777 Crash and Fire at SFO

[Evan]

By the way, have you've heard of a lot of MD-80s crashing becasue of that? And I bet that there have been much more landings in MD-80s than in 777s.

This is a piloting issue, definitely. I'm not trying to soften that reality. So, in the aftermath of AF447 we saw a renewed effort to reinforce unusual attitudes recovery and approach to stall and stall recovery training. It was a wake-up call to the industry that pilots need to be proficient at these things. Now, will we also see a similar reinforcement of automation modal and intermodal awareness? Will all pilots be required to be more deeply familiarized with the magic box on the dashboard? I'd have thought Turkish 1951 would be wake up call enough, but alas, nothing came of it.

[EconomyClass]

Was unaware of this crash, but this line from Wikipedia does demonstrate component failure:

A preliminary investigation found that the crash was caused primarily by the aircraft's automated reaction which was triggered by a faulty radio altimeter

[Evan]

Was unaware of this crash, but this line from Wikipedia does demonstrate component failure:

If you are referring to Turkish 1951, there is an interesting aspect to this automation failure. The crew was aware of the failed radalt; it was MEL'd. As designed, the autothrottle will not work with a failed radalt on approach because the logic compares both radalts and gives up if there is a significant disagree. Therefore, with a MEL'ed radalt, they should not have even attempted to use autothrottle on approach. But, whatever, as designed it is going to self-disconnect anyway.

But the A/T didn't self-disconnect as it is designed to in this case. What happened is that the logic failed to cross-compare and rule the radar alt unreliable, relying instead on the failed one alone which gave an erroneously low altitude, causing the autothrottle mode to change to FLARE (retard thrust) at the wrong moment. This logic failure was not unheard of. There had been something like 15 previous instances and Boeing was aware of the problem. As a fix, Boeing began using different hardware on post 2003 NG's and put out a bulletin for pre-2003 a/c to upgrade existing hardware.

But here's the thing: to a crew that IS properly aware of the system interactions on the NG, this is NOT particularly dangerous. They would not have been using the A/T on approach with a MEL'd radalt.

As the role of pilots is becoming more about monitoring and less about stick and rudder, training need to be more about deep understanding of the systems that they are there to administrate. Obviously, this is not being done enough.

You cannot fall into the FLCH trap if you know the system interactions because you would not set 0' on the MCP and expect the A/T to kick in anytime before then.

[Gabriel]

But here's the thing: to a crew that IS properly aware of the system interactions on the NG, this is NOT particularly dangerous. They would not have been using the A/T on approach with a MEL'd radalt.

Not to mention that any crew, even if NOT properly aware of the system, but that is FLYING the plane (becaus that's the pilot's main task, you know) will have recognized, even if they didn't understood why, that the A/T mode was in "flare", that the speed was going way below Vref, and that the pitch was stupidly high, just to mention the things that are shown in the PRIMARY FLIGHT DISPLAY in front of EACH PILOT, and would have acted, instead of waiting for the stick shaker to wake them up.

[Evan]

Not to mention that any crew, even if NOT properly aware of the system, but that is FLYING the plane (becaus that's the pilot's main task, you know) will have recognized, even if they didn't understood why, that the A/T mode was in "flare", that the speed was going way below Vref, and that the pitch was stupidly high, just to mention the things that are shown in the PRIMARY FLIGHT DISPLAY in front of EACH PILOT, and would have acted, instead of waiting for the stick shaker to wake them up.

And this is, I presume, what happened in those other 15 or so incidents (in which the crews may have also been unaware of a failure in one of the radalts).

But that's not my point. My point is that, with proper DEEP understanding of system interactions, Turkish 1951 would not have been on autothrottle and would not have crashed.

You see the issue here: the partnership between pilot and autopilot. The crew in both the Turkish crash and this one ASSumed that the automation was taking care of thrust. In both cases, a better understanding of their automation partner would have prevented the accident.

Stlil—veryfrustratingly the forums continue to address this partnership as a rivalry for control. Well guess what: managing the A/P IS hand flying. It is flying with your hands. Instead of a yoke you have a dial. You had better know how to use that dial every bit as well as the yoke or you are not a qualified pilot in a modern airliner. And again and again we have these crashes where pilots with PLENTY of stick and rudder experience don't know how to use the dials.

I want my pilots who hand fly to fully understand the aerodynamics behind what they are doing. And when they are using automation I want them to fully understand the technical interactions behind what they are doing.

Is that so much to ask?

[Gabriel]

I agree.

It's not one thing or the other. It's one thing AND the other.

These pilots failed at both: They didn't have a good understanding of the automation AND they didn't do their job at monitoring the flight parameters.

Both failures were required to produce these crashes.

[EconomyClass]

Very parallel. Are there airlines putting their planes in barely qualified hands either to keep the books in balance or remedy a shortage of people who want to work for a less-desirable employer? "Deep understanding" sounds like a challenge for the majority of airlines. I'd say the ones most likely to give that to the pilots are the ones whose jobs are prized and therefore worth working extra hard for. Now, to me, that should include Air France. But Asiana or Turkish? Call me ignorant if you must, but I'm gonna say they don't get the prize of the litter.

[Evan]

I'd say the ones most likely to give that to the pilots are the ones whose jobs are prized and therefore worth working extra hard for. Now, to me, that should include Air France. But Asiana or Turkish? Call me ignorant if you must, but I'm gonna say they don't get the prize of the litter.

You're a Korean looking to fly long-haul for a world-class operator, you've got two options. If you're a Turk you've got one. I'd say both Asiana and Turkish Airlines are the 'get' that you would work extra hard for.

Go back and read BB's post from Tom, the Korean sim instructor. The problem as he sees it is a combination of poisoned management culture favoring military pilots regardless of demonstrated talent (politics) and a deep cultural tendency to learn by rote rather than to comprehend systemic concepts and solve unscripted problems.

I think the best pilots are those who have a natural curiosity that exceeds their training requirements, who spend a good deal of free time off the job reading about aviation, aerodynamics, incident investigations and get to know their current aircraft systems from stem to stern.

The ones who bother me are the ones who learn only as required. Sooner or later that means learning the hard way.

[EconomyClass]

Asiana has been at it for a while. Do they have a bad safety record?

[Myndee]

I tried the nice approach.

Repeating, you really do make overly bold posts and assertions.

And Boeing Bobby was actually pretty correct in his assesment of you (he just happend to use hard honesty, and the truth does sometimes hurt).

The main thing I will bring up is that when you did reply to my "nice post", you did not touch on any of the the actual suggestions or criticisms of your posts. That would be a flaming example of a totally closed mind, ignorance, and inability to listen.

Yeah, Boeing Bobby can be a grumpy old man and I even made the nasty jab about ED which he denies and I genuinely belive him (just like I belive his statements about his credientials)

He's awfully darn factual, accurate, correct and a decent judge of folks posts...and the sucker listened and turned the grumpiness down 1.5 notches and suddenly we all love him. (Not only that, he's on record supporting ITS).

...you are the one who is a troll who is coming on here spewing over the top bullcrap.

But sadly, you will again deny it.

NICE? HAHAHAHA

3WE, you are not only a troll, but you are a bully that doesn't like women. Get over it.

I'm certainly not a troll, and I'm not spewing anything over the top. If I ever post something incorrect or I misspeak as I did in the referenced post, several are quick to correct me. I know what I know and what I don't know, I learn.

We all aren't born geniuses like you 3WE.

I never claimed to be a pilot. Most of you are only sim pilots, many of which I have great respect for!

Just as you (or someone said) it is nerve-wracking as a pax to see your shadow catching up with you. I am very well aware the pilots have a much different view.

CONTINUING ON...

You all are aware that SFO has been undergoing construction to lengthen it's runways, yes? They can't build an extra one to stop the GDP nightmare but they have lengthened them. Was this runway already lengthened? I have been gone from work for a while so I don't know.

[Leftseat86]

You all are aware that SFO has been undergoing construction to lengthen it's runways, yes? They can't build an extra one to stop the GDP nightmare but they have lengthened them. Was this runway already lengthened? I have been gone from work for a while so I don't know.

28L was recently lengthened, but not by much. Runway length was not a factor in this accident whatsoever, in any case.

[EconomyClass]

This article brings up interesting questions. Studies on human perceptions have shown that mental constructs actually influence or even determine what is "seen". We always think our vision is empirical, that the brain simply processes the sense data. That seems to be wrong, it looks and confirms mostly. And that means sometimes it sees what is not there. If the data is close to something anticipated, the brain can "round" it to what it wants to see. The hard thing is to look at things naively as if they've never been seen before. To pretend perpetually to be ignorant.

http://www.coshoctontribune.com/usat...rticle/2680895

[EconomyClass]

Interesting commentary by a lawyer/pilot who specializes in crashes:

http://www.digitaljournal.com/pr/1430116

[Highkeas]

Interesting commentary by a lawyer/pilot who specializes in crashes:
http://www.digitaljournal.com/pr/1430116

I wonder if Patrick Bailey is related to F. Lee Bailey who also was a pilot (biz jet) and an aviation lawyer.

[Evan]

http://www.coshoctontribune.com/usat...rticle/2680895

I propose that the first question on the CTPL test should be:

Complicated things make me sleepy.

| | Yes

| | No

| | What was the question again?