News Asiana 777-200 Crash on landing at SFO

[Scav]

From Avweb:

http://www.avweb.com/avwebflash/news/Pilot_Called_Go_Around_220020-1.html

A pilot aboard Asiana Flight 214 called for a go-around 1.5 seconds before the tail of the aircraft struck a seawall off the end of Runway 28L and San Francisco International Airport Saturday morning. At a news conference Sunday, NTSB Chairwoman Debra Hersman said the cockpit voice recorder also recorded a crew member calling for more speed seven seconds before the aircraft struck the jetty, tearing off the tail and resulting in the deaths of two passengers. Also, CNN has obtained amateur video of the crash sequence Meanwhile, airport officials have confirmed the glideslope of the ILS system for Runway 28L at San Francisco International Airport wasn't working at the time of the crash. That means the Boeing 777's autoland system would not have been available to the crew and they would have had only localizer guidance to the runway. Visibility was unlimited and winds were light when the 777 touched down about 1,000 feet before the normal landing point, leaving its horizontal and vertical stabilizers on the threshold before careening off the runway. Two 16-year-old Chinese girls were killed and more than 180 others were hurt, about 50 of them seriously.



A NOTAM issued in June (scroll down to the 19th NOTAM) advises the glideslope will be unavailable until Aug. 23. A pilot familiar with SFO (whose name we agreed not to publish) said the inoperative equipment has challenged many pilots who have grown accustomed to the electronics flying the approach, regardless of the weather. "It can be exciting for an airline crew to have to fly a LOC approach with step downs and everything, after not having done anything but ILSes for the last gazillion approaches," he said.

 

[Screamer7]

Check NTSB Twitter. They are posting live feed of the ongoing investigation:

https://twitter.com/NTSB

 

[n122vu]

News from Reuters that the pilot who was flying the jet was in training, experienced pilot, but only had 43 hrs in the 777:
http://www.reuters.com/article/2013/07/08/us-asiana-korea-idUSBRE96701620130708

 

[Andy44]

There is news that the coroner is investigating to see if one of the two dead passengers was run over by an emergency vehicle responding to the crash:

http://news.yahoo.com/official-probes-rescuers-ran-over-crash-victim-235843380.html

 

[Suzy]

A lot of serious injuries for the survivors: Spinal fractures a common injury after Asiana plane crash

 

[Andy44]

A lot of serious injuries for the survivors: Spinal fractures a common injury after Asiana plane crash

Not surprising after seeing that video of the plane bouncing up and slapping back down. If you were at the upper end of the fuselage when it slammed down onto the pavement you'd be hurting.

I have never felt safe with just lap belts, but I am a pilot and I think too much about things like that. If I had my druthers I'd be wearing a five-point harness with a head restraint...

 

[garyw]

I have never felt safe with just lap belts

Even with the deceleration of a normal landing you'll get thrown forward because the only thing holding you in place is the seat belt. I'd prefer something a little sturdier myself.

This is looking more and more like human factors with poor CRM and revision to previous aircraft types under the high workload of the manual approach after flying for 10 hours+

Asiana spokeswoman Lee Hyomin said that Lee Gang-guk was trying to get used to the 777 during Saturday's crash landing. She says the pilot had nearly 10,000 hours flying other planes but had only 43 hours on the 777.

I wonder if that's the Captain rather than the pilot flying? even 10,000 hours for a captain sounds a bit low. Most AAIB reports list the captain with 20,000 hours....

The NTSB have released quite a few photos:

 

[Screamer7]

Is this crash not related to the crew become to independant on the aircraft ILS system?
I watch a lot of landings on You Tube, and what fascinating me the most is the CAT 3 c landings in zero visibility.
The autopilot actaully disconnect when they are already on the runway.
I know the visibilty at the time of the crash was good, but maybe they get out of practise landing the plane with only the localizer aviable at that time.

 

[garyw]

The autopilot actaully disconnect when they are already on the runway.
I know the visibilty at the time of the crash was good, but maybe they get out of practise landing the plane with only the localizer aviable at that time.

Quite probably but if this was a step down approach those procedures are also available in the FMC and the aircraft can fly most of the approach itself even without an ILS.

I wonder if SFO did a late switch to the approach process and the crew had prepared for one approach but decided to press on with the second anyway then allowed the plane to get ahead of them.

 

[Screamer7]

That is how cruel life can be sometimes.
Just one moment of lack of concentration can alter your life forever.
No second change.

 

[Urwumpe]

Quite probably but if this was a step down approach those procedures are also available in the FMC and the aircraft can fly most of the approach itself even without an ILS.

It can't do that well. The localizer beam was there, but the glideslope was deactivated. So, it had horizontal guidance, but the pilot had to fly the aircraft in the vertical.

Aircraft can not accurately deduce their position relative to the runway by GPS (thus the idea to install GBAS/LAAS) and the DME distance by the localizer beam is much more inaccurate than GPS. Radar altitude can tell you how high you are and how fast you are dropping, but not how high you should be. And if the plane was flying to slow, the nominal descend rate for landing would have been too steep.

It looks like the crash was caused by pilot error and bad crew management, especially by being too slow on final approach for quite a long time during the approach.

Calling for go-around 1.5 seconds before crash is too late anyway for such a large plane, that the engine shad been idling made things worse (If I remember correctly, such engines need almost 4 seconds to go from idle to TO/GA power). Even a fighter would continue on its path for almost a second before the slowly reacting engines provided enough work to overcome the inertia.

But otherwise, it had been perfect conditions for a visual landing in SFO. Much better than the usual conditions.

And if one of the victims was really killed by an ambulance car, I really hope some heads will be rolling there. It is part of the ambulance training to avoid parking on the victims.

 

[RGClark]

Good grief... the crew really did stall it on approach. Now the question is one of 'how the hell did a trained crew stall a plane on approach'.
I have a feeling that this crash is going to become a study in CRM and communication.

Perhaps the pilot at the controls, who was in training on 777's, was the more senior pilot in the company? In that case the more experienced pilot who was conducting the training might defer to him anyway.

Bob Clark

 

[Urwumpe]

Perhaps the pilot at the controls, who was in training on 777's, was the more senior pilot in the company? In that case the more experienced pilot who was conducting the training might defer to him anyway.

Bob Clark

As far as it sounds, the pilot flying the landing was no real veteran and still in training for this aircraft. He had 10,000 hours of flight experience mostly on the Boeing 747-400, but only 43 for the 777. The 747-400 is a completely different aircraft in that context despite being in the same weight class, especially it lacks the fly-by-wire that the 767 and 777 has ve( The -400 only uses glass cockpit technology, but retained the control systems and autopilots). It was the first landing in SFO for the pilot with a 777.

PAPI had also been deactivated in SFO. Also, pilots complain about the controllers at SFO, that they constantly direct them with too high descend rates to the landing, probably for noise reduction.

 

[Fabri91]

IIRC the 777 was Boeing's first FBW aircraft, with the 767 (and its smaller twin, the 757) having conventional controls.

A contributing factor might be the fact that outside of North America a visual approach is very much non-standard, or at least it is in Europe and Asia.

 

[Urwumpe]

IIRC the 777 was Boeing's first FBW aircraft, with the 767 (and its smaller twin, the 757) having conventional controls.

Yes, I also just read it. The 767 was the first with glass cockpit. I thought Boeing had reacted faster to Airbus with the A320. After all, it had been 10 years from A320 to 777.

A contributing factor might be the fact that outside of North America a visual approach is very much non-standard, or at least it is in Europe and Asia.

What are the differences?

Sullenberger, the pilot who ditched in the Hudson River, also noted that the visual approach over the water is much harder since there are no usual visual cues to see the altitude for the pilot.

 

[mojoey]

Sullenberger, the pilot who ditched in the Hudson River, also noted that the visual approach over the water is much harder since there are no usual visual cues to see the altitude for the pilot.

Aside from the water skier now embedded in the windscreen...

 

[Urwumpe]

Aside from the water skier now embedded in the windscreen...

And the sudden engine failure during pre-flare because of herring strike.

 

[mojoey]

And the sudden engine failure during pre-flare because of herring strike.

Attention passengers, in a few moments the flight attendants will be serving fresh herring fillets. Please be mindful of the tiny metal shards and fish bones though.

 

[Fabri91]

What are the differences?

The problems stemming of a possible excessive reliance on automation during flight, and especially during approach, might be exacerbated by the fact that operating procedures or company policies to which the pilots were accustomed treated visual approaches like something to be avoided at almost any cost.

ATC Recording
2:07 initial call
3:12 Asiana flight calls short final

 

[garyw]

So, it had horizontal guidance, but the pilot had to fly the aircraft in the vertical.

It's still possible to fly a VNAV coupled with the FMC flying the approach plate STAR. Of course, SFO controllers will probably have requested a change to the published approach which caused a high workload on the flight deck.