, thanks.

[golden hawk]

http://news.airwise.com/story/view/1415969797.html

Asiana To Halt San Francisco Service For 45 Days
November 14, 2014


South Korea's transport ministry has decided to suspend Asiana Airlines' Incheon-San Francisco route for 45 days in relation to the plane crash in 2013.

An Asiana Airlines Boeing 777 crashed at San Francisco Airport in July last year, killing three passengers and injuring more than 180.

Asiana, which flies the route daily, is expected to lose about KRW15 billion won (USD$13.64 million) in revenue as a result of the suspension, Korean wire service Yonhap reported without citing a source.

(Reuters)

Nothing like quick, decisive action!

[B-rad]

Information from another thread that is worth linking to this one.

http://www.flightglobal.com/news/articl ... rs-406268/


DOT requires 777 operates to inspect pitch rate sensors
By: JON HEMMERDINGERWASHINGTON DC Source: in 2 hours
A new airworthiness directive issued by the US Federal Aviation Administration requires Boeing 777 operators to inspect and potentially replace the aircraft's dual pitch rate sensors (PRS).

The directive, which applies to 777-200LRs, -300s, -300ERs and 777 freighters, comes in response to reports received by the agency of dual failures of the sensors.

When both sensors fail, the aircraft's primary flight computer transitions from primary to secondary mode, which causes the autopilot to disengage, says the directive, made public by the US Office of the Federal Register on 19 November.

"We are issuing this AD to prevent a dual PRS failure that could cause an automatic disengagement of the autopilot and autoland," says the DOT's directive.

Losing both autopilot and autoland can be dangerous, particularly if it "occurs at low altitude and the flight crew is unable to safely assume control and execute a go-around or manual landing", adds the document.

The directive will be published in the Federal Register tomorrow and becomes effective 35 days later.

The rule will require US operators to inspect, within 60 months, all four PRS on their 777s to determine the sensors' part number.

Then, prior to further flight, operators must replace any PRS with part number 402875-05-01, says the directive.

DOT estimates airlines will spend $170 to inspect each aircraft and $340 to swap parts.

The new requirements come roughly 16 months after a Boeing 777 operated by Asiana crash in San Francisco after the pilots failed to understand complexities of the autopilot systems.

The National Transportation Safety Board pegged crew failures as the ultimate cause, but added that the pilot falsely assumed that the autothrottle was maintaining speed.

The aircraft's speed deteriorated when the autothrottle transitioned to "hold" mode while the flight computer was in "flight-level change mode", a configuration under which the autothrottle does not maintain speed.

The crash led to the death of three passengers.

[Eric Janson]

Information from another thread that is worth linking to this one.

http://www.flightglobal.com/news/articl ... rs-406268/

Not related to this accident in any way as far as I can see.

"We are issuing this AD to prevent a dual PRS failure that could cause an automatic disengagement of the autopilot and autoland," says the DOT's directive.

Losing both autopilot and autoland can be dangerous, particularly if it "occurs at low altitude and the flight crew is unable to safely assume control and execute a go-around or manual landing", adds the document.

There is nothing particularly difficult about a manual landing or a manual go-around from low altitude - provided you have some basic flying skills.

What has this Industry become?

[B-rad]

Not related to this accident in any way as far as I can see.

I provided a reference link and copied out the article and still you reply without even reading the damn thing.
I know the one page might be a long read for you but perhaps read the whole article. Better yet, I'll show you the connection so it's easier for you to understand. There is mention of some human factors that are connected to this AD specifically in the Asiana San Francisco flight.

The new requirements come roughly 16 months after a Boeing 777 operated by Asiana crash in San Francisco after the pilots failed to understand complexities of the autopilot systems.

The National Transportation Safety Board pegged crew failures as the ultimate cause, but added that the pilot falsely assumed that the autothrottle was maintaining speed.

The aircraft's speed deteriorated when the autothrottle transitioned to "hold" mode while the flight computer was in "flight-level change mode", a configuration under which the autothrottle does not maintain speed.

The crash led to the death of three passengers.

Capisce?

[Eric Janson]

Not related to this accident in any way as far as I can see.

I provided a reference link and copied out the article and still you reply without even reading the damn thing.
I know the one page might be a long read for you but perhaps read the whole article. Better yet, I'll show you the connection so it's easier for you to understand. There is mention of some human factors that are connected to this AD specifically in the Asiana San Francisco flight.

The new requirements come roughly 16 months after a Boeing 777 operated by Asiana crash in San Francisco after the pilots failed to understand complexities of the autopilot systems.

The National Transportation Safety Board pegged crew failures as the ultimate cause, but added that the pilot falsely assumed that the autothrottle was maintaining speed.

The aircraft's speed deteriorated when the autothrottle transitioned to "hold" mode while the flight computer was in "flight-level change mode", a configuration under which the autothrottle does not maintain speed.

The crash led to the death of three passengers.

Capisce?

Let me try this again.....

You linked an article that refers to a failure mode causing the autopilot to disconnect. This was not what happened in the Asiana crash.

The article then goes on to discuss the autothrottle "Hold" mode - even though that is unrelated to the previous failure. For some reason the author tries to link two separate issues together.

In the Asiana crash the autopilot and autothrottle functioned as they were designed to do - the Pilots were not aware of what was going on.

For the record:-

I am perfectly capabable of reading and understanding what is written

I have 10 years experience on Boeing aircraft - I know what the autothrottle "Hold" mode is.

[Gannet167]

And to be a qualified "pilot," I know what the go fast levers do. These guys screwed up basic power vs attitude. Stuff that's taught on lesson number one in a C150. Remember that old chesnut: "Aimpoint, Airspeed, Aimpoint, Airspeed"?

Maybe the automation sh!t the bed. Maybe some mode screwed up. Maybe they didn't understand the automation. Maybe that's a contributing factor. It looks very much like culture played an unfortunate role as well. All of that sucks but it's a poor excuse for a guy with wings on his chest.

If you're a "Pilot" you should be able to see that the airplane isn't flying as it should be. That's when you get nervous and suspiciously start thinking to yourself "What the F$ck?" and about 0.5 seconds later you do something - anything - but probably hit the disconnect button and ... wait for it ... FLY. Less worrying about what other guy thinks of me, more FLYing would have helped.

You can argue the details to death, and they're interesting to understand and learn from. But at the end of the day, a perfectly serviceable aircraft with a lot of very nice bells and whistles designed to make flying way easier, was flown into the ground by 3 guys who watched it happen and couldn't manage to do with their super-mega fancy airplane what any post-solo PPL student can do in a 50 year old Cessna.

[Cat Driver]

If one of them actually looked out of the windshield wouldn't he have wondered why the runway was way up there?

[Flying Low]

If one of them actually looked out of the windshield wouldn't he have wondered why the runway was way up there?

Actually they were so nose up the runway was probably way down there! Again looking out the window might have given a clue...

[pdw]

... but then again no ... looking out the left window there was glassy water to the PIC's left, as they are approximately two Kilometers lee of the left-hand shore ... where to their right is only bay.