Korean 777 Down in SFO

Did FLCH mode cause the crash?
IMAO. No. It's a good chance it was a causal factor. However I don't believe it in itself, was the cause.

I think it was pilot error. However I think automation dependency is part of the cultural problems. FLCH mode being part of the problem.

By cultural, I don't mean race or country; though those could be and sometimes are part of cultural problems.

By cultural, I mean the aviation community as a whole. The governing bodies, the manufacturers, the airline industries, and air crews.

While I think the pilot(s) are at fault, I don't think the buck should stop there. Though I suspect the pilots are at fault, they shouldn't be used as scapegoats. I think there's plenty of blame that should be spread much farther than the pilots.

Though I have little doubt that ambulance chasers will try to single out FLCH as a cause to try to put the blame on manufacturers as they are easier to sue and have deeper pockets.

I think to some extent automation can make things safer. However almost every solution, has a subset of unintended problems/dangers.
 
Last edited:
Don't become overly fixated on the "FLCH Trap" . I don't think it is as simple as blaming the "FLCH Trap" .

I suspect that the "FLCH Trap" is one of many things that are causal factors in the crash.

I think there are complex multiple causes and contributing factors to the crash.

Overall and most importantly I feel there is a cultural problem. While automation prevents a lot of accidents; it also introduces some dangers. I feel that automation is contributing to complacency and a lack of basic flying skills.

One way I put it is; pilots are letting the planes fly pilots, rather than the pilots flying the planes.

Allegedly according to flight data recorders, 8 seconds before the crash, the throttles were advanced. If I understand correctly one of the pilots advanced the throttles. If I understand correctly, this would have disengaged the autothrottle. Therefore from that point on I would think "FLCH" would be disengaged.

Allegedly according to flight data recorders, 7 seconds before the crash, a crewmember called out to increase speed.

I don't know how long it takes for these engines to spool up. But I would think 8 or 7 seconds would be enough time to get enough power to have prevented the crash.

I suspect that not enough power was added.

I was taught to largely land airplanes with my hands on the stick and throttle. I think the pilots in the crash were fixated on using attitude/pitch to control the flight path, thusly neglected/forgetting thrust.

I get the impression that many modern pilots have become addicted/dependant to the use of autothrottles and have become complacent with monitoring their speed and throttle. I get the impression that the airline industry often presses pilots into situations that encourage pilots to become auto throttle addicted/dependent. IE: Flying on the edge of flight envelopes to save fuel, time & money.

I get the impression that the *culture for landing large modern aircraft have fallen into the **routine of using hands the stick (yoke) only to land the planes. (As opposed to hands on stick and throttle landings)
*(pilots, airlines, governing agencies, and manufactures)
** (a trap of complacency and dependency).

CRM (Crew Resource Management) seems to have been lacking. Three pilots in the cockpit, yet the aircraft was allowed to get too slow and too low. An instructor with not much time on type of aircraft.

I think automation has increased aviation safety, however I think automation has inadvertently introduced some of its own dangers. There is no perfect solution.

If reasonably possible; training, systems and procedures should be revamped to try to prevent the "FLCH Trap".

More emphasis should be made on monitoring and maintaining the glide path, airspeed and descent rate during landing. (Basic flying skills) Policies should be improved to qualify, train and maintain these fundamental skills.

The above is just my personal opinion, with what little is known publicly.
 
For those who don't have a clue what FLCH is, it stands for "flight level change" and is what I think of as "indicated airspeed". It screwed with my mind during 680 initial because I had no idea what "flight level change" could mean. Finally someone told me it was the same thing as IAS on the airplanes I had been flying. Does anyone have an explanation why they call it "flight level change"?
 
In the Gulfstream it's a preprogrammed set of climb/descent speeds based on altitudes and flight conditions. Instead of FLCH we often used v/s mode set to 2,600 fpm after takeoff for a smoother transition to I climb and the same technique at TOD (but lower rate) when starting down.
 
In the Gulfstream it's a preprogrammed set of climb/descent speeds based on altitudes and flight conditions. Instead of FLCH we often used v/s mode set to 2,600 fpm after takeoff for a smoother transition to I climb and the same technique at TOD (but lower rate) when starting down.
In the 680 if we take off and hit FLC at a certain airspeed it will stay at that airspeed if it can. If we hit VNAV it will go to the preprogrammed speed (200 then 250 then .64 in the climb). We can use V/S too. OK it's slow (slower than a Gulfstream) and it doesn't have autothrottles.
 
For those who don't have a clue what FLCH is, it stands for "flight level change" and is what I think of as "indicated airspeed".

I am not sure what you mean by this other than in FLCH, the autopilot will hold the speed set in the MCP, or Mode Control Panel.

Does anyone have an explanation why they call it "flight level change"?

The only time you use FLCH is when you are changing altitudes, typically in descent, but it is possible, IIRC, to use FLCH in a climb. When using FLCH, the autopilot will hold the set airspeed at either max continuous thrust in a climb or flight idle in a descent, unless the pilot manually sets the throttles at another setting.

In the case of a descent, the airplane will continue in FLCH until the altitude set in the altitude window is reached. It will then capture the altitude and continue in level flight at the speed set in the MCP.

IF the altitude is set to zero, or field elevation, there is no protection as far as altitude is concerned. The auto pilot will merrily fly the plane into the ground. Our SOP, as far as FLCH is concerned, is to set the MCP altitude no lower than Glideslope Intercept Altitude, when being vectored for the approach.
 
Overall and most importantly I feel there is a cultural problem.

Can't disagree with that.

Allegedly according to flight data recorders, 8 seconds before the crash, the throttles were advanced. If I understand correctly one of the pilots advanced the throttles. If I understand correctly, this would have disengaged the autothrottle.

Only if the pilot manually disconnected the autothrottles. Otherwise, the autothrottles are just being overridden by the pilot.

Therefore from that point on I would think "FLCH" would be disengaged.

No, FLCH is still active until either altitude capture or some other pitch mode is captured, IE VNAV, Vertical Speed or Approach mode.

I was taught to largely land airplanes with my hands on the stick and throttle.

We still do that in the terminal area.

I think the pilots in the crash were fixated on using attitude/pitch to control the flight path, thusly neglected/forgetting thrust.

Can't disagree with that.

IE: Flying on the edge of flight envelopes to save fuel, time & money.

Well, actually as close to the MIDDLE of the envelopes, but semantics, I guess.

I get the impression that the *culture for landing large modern aircraft have fallen into the **routine of using hands the stick (yoke) only to land the planes. (As opposed to hands on stick and throttle landings)
*(pilots, airlines, governing agencies, and manufactures)
** (a trap of complacency and dependency).

Not at UAL

An instructor with not much time on type of aircraft.

3000 hours, IIRC. How much time do you WANT him to have?

I think automation has increased aviation safety, however I think automation has inadvertently introduced some of its own dangers. There is no perfect solution.

Now THAT statement I can 100% agree with.

If reasonably possible; training, systems and procedures should be revamped to try to prevent the "FLCH Trap".

We have SOPs that prevent that.

The above is just my personal opinion, with what little is known publicly.

Yeah, and what little you personally know about how the industry operates. Not a slam, per se. Just saying that a lot of what you suggest is already what we do, at least at UAL.
 
We have SOPs that prevent that.

You guys do, but one of the things I learned from some of the professional discussions on this are that some carriers (FEDEX is apparently one of them that does this in the 777) that use the practice of setting Zero for the altitude vice MDA which is exactly what the article is describing:

Boeing recommends setting the minimum descent altitude for non-instrument approaches in the MCP altitude window to ensure the aircraft levels off and maintains speed, though some airlines will set “0 ft.” in the altitude window, the experts say.

Be interesting to see if FEDEX changes that practice after this.
 
I am not sure what you mean by this other than in FLCH, the autopilot will hold the speed set in the MCP, or Mode Control Panel.
The speed you set is the indicated airspeed, isn't it? Anyway, that same mode was called IAS in airplanes I flew previously. FLC which is what they call it in the Sovereign does the same thing.

The only time you use FLCH is when you are changing altitudes, typically in descent, but it is possible, IIRC, to use FLCH in a climb. When using FLCH, the autopilot will hold the set airspeed at either max continuous thrust in a climb or flight idle in a descent, unless the pilot manually sets the throttles at another setting.
I see where part of the confusion is. We don't have autothrottles so we would never set an airspeed to cruise. We have to do that by setting the throttle ourselves. Pilot who used to fly the Airbus complained about that when we first got the airplane and still mentions it sometimes. :rofl:

In the case of a descent, the airplane will continue in FLCH until the altitude set in the altitude window is reached. It will then capture the altitude and continue in level flight at the speed set in the MCP.
That's what ours does except it obviously will not continue in level flight at the speed selected.
 
True enough. I can't wrap my brain around you guys not having autothrottles. :D

I hear the new ones after the block change are supposed to have them. I don't know how that project is going, though.
 
Are there any short term ill effects from being covered in this stuff, if you were already a trauma victim?

The diagram looks like someone could make a plausible case that the battalion chief was culpable since he directed foam to be sprayed in front of the left wing, ludicrous but this is the American legal system none the less.
 
If you mean the foam , I really do not think so . It may burn a bit if you have open wounds . I'd have to go over to the fire station next door and look at the MSDS for the class A foam we use . I do not know if it is the same stuff they would use or not .
 
Sounds like the airline is taking some worthwhile steps:

Asiana, South Korea’s second-biggest airline, plans to expand instruction for air crew and begin an outside review of safety standards, it said last month. Pilots will get more hours in flight simulators to prepare for approaches to airports without landing guidance systems.

The carrier said it will also hire another company to evaluate its procedures, add safety specialists and boost maintenance.
 
A visual approach can absolutely be stressful. Especially if they dropped them in like they say they may have and the airplane was in a high energy state and had a while to descent to make their altitude restrictions. Also consider that they had be flying a very long flight essentially all of which would have been on autopilot apart from the takeoff and initial climb out. Add all those things in with the confusion about the FLCH function this is a combination for an accident.

Not excusing what happened here. I am saying that we should point the finger, and say how could they have done something that stupid. The fact of the matter is we are all human and we all have good and bad days. The real error and tragedy here is in the fact that it fell through every filter that was there to stop it.
 
If I were to go out today, and hop in my plane for a lengthy flight in IMC, concluding with an approach to minimums, that would be "very stressful" to me, because this is at the edge of my abilities and proficiency as I sit here writing. For me, flying is a hobby, not my vocation.

If, on the other hand, I intended to make my living as a professional pilot, accepting therewith the responsibility to properly train and maintain proficiency, I also accept the responsibility of conducting myself appropriately. If the slamdunk approach was overloading the flight crew's ability to maintain proper aircraft control and situational awareness, it was their responsibility to recognize that situation, and react appropriately (go around, request additional vectors, whatever it takes).

If the "stress" was because the captain was simply inadequately trained, or simply incapable, the responsibility lies squarely with the air carrier.

"Stress" is no excuse for crashing a perfectly-functioning aircraft, in flawless weather.
 
While I agree with most of the points you make I think it is important that we do not make the mistake of assuming that flawless weather makes things easy. In fact you can argue the exact opposite here because if the weather had been lets say IFR/IMC then the aircraft would've been spaced further apart, the aircraft would have had a much more shallow descent profile to the final approach, the crew would have flown the instruments, etc etc.. Now would you fly instruments for a while you can easily lose your touch for visual flying. For example I flew a local flight doing approaches the other day. Did 5 approaches in 2.0 hours with the foggles on for 1.8 of those hours. By the time it came time to take off the foggles and actually fly the airplane home on the visual I was completely disorientated by the visual environment and elected to have my friend fly the airplane. The small corrections of instrument flying can lead you to make those corrections in visual flying if exposed to instrument flying for a prolonged period and/or as a normal operating method.

Again I am not excusing anything that happened here. I am just suggesting that it is important that we consider that we to are susceptible to the same errors. I am not an expert on the case by any means. Simply speaking hypothetically here. Clearly this could have been avoided but....hind sight is 20/20.
 
Pilot said " airplane too big and heavy for visual approach....." Made him ......" "Nervous" Unquote.

It's a good thing he didn't have to return to land with full fuel at the beginning of his 11 hour flight.

He managed to end up only 800 feet short at LAX when he was relatively light ...
 
It's a good thing he didn't have to return to land with full fuel at the beginning of his 11 hour flight.

He managed to end up only 800 feet short at LAX when he was relatively light ...
If he was landing LAX then he missed by more than 800 ft...
 
The trouble here is that they are not hiring pilots, who have learned how to FLY an airplane. How many flew Beaver floatplanes every summer for 10 years before they became "airline" pilots?

Button pushers................
Visual approach stressful in perfect weather, only to a button pusher.
 
Pilot said " airplane too big and heavy for visual approach....." Made him ......" "Nervous" Unquote.

Which begs the question. Are we bring unfair by expecting them to handle much larger complex planes without using all of the fancy gizmos?
 
Let's see. One of my very first instruction on landing was "If your aim point is stationary on the windscreen, that is where you will land." Didn't seem all that stressful, even in an F-14 (on a calm sea) or a C-17. ;)

Of course I learned that before PAPI, et al.

Cheers
 
A visual approach can absolutely be stressful. Especially if they dropped them in like they say they may have and the airplane was in a high energy state and had a while to descent to make their altitude restrictions. Also consider that they had be flying a very long flight essentially all of which would have been on autopilot apart from the takeoff and initial climb out. Add all those things in with the confusion about the FLCH function this is a combination for an accident.

Not excusing what happened here. I am saying that we should point the finger, and say how could they have done something that stupid. The fact of the matter is we are all human and we all have good and bad days. The real error and tragedy here is in the fact that it fell through every filter that was there to stop it.

I read the whole CVR transcript, they didn't sound stressed out to me except maybe unrelated job/business stress. They made their altitudes and speeds just fine, they just didn't clue into the visual perspective cues on short final and got too slow.
 
This was not an accident.
At best it was criminal negligence.
It meets the standard for depraved indifference and homicide.
A cockpit full of pilots (supposedly) sat there like frogs on a log and let the airplane just fall out of the sky because the "captain" was superior to them (nasty Asian cultural habit).
He was nervous about landing without the Cat III in operation because he knows that at best he is incompetent.
Were I the Prosecutor in that county they would not have been allowed to pack up and leave anymore than a truck driver who blew through a red light and killed people would have been allowed to zip back to Mexico.
There are accidents and then there are felonies.
 
I love this; the Asiana Captain considered it "very stressful" to have to fly a visual approach (in excellent weather!).

Could that be partially due to living in a country that doesn't have a viable general aviation sector?
 
Last edited:
Back
Top