777 Crash and Fire at SFO

[Gabriel]

So there is no way to establish the elapsed time between these points? I don't really understand from what source the information is received. Does it not come in regular time intervals? I assume it is datalink info and not radar returns. ACARS? VDL? ADS-B? Maybe you know?

While they do call it 'preliminary data', it is strange that the NY Times would state 9 seconds in a published infographic if there is no way to establish that. They typically are notorious fact-checkers on things like graphics. Perhaps they have worked with additional sources on this...

AFAIK, FlightAware works with a network of voluntary users that receive ADS-B data in their computers (using a specialized but not very sophisticated or expensive equipment) and upload it in real time to the FlightAware network, very much like LiveATC.

I guess that the emission of each signal is every time the transponder is interrogated by a radar or another airborne equipment (another ADS-B, or a TCAS). If that's correct (and this is really a guess because I have no knowledge), then the data will not be equally spaced in time.

I don't know if the ADS-B data has a tag time in it. I don't think so. So the time tag may be put by someone downstream (the receiver, the user's compurter, FlightAware servers...).

So the answer to your question is... I don't know!

[Gabriel]

So there is no way to establish the elapsed time between these points? I don't really understand from what source the information is received. Does it not come in regular time intervals? I assume it is datalink info and not radar returns. ACARS? VDL? ADS-B? Maybe you know?

While they do call it 'preliminary data', it is strange that the NY Times would state 9 seconds in a published infographic if there is no way to establish that. They typically are notorious fact-checkers on things like graphics. Perhaps they have worked with additional sources on this...

Did you see LeftSeat's post #243?

They also claim to use FlightAware as the data source, and sure it doesn't look like there is any 4000 or 6500 fpm descent there.

[Evan]

Did you see LeftSeat's post #243?

They also claim to use FlightAware as the data source, and sure it doesn't look like there is any 4000 or 6500 fpm descent there.

He refers to using radar returns but he is actually using the same data stream from Flightaware, not radar. It doesn't show the kind of sudden drop the NY Times is showing (which, in the absence of pax reports of a sudden drop does seem unlikely I admit), but it seems to show the same basic scenario: high and fast decending into the glidepath, falling through and unable to arrest the descent... late on power..

In any case, if either are remotely accurate, failure to abandon an unstable approach is probably going to be causative factor #1 here.

[Evan]

I don't know if the ADS-B data has a tag time in it. I don't think so. So the time tag may be put by someone downstream (the receiver, the user's compurter, FlightAware servers...).

ADS-B is supposed to transmit every second. So the intervals must come from the relaying receivers or FlightAware's own algorithm.

[Black Ram]

No offense or disrespect meant, and no intention of starting an argument, but I have this one question, which has been bugging me. We've talked a lot about different philosophies and how "unnatural" the Airbus flight control philosophy can be. No doubt this 777 crash looks like nothing more than very poor airmanship. But if they lost track of their speed some 35 seconds prior to impact and continued decelerating, would Alpha Floor have prevented this? Does Boeing have an equivalent to Alpha Floor? And what happened to the seat-of-the-pants advantage of using a yoke with feedback and manual trim?
Again, don't want to start a general argument, just curious about the mechanic of this one accident. Maybe a little early for this, but for me this is almost without question poor airmanship, poor CRM and the lately very popular "failure to monitor instruments and flight parameters".

[TheKiecker]

I will do one better than that, I will leave you all to your ignorance.
When I see what most of you write here, it is no better than the press.

Enjoy

NTSB: Asiana Aiways Boeing 777 auto-throttle was activated before the crash

NTSB: The Asiana pilot flying at the time of the SFO crash flew Airbus jets before switching to the Boeing 777

NTSB: Crew member assumed the auto-throttle was maintaining the speed before the 777 crashed

International Desk @CNNInternatDesk 2m
Two flight attendants aboard Asiana Airlines Flight 214 were ejected when the plane slammed into the runway, NTSB says.

[phoneman]

I work for an airline, and I have been studying aircraft accidents since I was nine years old and witnessed the Delta 191 crash victims through my playroom window at Children's Medical Center/Parkland. The heli pad was right outside and I saw things that nobody should ever have to see. Parkland is the burn unit in DFW. I was traumatized, and the only way I was able to deal with it was to try to understand how it happened. I was taken to the crash site a couple of days later. The smell on the air was horrible.

Anyway, I figured most of the pilots would come out in droves to defend their kin.

They were morons.

Sorry.

Who are you referring to,the 191 pilots or their kin?

[Gabriel]

NTSB: Asiana Aiways Boeing 777 auto-throttle was activated before the crash

NTSB: The Asiana pilot flying at the time of the SFO crash flew Airbus jets before switching to the Boeing 777

NTSB: Crew member assumed the auto-throttle was maintaining the speed before the 777 crashed

International Desk @CNNInternatDesk 2m
Two flight attendants aboard Asiana Airlines Flight 214 were ejected when the plane slammed into the runway, NTSB says.

So several things here:

- Using automation doesn't relieve the pilots from being in control. If we relied A/T and A/P enough to leave them alone, we would not have two living beings in there. TK also had the A/T engaged.

- They had the A/T in FLCH mode. Let me give an introduction. Thrust, speed and climb rate are function one of the others. Fix two of them, and you get the third one. So there are two ways to manage climbs and descents with the automation: Select the speed and climb rate, the A/P will hold the selected climb rate, the A/T will hold the selected speed, and the thrust will be whatever is needed to make it happen. There is another way: FLCH: Select the speed and thrust. The A/T will hold the thrust, the autopilot will hold the speed (by raising the nose to slow down and lowering it to speed up) and the climb rate will be whatever is needed to make it happen.

I don't know how it works in the 777, but in other planes, you can program the autoflight to change mode when something happens. For example, you can select a given altitude, and no matter in what of the two modes you are climbing/descending, when the airplane reaches that altitude the system changes to ALT HOLD, so the A/P holds the altitude and the A/T holds the speed. More or less the same with approach. When the plane intercepts the glide slope, the automation goes into approach mode, so the A/P tracks the glide slope and the A/T holds the selected speed.

Now, what happens if you select FLCH but are manually controlling the pitch? Well, in that case you should be doing the job of the A/P in this mode: hold the pitch as needed to keep the selected speed, and the flight director command bars will lead you to do just that. But if you don't obey and pull up instead, NOTHING will add thrust, since you've asked the A/T to hold the thrust, NOT the speed.

YOU CANNOT USE FLCH MODE TO TRACK ANY VERTICAL TRAJECTORY BECAUSE THE DESCENT RATE IS THE OPEN VARIABLE IN THIS MODE.

And using FLCH when flying manually doesn't make a lot of sense anyway, in my opinion, since the A/T will do basically nothing: holding the thrust means just holding the thrust levers where they are, just as they do if you have the A/T off and put the levers in any given position: they will stay there.

[3WE]

I will do one better than that, I will leave you all to your ignorance.
When I see what most of you write here, it is no better than the press.

Enjoy

Good riddance.

From your record you posted earlier in this thread, it seems clear that you are highly experienced (i.e. old), highly senile and probably have ED.

I've known some extremely good trolls. They can be quite infuriating, but deep down there's some humor and desire to educate from them (ITS, Flyboy).

However, your posts, just plain stink, and are nothing but put downs and declaratoins that since you are old, you must be smarter than us.

[Gabriel]

No offense or disrespect meant, and no intention of starting an argument, but I have this one question, which has been bugging me. We've talked a lot about different philosophies and how "unnatural" the Airbus flight control philosophy can be. No doubt this 777 crash looks like nothing more than very poor airmanship. But if they lost track of their speed some 35 seconds prior to impact and continued decelerating, would Alpha Floor have prevented this? Does Boeing have an equivalent to Alpha Floor? And what happened to the seat-of-the-pants advantage of using a yoke with feedback and manual trim?
Again, don't want to start a general argument, just curious about the mechanic of this one accident. Maybe a little early for this, but for me this is almost without question poor airmanship, poor CRM and the lately very popular "failure to monitor instruments and flight parameters".

I understand that's correct: The Airbus FBW would have added thrust (but without moving the thrust levers) even with the A/T off.

Maybe the 777 would not have, but the philosophy is different.

In the Airbus, even in manual flight, the plane will behave as if it was on A/P in vertical speed mode, and the yoke works like the vertical speed knob: Pull to increase the vertical speed to the desired value. Push to increase the vertical speed. Center the stick to keep the speed. With the stick centered, the vertical speed will be kept by the FBW. That means that if you slow down, the FBW will pitch up by itself (with no pilot input, and with the A/P off) to keep the vertical speed. In this context, an inadvertent stall is a likely mistake, since there is no feedback to the pilot (except the airspeed indicator) about the deteriorating speed.

A non-FBW plane (for example, a balsa glider, a Cessna 172, a 767, or the 777 which is FBW but kind of mimics the natural response of a non FBW plane) will do EXACTLY THE OPPOSITE (I would in fact say that the Airbus FBW will do exactly the opposite to the natural response of a plane, which is why I don't like it): It will keep the AoA and the speed, and that means lowering the nose and increase the sink rate in conditions where the speed would tend to deteriorate. That means that, to keep the sink rate (or even the pitch) constant, the pilot needs to hold an ever-increasing pull-up displacement and force on the yoke, or constant nose-up trim inputs, which is a VERY STRONG feedback that you are slowing down and increasing the AoA and, while it might look like a confusing feedback for the layman, it's perfectly natural and instinctive for a pilot because it's what he is used to since the hour number zero in the J-3. And unlike non-FBW planes, the 777 "exaggerates" the pull force needed to slow down below certain point, in Boeings way of envelope protection (make corrections but leave full authority to the pilots).

That said, maybe a thrust increase (which would still be overridable) when certain condition is met might not be a bad idea (if it's not implemented).

[Evan]

So several things here:

- Using automation doesn't relieve the pilots from being in control. If we relied A/T and A/P enough to leave them alone, we would not have two living beings in there. TK also had the A/T engaged.

- They had the A/T in FLCH mode. Let me give an introduction. Thrust, speed and climb rate are function one of the others. Fix two of them, and you get the third one. So there are two ways to manage climbs and descents with the automation: Select the speed and climb rate, the A/P will hold the selected climb rate, the A/T will hold the selected speed, and the thrust will be whatever is needed to make it happen. There is another way: FLCH: Select the speed and thrust. The A/T will hold the thrust, the autopilot will hold the speed (by raising the nose to slow down and lowering it to speed up) and the climb rate will be whatever is needed to make it happen.

I don't know how it works in the 777, but in other planes, you can program the autoflight to change mode when something happens. For example, you can select a given altitude, and no matter in what of the two modes you are climbing/descending, when the airplane reaches that altitude the system changes to ALT HOLD, so the A/P holds the altitude and the A/T holds the speed. More or less the same with approach. When the plane intercepts the glide slope, the automation goes into approach mode, so the A/P tracks the glide slope and the A/T holds the selected speed.

Now, what happens if you select FLCH but are manually controlling the pitch? Well, in that case you should be doing the job of the A/P in this mode: hold the pitch as needed to keep the selected speed, and the flight director command bars will lead you to do just that. But if you don't obey and pull up instead, NOTHING will add thrust, since you've asked the A/T to hold the thrust, NOT the speed.

YOU CANNOT USE FLCH MODE TO TRACK ANY VERTICAL TRAJECTORY BECAUSE THE DESCENT RATE IS THE OPEN VARIABLE IN THIS MODE.

And using FLCH when flying manually doesn't make a lot of sense anyway, in my opinion, since the A/T will do basically nothing: holding the thrust means just holding the thrust levers where they are, just as they do if you have the A/T off and put the levers in any given position: they will stay there.

Gabriel, FLCH is not an A/T mode, it is a pitch mode. When FLCH is selected, the A/T is in THR mode - unless - as I think it was in this case - it goes to IDLE - whereupon it changes to HOLD mode. There has been a lot of chatter on other forums about the 'FLCH trap'. As best I understand this, it happens when you have selected the FLCH button in descent while trying to get rid of altitude and you therefore send the A/T into IDLE and then HOLD and then you disconnect the A/P, thinking the A/T will catch up but it's still holding at IDLE. I think this is how it works, but I have less than 43 hrs on the 777 .

[Gabriel]

Gabriel, FLCH is not an A/T mode, it is a pitch mode. When FLCH is selected, the A/T is in THR mode - unless - as I think it was in this case - it goes to IDLE - whereupon it changes to HOLD mode. There has been a lot of chatter on other forums about the 'FLCH trap'. As best I understand this, it happens when you have selected the FLCH button in descent while trying to get rid of altitude and you therefore send the A/T into IDLE and then HOLD and then you disconnect the A/P, thinking the A/T will catch up but it's still holding at IDLE. I think this is how it works, but I have less than 43 hrs on the 777 .

Okay, I understand my mistake. But that's the idea.

[AVION1]

And what exactly is your job title at this airline? Do you think because you witnessed another aircraft accident you have the qualifications to justify your remark validating AVION1's statement for revoking an airlines right to fly into the U.S and having the certificates of two flight crew members revoked?

You are aware they call this an accident right? I seriously doubt these guys decided to do this for fun.

Sorry if I open my mouth here again, but I am an FAA Accountable Manager, I run a Part 145, dealing with the FAA everyday is my job and the CFR regulations is my Bible. And revoking airmen licenses and certificates is more common than you think, specially with the new regulations and enforcement around.
Those pilots are not only being investigated by the FAA and the NTSB, but an enforcement action is around the corner, specially if they don't know how to deal with the FAA.

[Evan]

Latest from Reuters:

• 82 seconds before impact: Plane's altitude is about 1,600 feet. Autopilot is disengaged. Speed not given. The plane is configured for approach and proceeds normally with "no discussion of any aircraft anomalies or concerns," NTSB Chairman Deborah Hersman later says.
• 73 seconds: Air speed slows to 170 knots. Altitude is about 1,400 feet.
• 54 seconds: Speed falls to 149 knots, altitude drops to 1,000 feet.

So, a drop of only 200 ft in the 9 seconds after A/P disconect, and a drop of 600 ft in the 28 seconds after A/P disconnect. So NY Times graphic wrong, point Gabriel.

[3WE]

...They were morons...

Three comments for you Myndee:

1) You make too many statements that are exceptionally bold, over the top and not thought out
2) I bet you make mistakes occasionally
3) There's a $20.00 word called insidious

Pretty much every commercial plane crash comes with highly trained pilots at the yoke.

These guys have done who knows how many zillions of landings, the autothrottles ALWAYS take care of their speed- and I'm reading that the autothrottles were set. 140 knots is pretty slow and mushy...probably doesn't feel all that different from 103 knots

Do you know if they were well rested?, or worked a tough schedule?, or maybe- purely randomly- they both had insomnia the night before?

Have you ever crossed the center line, or had a 10 mph speed variation while driving your car? Ever caught yourself daydreaming for just a second? Ever been tricked by a mental illusion? Ever been wrong when you knew with 100% certainty that you were right.

If you say "no" to the above, I'll give you a fourth comment that you are delusional.

The good pilots who post on this forum can all tell you of mental lapses and mistakes they have made. And aircraft mechanics have made their mistakes: AA 191, The ASA brasilia with a panel that wasn't screwed on right, Miss-trimmed Beech 1900, numerous instances of cross controlled ailerons.

The system is pretty good at catching those lapses, but occasionally, all the holes in the slices of swiss cheese align just so- and if it's your turn, Ms Myndee- it could be that YOU are the one who forgot to tighten some nut on the engine and now there's a smoking hole in the ground.