Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

Asiana flight crash at San Francisco

Wikiposts
Search
Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

Asiana flight crash at San Francisco

Thread Tools
 
Search this Thread
 
Old 12th Jul 2013, 14:49
  #1861 (permalink)  
 
Join Date: Jul 2013
Location: Aidensfield
Posts: 6
Likes: 0
Received 0 Likes on 0 Posts
jientho 12th Jul 2013, 13:49:
This seems to be a classic case of lack of real-world awareness in software design. What POSSIBLE real-world condition would justify allowing a flight-level-change target of FL000?
There is no such condition. Nobody does that.

Airbus works just same way on "open descend" mode.

It is possible to reduce speed below lowest acceptable when flying manually against FD guidance with both Boeing and Airbus, if autoflight computer guidance is on FLCH or open descend mode, that is not Boeing-specific thing.

That is not the case here, I'm pretty sure of that. Nobody selects 0 feet in alt selector in any situation. Never. That happens only in wannabe pilots' nightmares, never in real world.

Let's wait until we know what really happened. So far we only know that the approach was unstable, "hot and high", and that the control of the plane was not in acceptable level when the plane finally went through glide slope with reducing speed and engines on idle.
AF Eagle is offline  
Old 12th Jul 2013, 14:53
  #1862 (permalink)  
 
Join Date: Jun 2001
Location: Rockytop, Tennessee, USA
Posts: 5,898
Likes: 0
Received 1 Like on 1 Post
I hope the quote by Lorimer from Deborah was inaccurate - "This was a visual approach which is a completely normal approach flown on a nice day with 10 mile visibility." - I would not classify a 'request' for 3-400 ft high and several kts fast at 5 miles to be 'a completely normal approach' and I hope the NTSB comment on this SFO procedure. I await the transcript/video of the conference. It is certainly another of those 'holes in the cheese'.
You can see Debbie (as she usually introduces herself at the podium) respond to a final question about automation starting at about 27:20 in this video:

Raw Video: NTSB Final Update On Asiana 214 Crash Investigation - Pt. 2 « CBS San Francisco

Her comment about a manually flown approach being normal and intended coming into an airport like SFO with the glideslope out on a clear, 10 mile visibility day is at 29:10.

You will have to let the video load for a minute or two before you can view the final minutes of the clip. Also, if you don't stop the video prior to the very end you will have to recue the clip starting with the commercial to review part of it.
Airbubba is offline  
Old 12th Jul 2013, 14:56
  #1863 (permalink)  
 
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes on 0 Posts
I find it hard to imagine a justification for bestowing "qualified" status to any position that would be at the controls of any aircraft that did not include the ability to shoot a visual approach in VMC,

Agree. However the requirement for test would need to be instigated by XAA's and they don't. Further more the airlines actively discourage visual approaches as crews without training and practice tend to screw up and cost time and money. The gulf in 'piloting' ability between the major airport iLS jockeys and the charter carrier Greek island visual pilots is greater than it should be. I used to fly for one of the 'discouraging airlines. Occasionally an F/O would ask permission on a severe clear day to conduct a downwind visual circuit to an ILS. Be my guest. The ensuing horlicks was interesting. I give credit that they wanted to try; they explained that with so little training or practice it was difficult etc.
The TQ courses designed by the airlines and approved by the XAA's are both good & bad at the same time. They do not produce pilots and their line flying does not either. I reckon that give a modern young pilot a visual approach from 20nm 6000' 210kts on a severe clear day, with no ILS or PAPI's they could not do a CDA and be stable at 1000'. I doubt many new young captains could either.
The same true for a descending visual circuit from 3000' abeam. I find that very very sad; indeed perhaps unacceptable. I will not speculate on the pax opinion. It is a sad state of affairs and XAA's and airlines must take some responsibility.

The engines would need five seconds or more to produce significant thrust

I can't remember from my CF-6 days, but on CFM-56's I used to suggest a min of 40% N2 on finals, for instant response, but if idle was needed to ensure that you had 40% when your speed was target +5kts at the latest, again to give instant response to freeze you at target speed when it was reached.

These protections activate only in situations where the correct control of the plane has been lost, alfa floor activates when the plane approaches stall AOA.

I'm not a FBW pilot, but is this not similar to the A320 crash at Basle? They descended onto the trees expecting the A/T to spool up. An Indian pilot did exactly the same thing a couple of years later and planted the a/c short of the runway, watching & waiting but doing nothing, as was the A320.

Capable, well trained and current truck drivers is the answer.......but who pays for the training?

The analogy of the truck drivers and the automated parking system is excellent. Money talks: Who pays for the training should not matter. If the training is necessary for the qualification then it gets paid by necessity. How, is a debate, but where we are now is that the training is so expensive, and pilots are self-funded, that basic elements have been chopped. Even the basic CPL is 100hrs less than 40 years ago. Sims can do some IR training, but not all. Just watch the ladies in the streets trying to park the SUV without parking sensors. Good game, but we are discussing something more critical than a dented shopping car.

Three pilots let a perfectly good B777 fly into the rocks because they didn't know how to fly an airplane. They were trained to program a computer, not fly an airplane.

In many TQ courses I've given the FBS is dedicated to systems knowledge and operation, both normal & non-normal and all on autopilot, except for takeoff. 20hrs of button pushing. FFS then commences with a token 2hrs of manual flying ticking the boxes of turns, stalls, buffets, FD ILS's & raw data ILS's. Once that's done the next 32hrs of PF/PNF is taken up with autopilot operation, but manual landings, while various QRH items are ticked off and the accompanying emergencies are also ticked off. Thus the total manual flying in a TQ course, before base training, is 2-4 hrs. assuming a PCST session of 1 hr. On the line it is autopilot all the time. They think it is 'sportive' to disconnect on an ILS at 2000' if visual; or perhaps 1000' if a x-wind. Airlines often prohibit visual circuits with finals <4nm, yet base training used 800'. Not only that, but they want an LNAV/VNAV guidance to he visual finals. Give me strength!
In one TQ course a very intelligent lad, and well prepared in his studies, sharp and very computer literate: the a/c was not doing what he wanted on autopilot. I suggested LVLCHG or HDGSEL would help solve the problem quickly. No; he was straight into the FMC and dancing over the keys with gay abandon. Head down and wanting Bill Gates/Mr. Mac to give him guidance. I suggested rather disappointedly that as he wanted to be a pilot perhaps he might like to fly the a/c out of the problem; or at least use the MCP directly to have control. This was not impressive. In 5 years he and his ilk will be captains teaching new wannabes. More AGH! & Ouch! Get me out of here.

But surely there is a massive wake-up call here. I find it trite to read all the ergonomists and social behaviour gurus talking about 'training pilots in different skills; humans are bad automation monitors etc; the interface of human and automation needs to be better understood; perhaps even that the pool of pilots should come from a different back ground than previously'. All fine and dandy, but that does not excuse an industry from failing to train a pilot to fly an aeroplane. Once you can fly it you then learn how to operate it. The flying bit has been by-passd and needs recovering. Base training is insufficient.

Last edited by RAT 5; 12th Jul 2013 at 15:09.
RAT 5 is offline  
Old 12th Jul 2013, 15:44
  #1864 (permalink)  
 
Join Date: Mar 2002
Location: Surrey, UK
Posts: 897
Received 10 Likes on 6 Posts
They were trained to program a computer, not fly an airplane.
Were they? It sounds more like they learned to use one, badly. The degree of systems knowledge a programmer might have seems to be lacking.
steamchicken is offline  
Old 12th Jul 2013, 15:46
  #1865 (permalink)  
 
Join Date: Nov 2005
Location: Potomac Heights
Posts: 470
Likes: 0
Received 0 Likes on 0 Posts
Time for response

One point that I have not seen adequately discussed is the amount of time required for emergency services to respond to the crash. Some have suggested that it seemed to take an inordinately long time. But I believe the video in Post 1819 provides the answer.

San Francisco Boeing 777 Plane Crash - HD 7/6/13 (Part 1) (Actual Crash Footage) - YouTube

You will note that at about 7 seconds from the beginning of the video, the escape slides are deployed. We know that this occurred roughly 90 seconds after the crash. You can see that at about 1:30 into the video, the fire trucks begin to arrive. This suggests a total time of 3 minutes from crash to emergency arrival. Given that the crash took place without warning and at the far end of the runway by threshold 28L, almost 2 miles remote from the nearest fire stations, to me this suggests a pretty immediate response.

See KSFO map at http://aeronav.faa.gov/d-tpp/1307/00375AD.PDF
SeenItAll is offline  
Old 12th Jul 2013, 16:04
  #1866 (permalink)  
 
Join Date: Jul 2013
Location: Aidensfield
Posts: 6
Likes: 0
Received 0 Likes on 0 Posts
Money talks: Who pays for the training should not matter. If the training is necessary for the qualification then it gets paid by necessity.
FAA begins to require all new airline pilots at least 1500 flight hours before they are allowed to fly an airliner. That is a lot, requirement used to be 250 hours.

Meanwhile in Europe... ICAO Multi-Crew Pilot Licence is implemented here. That means new pilots need to have at least 240 hours flight time, and as far as I understand, approximately two thirds of that can be flown in flight simulator.

What does that mean? That means the airlines want to get pilots with as cheap training as possible, and European Union has listened to them and accepted this.

USA was the first continent to face the changing world of the airline business, towards low-cost operations and low-cost pilots. This new 1500 hour minimun requirement is a large step back, now it is almost impossible to begin your career as airline pilot in USA, and you won't have the cheapest pilots there anymore.

But we get cheaper and cheaper pilots here in the other world. So you can say whatever you want about necessity of expensive training, but money talks. Airlines lease their pilots via contract agents that work in countries with laws and taxes that are cheapest and easiest for the airlines. The most important thing in recruiting is that the work force is cheap.

The reputation of Asian airlines is such that they need to pay more than the others for Western pilots. Western pilots are not loved members of the work force in Asian carriers, I believe, and both Asian pilots and management of the companies want to get their own people flying their planes as soon as possible.

Maybe you should close the borders and not accept non-American airlines to fly there at all? You can recruit the best pilots from other world after they have flown their first 1500 hours somewhere else. But be careful, we pilots in the other world are going to be worse and worse, you should really think twice before accepting us to fly in your airspace.
AF Eagle is offline  
Old 12th Jul 2013, 16:15
  #1867 (permalink)  
 
Join Date: Jul 2013
Location: CA
Posts: 16
Likes: 0
Received 0 Likes on 0 Posts
jolihokistix
If Asiana pilots really want to gently practice visual hands-on approaches coming in over water, they have the best opportunities and the best place in the world to do it, Incheon
EXCEPT Incheon ATC probably doesn't give out approach instructions like the ones routinely given out by SFO. Seems like many foreign pilots flying into SFO have called SFO ATC's approach instructions as 'stupid'.
dba7 is offline  
Old 12th Jul 2013, 16:21
  #1868 (permalink)  
 
Join Date: Jul 2013
Location: Cohoes, NY
Posts: 23
Likes: 0
Received 0 Likes on 0 Posts
jientho 12th Jul 2013, 13:49:
Quote:
This seems to be a classic case of lack of real-world awareness in software design. What POSSIBLE real-world condition would justify allowing a flight-level-change target of FL000?
There is no such condition. Nobody does that.
Let me clarify -- I am not saying that FL000 setting happened in the Asiana approach (as you say, we shall see). I was referring to the capability for a pilot to set such a thing (and the system to then "use" it). Maybe I'm wrong, but I understand from comments here that that capability does exist on the 777? And as anyone from the systems world knows, anything that can be input by "users" must be assumed will be input by someone somewhere sometime, especially in life-critical systems such as those used in transport aircraft. Now, you seem to agree that there is NO condition where it makes any sense, so WHY IS IT ALLOWED BY THE SYSTEM to be entered if there is risk (even the tiniest bit)? That is my point.
jientho is offline  
Old 12th Jul 2013, 16:29
  #1869 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
Thanks for the video link, airbubba. Again an impressive performance by Deborah. It would seem that the quote by Deborah from Lorimer was a generic for 'visual approaches' and did not specifically refer to this SFO visual approach which I trust the NTSB will look at for suitability.

I have to say that the Boeing airframe appears to have survived remarkably well. The news that the floor had collapsed behind door 4R is not pleasant and it is a wonder how folk survived in that area, since I assume the seat supports had gone as well. There appear to be a significant number of seat rows behind doors 4..

The support from other agencies including UAL appears to have been excellent.
BOAC is offline  
Old 12th Jul 2013, 16:30
  #1870 (permalink)  
 
Join Date: Nov 2009
Location: Cambridge, UK
Age: 56
Posts: 7
Likes: 0
Received 0 Likes on 0 Posts
Slightly out of the general flow, but I found this interesting - it's a Korean's view of Malcolm Gladwell's view of Koreans. Indirectly linked (i.e. you'll need to click through to the main article from the following link) as the forum wouldn't allow the original link to be posted.. Gladwell's Culture & Air Crashes Analysis Badly Flawed - Slashdot
DaveKnell is offline  
Old 12th Jul 2013, 16:35
  #1871 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,197
Received 393 Likes on 244 Posts
dba:
Interesting point there.

It seems to me that each airfield has its own interesting features, be they geographic, navaids, routing, approaches, runway/taxiway, NOTAM that day, or ATC sequencing habits. Whatever. As a professional, I'd think that if one's destination is deemed by one's company (or by one's own experience) to be a difficult one, that becomes a briefing item, and a prep item. In a good organization, it's added to the company "smart book" for the best ways to deal with all the odd ins and outs of a given destination or mission.

A few pages back, one of the pilots calls that approach the "threat briefing." I'd say he was on the mark. In a generic sense, it's part of the risk management feature of any mission briefing.
"Your mission, should you choose to accept it, is to get this group of passengers to their destination safely."

Those complaining about SFO being "stupid" strike me as stupid: one needs to brief the flight for what is, not for what one wishes it were.
It's why we brief and plan ... to reduce the number of things that can surprise us or go wrong during the flight.
FBW or not
Short or long haul
IFR or VFR
Pax or Cargo
domestic or international

Brief the flight, prepare the flight, and address risks to the flight, in order to get the most out of the flight and to stack the odds in favor of the flight coming off as "routine" regardless of how much effort it takes to make it look that easy.

From the AF 447 CVR, there was an discussion between the captain and one of his FOs on the temp not being as forecast at altitude, hence the scheduled climb to FL 390 not possible as planned ... a perfect example of how valuable the planning process is to executing the flight as one encounters it, not as one wished it were. (That the two FO's later made fatal mistakes in flying is another matter)
Lonewolf_50 is offline  
Old 12th Jul 2013, 16:37
  #1872 (permalink)  
 
Join Date: Mar 2010
Location: Often in Jersey, but mainly in the past.
Age: 79
Posts: 7,808
Received 135 Likes on 63 Posts
ATCO tiptoes in again.

What sort of procedure calls for 1800 ft at 5 miles, to achieve a <3º glide path, with a request to maintain 180 kts until that point?

Oh, SFO does … with 5 miles to go, you are now c. 40 kts fast and already above the GP.
As a radar controller, if I ever demanded someone do that, my ears would burn.
And then the FD is expected to stabilise?

May I ask who signs off the plates for a visual to SFO 28L?
Regardless of the ins and outs of this case, that seems to me to be … ummm … stupid?

Last edited by MPN11; 12th Jul 2013 at 16:44.
MPN11 is offline  
Old 12th Jul 2013, 16:47
  #1873 (permalink)  
 
Join Date: Jul 2013
Location: on an island
Age: 81
Posts: 27
Likes: 0
Received 0 Likes on 0 Posts
Lonewolf 50

Brief the flight, prepare the flight, and address risks to the flight, in order to get the most out of the flight and to stack the odds in favor of the flight coming off as "routine" regardless of how much effort it takes to make it look that easy.
Oh how many mishaps I have investigated where the real "slice of cheese" at the heart of causation was a poor or non-existent pre-flight preparation and briefing.

On that subject, our crusty old safety officer in primary flight training (WWII, Korea, just back from VN) said, "No flight should be considered routine until you are giving the post flight debriefing. Otherwise, you are setting yourself up for a mishap."
tilnextime is offline  
Old 12th Jul 2013, 16:49
  #1874 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,197
Received 393 Likes on 244 Posts
MPN, the captain is always in the position to decline, or say "unable."

As you will note from this thread, any number of Captains who post here do that when they get such a request.

You do raise a valid point.

After I looked at some back of the napkin math in terms of the time and distance, and deceleration to stable approach asked for in the 180/5 call, I'd have to ask you this.

Doesn't the controller usually know what kind of aircraft he is tallking to? (local or not, large medium small, etc)
Does a controller have a "smart book" or "smart card" with expected approach speeds for a given heavy handy at his station?

Is the controller taught to rely on a crew telling him if they can or not?

EDIT: tilnextime. Amen. The mishap investigation we did on some unbriefed formation flying (which ended in tears) remains one of the most depressing tasks I ever got involved in.

Last edited by Lonewolf_50; 12th Jul 2013 at 16:52.
Lonewolf_50 is offline  
Old 12th Jul 2013, 16:50
  #1875 (permalink)  
Longtimelurker
 
Join Date: Nov 1998
Location: killington Vt
Posts: 391
Likes: 0
Received 0 Likes on 0 Posts
Seth
Funny how different airlines do things. You are terminating somebody for something I teach on OE. Mind you MCP alt alert is set 1000AGL or at FAF alt if IMC and we would never want to see use of V/S down low. Our AB guys do same thing too.

Last edited by filejw; 12th Jul 2013 at 16:52.
filejw is offline  
Old 12th Jul 2013, 17:05
  #1876 (permalink)  
 
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes on 0 Posts
RAT 5:
The engines would need five seconds or more to produce significant thrust

I can't remember from my CF-6 days, but on CFM-56's I used to suggest a min of 40% N2 on finals, for instant response, but if idle was needed to ensure that you had 40% when your speed was target +5kts at the latest, again to give instant response to freeze you at target speed when it was reached.
I submit you probably mean 40% N1 (fan rpm).

Core speed N2 at ground idle won't be less than about 60% for either CF6 or CFM56.

Last edited by barit1; 12th Jul 2013 at 17:07.
barit1 is offline  
Old 12th Jul 2013, 17:12
  #1877 (permalink)  
 
Join Date: Jun 2001
Location: Surrounded by aluminum, and the great outdoors
Posts: 3,780
Likes: 0
Received 0 Likes on 0 Posts
Mind you MCP alt alert is set 1000AGL or at FAF alt if IMC and we would never want to see use of V/S down low. Our AB guys do same thing too.

Many ways to skin a cat, our Boeing procedures were MCP altitude set to GA altitude passing the FAF on a VOR approach/GS capture on an ILS...if MCP ALT was set to 1000' AAL then ALT CAP would occur if using VS when not locked on to a GS..Airbus had the advantage of FPA which lessened the workload and allowed for more precise flight path control
ironbutt57 is offline  
Old 12th Jul 2013, 17:50
  #1878 (permalink)  
 
Join Date: Jul 2013
Location: Aidensfield
Posts: 6
Likes: 0
Received 0 Likes on 0 Posts
Just to clarify something assuming FLCH on a Boeing is equivalent to OP DES on the Airbus: Airbus prohibits use of OP DES (on the A320 and I imagine across the board) if altitude in the FCU is set lower than 500' or MDA/H, whichever is highest.
Ok, that's good. Having flown both A and B aicraft I have never tested that, and I can't see any situation where I would use FLCH or open descend with selected altitude lower than FAF alt. And I didn't know there is such logic that you described in Airbus even though I believe I know the Airbus logic quite well.

Why don't they fix the bug in Boeing? Why don't they build planes that can't be flown to any unsafe conditions? Why is it possible to shut down all engines airborne? Why is it possible to switch off flight instruments and IRU's airborne? Why is it possible that both pilots open their seat belts at the same time if the plane is flown in severe turbulence?

Why do they build cars that can be driven straight to a stone wall?

It is not possible to build such automation that lets crew fly the plane in all possible cases, but prevents them to do anything that might lead the plane to danger. Understanding and controlling plane's energy level while flying an approach is still pilots' job, automation doesn't understand what lies ahead. There are many cases where the automation doesn't work as it should, and the automatic systems are very stupid if things don't go as they should. That's why the pilots are needed in the cockpit. And the pilots shouldn't be stupid or sleeping. This "FLCH trap" is not something that can ever happen in final approach if the pilot's have brain in their heads.

Last edited by AF Eagle; 12th Jul 2013 at 17:54.
AF Eagle is offline  
Old 12th Jul 2013, 17:57
  #1879 (permalink)  
 
Join Date: Jul 2013
Location: on an island
Age: 81
Posts: 27
Likes: 0
Received 0 Likes on 0 Posts
MPN11:
2. Controllers know what aircraft type they're controlling. They are not, surprisingly, trained to know what the aircraft/crew can/cannot do.
Not surprisingly at all. The Pilot in Command is the final authority on what a given aircraft and crew can or cannot do. ATC deals in general procedures, the PIC deals in determining if he and his crew can handle the aircraft to comply with those procedures. If not, it's the PIC's responsibility to request an alternative.

If automation is making some of the decisions, and ATC is making some of the decisions, what the hell are we pilots doing?

Last edited by tilnextime; 12th Jul 2013 at 18:00.
tilnextime is offline  
Old 12th Jul 2013, 18:30
  #1880 (permalink)  
 
Join Date: Jan 2008
Location: Atlanta, GA, USA
Posts: 349
Likes: 0
Received 0 Likes on 0 Posts
Confirmed - one victim hit by a responding truck

Police: Firetruck hit girl after Asiana Airlines crash

Inexcusable.
deSitter is offline  


Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.