infrequentflyer789

It's all working here.

Only problem is you make it look too easy - the still photo of the wing always looked seriously scary (at least to me), but in the video you make it look like a non-event. Makes it all the more impressive.

gums

Thanks Infrequent and JC.

You must realize that I did not turn on the VCR we had ( first jet with the "home movies" for de-brief, heh heh) until I had done a few things and then had the gear down. All what I did was in the after action report for other pilots with the same problem.

I also apologize for poor R/T with tower, and I also had the private VHF with Rooster for stuff, so some radio is private and other is to the tower.

My basic point is to show how one can handle a serious problem without doing a lotta stupid things. It is called airmanship. Also did not turn on the VCR until base leg with gear down and a fair semblance of control ( 17 pounds of authority and about 15 - 16 pounds required to hold wings level according to the recorders). I then turned on the narration for last two miles in case I had to jump, and I wanted it for the accident board. Silly me.

Funny thing was using the auto-ack radar to find the C-141 who was trying to land the same time I was, heh heh. So that box with the "x" you see for a few seconds was the lock-on, and I kept looking for that sucker because the jet seemed O.K. at that time.

All is well that ends well.

It was the only time I ever came home and my wife knew that something serious had happened. You damned right. First and only time my maintenance guys let me down, and when that senior NCO showed me the missing cotter key my legs got all gimpy.

Oh well a good thing that saved a few later folks with the same problem - most mechanical failures and not maintenance.

Lyman

Owain

"I think you've got it about face again - +ve sidestick is ND - he's lingering in NU."

Exactly. Precisely. Clearly. With a focus on that, let's compare what happened at the STALL "BREAK" with what happens at the unlatch of the autopilot.......

Just as PF has got the flight path mastered, (per BEA), the Bars return, and he follows with aft stick (he is told he is LOW), BUT, with a trained in response that did them in he selects TOGA, and just enough NU from the Thrustline to get to AoASTALL 9.6. and bob's your uncle.

That is how they STALLED. How did they CLIMB, to get into such a precarious position in the first place? The same way. The same way the other flight climbed, they pulled.

BUT WHY? Following "orders"? Of course, how else? I have been trying to get it into the discussion for a while, let's take an honest look at the pilotage. It was not impatience, wholly, or even partially that started the climb to disaster, it was the data on the Flight screens. Garbage in, garbage out, climb back up to the flight path. It was not Turbulence, nor an UPdraft, it was the boxes.

Why would the pilots insist on climbing, especially when the REC was discussed, and agreed to be not partial to ascent? Why would they persist in the climb?

Follow me, flyboy..... The siren song of an aircraft who cannot tell the truth. It is not her fault, these things happen. The FD bars were lying at the outset, it is what got them into the zoom to begin the disaster, and when the PF had it sorted, they returned, just at the critical point when the Pilot had it "mastered"

You all believe what you will, or what you are told. The first response to the bars in PITCH and RUDDER got them onto the dirt road to HELL, and they were close to getting back on the tarmac, but fate....

Look at the teeny little hook on mm43's excellent graph, where the FD comes back just prior to the STALL...That's TOGA, a pinch more NU, and "crazy speed" follows, STALLED. Of course he was flying the FD's, they were on, and the a/c doesn't show them if they "lie".... But at the first? What 'started' the accident?

To me, I always wondered why everyone's attention was deliberately focused downstream, to get into reams of post STALL/ZOOM conjecture.

That is where the crash began, when the pilot followed his data, to climb. And climb, and.....climb. At first the a/c was hesitant, almost sentiently resisting the ascent, but in the end, she Nosed up, and that was that. Gargabe in, garbage out.

"Why would an experienced pilot keep pulling, against his better judgment?"
To bubbers44: He DID, but not due lack of 'stuff'. You would do what he did, so would PJ2, so would gums, Machinbird, and on.... In the ****, you fly instruments. Everyone does, it is trained.

Of course there's more, but I prefer to start at the most important cause, and let follow the chain behind.

Lyman

Look at the Stall point. With his stick at aft 12 degrees, he moves it forward about 20 degrees, instantly, and he already has TOGA, so to me this shows a textbook STALL recovery. It also shows, via the AoA green line, that the a/c did break, her nose drops. Follow the AoA line, and mind the sequence of "bumps". I think each of these bumps is a secondary STALL, then Tertiary, etc. He is trying to "fly" on the "good" side of the Break. Each time it breaks, he thinks he is STALLING, then recovering. He has the authority to arrest the "descent" (stop the NOSE from falling further) each time it breaks, sad...... and thanks to the THS. The STALLWARN is counterintuitive, at this point. I think he hears the WARNING as an "About to STALL", instead of STALLED. This is how the a/c gets completely away from him, as all his forward speed goes away, and he refuses to allow the NOSE to drop seriously, into what may have been a recovery.

mm43. What an awesome graph. It is as if one is there, in the cockpit.

RetiredF4

Lyman,

i´m not following your asessment.
As a pilot you follow the "intruments" as long as it makes sense. To follow a FD bar would be totally out of this world, when Altimeter tells you you are climbing above max altitude, when pitch tells you your nose is far too high, when stall warning screams you are stalled, when VSI shows values not common in cruise flight, when the aircraft behaviour tells you you are flying my airframe outside healthy parameters.

I´m not saying that it might have influenced their thinking in some kind of way, but you can´t construct the whole accident around it.

gums

I am with Retired.

Hey Lyman! Are you postulating that the PF simply followed steering cues?

What about other indications? basic crosscheck. Granted, the speed and maybe even altitude were questionable, but simply holding the basic attitude of 2 minutes ago would have been the wise choice, IMHO. Don't fool with the power, don't change pitch attitude. Don't just do something, sit there and analyze the situtation and then proceed. And I did not see that.

On my LEF failure, I pulled power back and stayed where I was because the jet was not outta control at that point. Sure, I continued the climbout to gain feet - speed is life, but altitude is life insurance. I posted that video to show how one could handle a unique mechanical failure, and in marginal weather.

If the basic cause of the climb was the PF following steering from the jet, then we have to train the folks for unusual instrument failures. As far as I know, the attitude reference systems were working as advertised. So just hold what you got and take a deep breath while figuring out the next step.

The whole thing pisses me off. A simple failure due to icing or whatever, then an intentional change in attitude within seconds.

Lyman

hey gums,

He was flying attitude, obviously. He had a PITCH CUE, and a ROLL CUE, hence his inputs. Cross checking what? For all he knows it's a simple autopilot disconnect, UAS hasn't come up before, for him, and he's flying in NORMAL LAW, as he would be in for any autopilot drop not related to UAS.

Retired? Of course he's steering, that's what he and the autopilot have been doing for two plus hours. He doesn't move to deselect throttles, because deselecting throttle AUTO and deselecting FLIGHT DIRECTORS is part of UAS, of which he is unconcerned. WE know it was UAS, HE DID NOT. NEITHER did, not at that point. So he handles the ship off the FD, easy peasy. Again, why with the alarm? CAVALRY CHARGE and MASTER CAUTION happen at every AUTO quit, so he just wants to get to PARIS. I am saying his level of concern is not up to what you want it to be, or should have been, because he just wants to NAVIGATE.

Mistake? Boy Howdy, but how is he supposed to know? Besides, even if you're LOST, get a DF STEER or a PAR, right gums? No sweat.

I think the FD was worms, and he took bad data and turned it into confusion, alarms, and STALL, by virtue of his reactions to what he thought was happening, which was NOT UAS.

FD worms? Bad AirData? For how long, and how long was the ship "flying" with duff gunk?

We assume the flight degraded because we have the gift of time, and back slapping consensus, we think that's how it went. I say BS.

gums:

What about other indications? basic crosscheck. Granted, the speed and maybe even altitude were questionable, but simply holding the basic attitude of 2 minutes ago would have been the wise choice, IMHO. Don't fool with the power, don't change pitch attitude. Don't just do something, sit there and analyze the situtation and then proceed. And I did not see that.

This is all good, but coming from a pov that has the PF in the stockade, for doing dumb ****. He wasn't, yet most of us have him all "figured out".

It was Not until 17 seconds (check, I think actually eleven) after the loss of AUTOPILOT the PNF announced "we lost the speeds, then". "then" as in, "Oh, Hello, we've lost our airspeeds."

We need to look at all of it, including that PF was a simple line pilot, who wanted to get to France, and home, not some test pilot wannabe who wants to out maverick goose. He acted as though he was unaware that UAS was the call, did he not? His call seems to have been a vanilla a/p drop, and "let's us stick with the flightpath", right?

gums

c'mon, Lyman.

I was not an official test pilot, but flying the first few A-37's and F-16's I had to be one, like it or not.

I am trying to predict the accident board findings, and I have the following findings for other pilots here to critisize:

- Loss of airspeed sensor due to icing or other factors.

- AP disconnect and then reversion to one of the alternate control laws

- A climb to 2,000 or 3,000 feet above the planned cruise altitude.

- Entering a stall due to airspeed decreasing and no "down" command on the sidestick

- Failure of the "cosmic" flight control system to limit AoA commands or actual flight parameters WRT AoA

- Convoluted flight control system laws regarding reversion when the primary mode is inop

- Failure of the crew to realize that they were stalled

- Failure of the crew to hold nose down inputs to get outta the stall

- Failure of the PNF to "help" the PF in a timely fashion

- Failure of the control laws to use AoA even when the airspeed data is unreliable

- training inadequate for the line pilots/crews when sh$$$t happens.

DozyWannabe

Gums, it's not that convoluted - Normal -> Alternate -> Direct. Couldn't be simpler. The only thing you need to remember is that protections are effectively off the table in anything other than Normal Law and more care must be taken when applying input.

The flight control computer system is not "cosmic" - it was only considered ground-breaking 24 years ago becuse it was the first, and the concept has formed the basis of most western airliners built since then.

I've already explained why the use of AoA without a cross-check from another data source is inappropriate in an airliner configuration. If the AoA sensors have failed at the same time as the pitot tubes, then control would potentially be limited in a way that prevents a successful recovery. It's better to put it in the hands of the pilots to make that call - with the caveat that the pilots must be trained properly and thoroughly.

Lyman

A bit more complex than that Doze, and for good reason, also.

Look at the timing and cause of Pilot's initial inputs, and tell me he wasn't flying FD and bars. UAS is not simple. Had it been simple, the Airline and the manufacturer would have designed and implemented a system that would have made it mpossible for Bonin to miss it.

A system that would have kept 447 flying, not guessing. You need to take a breath and back up a step, you assume a position of knowledge and design that is superior to the Companies you defend, and that is not the case. For all that, you don't have the answer, only a stubborn fallback to a system that is under scrutiny for autotrimming to and throug STALL, inadeguate warnings of degrade, and STALL WARN that makes sense only to non-pilots.

roulishollandais

@ gums : your post #767 Thread 8 AF447

Editing 19.5.2012
@ gums : Reading too quikly your post, gums, and others, I mismatched the informations, thinking only to connect the Master/Slave and soft- and hard-limitation ! Sure Your Viper is hard-limited ! Shame to me ! I also deleted smillies.

Your post shows your Viper is hard-limited (no possibility for the pilot to overcome the protection), as Airbus would be in normal law.

I Deleted also smillies.
End of editing.

When soft-limited, the Flight control system is slave, the PILOT is MASTER

When hard-limited (Airbus), the FLIGHT CONTROL is MASTER, and the pilot who is slave, has to look the film...

DozyWannabe

Lyman - I spoke to my friend and the TRE over email the other day and they concurred with rudderrudderrat's assertion that the FDs would have come back indicating the current attitude. Bonin was already pulling before the V/S started dropping away. I'm telling you - he could have been following the bars (although I think it unlikely - it has to be considered as a possibility), he could have been following the ADI (which is my pet theory), he may have been completely thrown and trying anything. We don't know and aren't likely to know until the final report is out.

There *was* a UAS pitch-and-power procedure that Airbus disseminated amongst the airlines, and we know AF had a copy. This was not followed, and we do not know how AF communicated that procedure to its crews.

For the last time I'm not defending any company, I'm simply keeping all options on the table. On the other hand I will refute folklore regarding FBW and Airbus if I know it to not be true, but that's about as far as I go. I have no "skin" in this game other than a desire for rational and evidence-based explanation.

Completely false. The pilot is "master" in a hard-limited system right up to the point where he or she commands a maneouvre that would stall or damage the aircraft - and the aircraft will follow that command right up to the safe limit. The computer is there to help, not to hinder.

Lyman

"Completely false. The pilot is "master" in a hard-limited system right up to the point where he or she commands a maneouvre that would stall or damage the aircraft - and the aircraft will follow that command right up to the safe limit. The computer is there to help, not to hinder."

Item: Autotrim. Soft limit? Silent helper? Hard limit, eg: "Take this trim and....."

Shall the master be given an "AUTOTRIM/DISABLE" toggle to prevent the Stabiliser from "Helping"? Can we see a priority alert for LAW CHANGE in concert with AUTOPILOT LOSS? Something different, and notable, not like the thirty seven items of maintenance logged? A STALL WARN that WARNS and an additional one that alerts "DEPARTED"? Dear DEPARTED, we are gathered here...... Focus, flyboy, you're ass is STALLED.....

mm43

@ CONF iture

The graphic has been updated to include at 02:11:32 the PF's, "I've lost control".

DozyWannabe

The master is supposed to be aware that once the protections are gone, the aircraft must be controlled with as much care as an unprotected aircraft - rocket science it ain't.

This is because with the conferrence of "master" on a pilot, with the power comes the responsibility. I'm completely on board with sympathising with the F/Os, who were thrust into a situation for which they were not trained - but that doesn't make the resultant aircraft handling any less inappropriate.

I realise that I may attract flak as a non-pilot saying this - but even if the FD is telling you to pull up at an altitude that is several thousand feet in excess of the safe limit you discussed only minutes previously (remembering that this is only a theory), then airmanship demands that you ignore the dn FD and just set pitch and power, then try to keep things straight and level until the the danger has passed and the other problems can be resolved.

HazelNuts39

Probably referring to roll rather than pitch?

EMIT

Hello Gums,

Impressive VTR of that LEF UP landing.
Good thing you kept AOA so low during the entire event.
I remember that in the first half of the eighties a Belgian F-16 was lost with the same failure (albeit that, IIRC, the torque tube there really failed mechanically, not just loosened because of a forgotten cotter pin).
The Belgian approach went well, but during the flare, the AOA went up too high and a quick roll over followed. Pilot ejected, but bailout trajectory was already about level (90 degrees of roll) so he did not survive.

You are correct in your evaluation that all that was needed in the AF case was to hold the usual ATTITUDE and THRUST for a couple of minutes. The correct words/commands would have helped - CHECK PITCH (ATTITUDE) or SET PITCH ON THE HORIZON, not - you are climbing - I know - you are still climbing (etcetera). The climb did eventually stop, but the pitch attitude was still way off the mark, of course.

mm43

You could be right, as I interpret that the pitch and roll oscillations shortly after CLmax were caused as a result of the CL shifting all over the place prior to the final break taking place. They didn't necessarily follow the SS commands.

Lyman

HazelNuts39

"I've lost control"
Probably referring to roll rather than pitch?

NO. (...) "I've lost control", (THEN)
"I don't have control of the plane, AT ALL".....

perhaps no controls?

DozyWannabe

Then why do the flight surface DFDR traces mirror the control inputs precisely? You're getting ahead of yourself - take a second to breathe and chill.