Shadoko

@ CogSim
"Why not just use descend in the translation then?"

For an old man like me, the PF speaks a very poor French. I think to make this understandable in English, they use what they think the simplest words. I am not enough fluent in English to tell if they achieve that.

takata

Sorry, I'm a little bit lazzy (or too busy), as I didn't quoted the reference. It is from my notes about AMM 27-90-00, but I suppose it was from an A320 document as the g-load limits for an A330 THS is rather 1.3 / 0.5 g (instead of 1.25 / 0.5 g quoted). But both THS are basically the same thing, as far as the operating system is concerned.

safetypee

Lyman, (#1804)
I suspect that BEA’s examination of ‘Human Factors’ will consider the many aspects within the scientific discipline. This should avoid the often bigoted dismissal of human factors, human error, etc, as a coverall for many complicated and interacting issues in such accidents.
Such an investigation might only be comprehendible with some speculation; this requires activity beyond the scope of ICAO guidance for accident investigation (factual reporting), but this work could provide a valuable resource for learning about the crew in AF447 or similar humans/situations.

The major threats in accident scenarios are in the extremities of the ‘SHEL’ human factors model. The risks are in the interface with the central ‘L’ – the human interaction with the situational environment.
The contributing causes to an accident can be found in all aspects of the model – threats, risks, situation, and management – human activity; the difficulty is deciding which had the greater contribution, or which might be more effective for improving safety.
A human factors view might only be another view, but in such a complicated accident another view might be valuable.
With hindsight, perhaps the crew of AF 447 would have benefited from another view.

xcitation

One thing that stands out is at circa FL350 the Captain should have taken his seat. Why would he not do that?
Why would a Captain decide to be only a back seat observer and commentator?
With my limited experience I have found that all the pro pilots (civy and military) are of the assertive/like to have control personality more than passive observer personality. (sorry I'm not a psychologist so I lack the fancy words).
Equally the PNF should have given his left seat back to his boss and insist he take it. Especially after the insane teenage nagging conversations he had with PF.
What am I missing here - is there an AF policy in an emergency for CAPT to stand back and oversee/bark orders?
I don't think trading seats would take that long given the adreniline.

jcjeant

Hi,

I think that at that time the rule in AF should be absent .. or if it existed .. it was not considered by some pilots
According to the Colin report .. it seems that a certain indiscipline prevailed (and perhaps still exists) in AF
Colin report (2006) available here (PDF french)
rapport_colin__juin_2006_.pdf

Extract

Raw google translation

wallybird7

It seems to me that what everybody is forgetting, is that the co-pilots have never had the opportunity to hand fly the aircraft at altitude nor in different modes/laws. And therefore all of the inputs were merely trial and error. At altitude and during turbulence and under stress.
How this could have been overlooked is literally unbelievable to me.

rudderrudderrat

Hi wallybird7,

I agree. According to: http://www.pprune.org/tech-log/376433-af447-132.html post 2632, the author points out the handling of the aircraft with UAS was different to the simulator.
I think the sims. will need better data to imitate the handling more truthfully.

CONF iture

No different as any conventional aircraft.
You did trim already in the sim when the aircraft was in direct law, didn't you ?
How was is any different ... ?

RetiredF4

O come on, you are kidding?

In a conventional aircraft you set desired attitude with stick input and relief the resulting stick pressure (either natural or artificial one) with trim.

would not work in an aircraft without feedback?

CONF iture

OK Franz, how is it different from direct law on the bus ?

1. Have you ever flown the bus ?
2. Have you ever flown the bus in direct law ?

RetiredF4

No, i´ve flown conventional aircraft with natural and artificial feedback,therefore i know that forcefeedback to the stick is an elementary necessity for trimming.

And i know also, that there is no feedback to the SS in airbus.

That is all i need to know, that trimming must be different.

I dont say, that it is not possible or more difficult, but it is not like trimming a conventional aircraft with feedback.

I would be very interested though in a detailled description, how it is done when autotrim is not available. I.e. concerning the pitch, do you steer pitch only with the trim like turning the trim wheel and look what it gives? Then readjust and finetune?

CONF iture

Same procedure is used with the FBW in direct law - As long as stick deflection is needed in order to maintain the choosen attitude, it means the aircraft is not properly trimmed.

To trim the Airbus feels as natural as trimming any other type.

takata

Is it another urban legend?



Thank you for mentioning that.

hetfield

Artificial feel - with a spring rod.

Aha.....

So it feels the same in a stall as with 340 kts.

3holelover

Artificial feel from springs is not the same as feedback. ...but yes, it should be sufficient to tell the pilot what forces he's having to maintain, and therefor, how to trim.

gums

Interesting about trim feel, and it could have played a role in Viper stick implementation.

The thing uses stick force per gee or roll rate, so it was real easy to trim off the control pressure to get a gee (gear up) or AoA (gear down). Max force was about 24 pounds in pitch and 16 or 17 pounds in roll. Neat thing about roll was the jet would command up to 180 deg per sec calculated aileron deflection at ZERO stick input to maintain zero deg/sec actual rate. This saved a jet and pl' Gums when one leading edge flap folded up on me after raising the gear one day ( see the pic in my profile bio).

So maybe force per aircraft movement could be emphasized for the 'bus.

RetiredF4

@takata and @ CONfiture

Please note again, that i didn´t comment at all on the subject, that the trimming feels like trimming an conventional aircraft or that it is worse or better or not possible.

I said (at least that was and is my intention), that it is not like concerning the process of doing it.

The SS has an artificial feel, i did not deny that. It tells you stick deflected NU, than trim NU to equalize, or vice versa. It´s not comparable to a tactile feedback (be it natural or artificial), where the amount of mistrim and therefore the amount of necessary correction can be felt in the hand on the stick due to stick deflection from neutral and force required to keep it deflected. It is therefore airload related feedback.

That is IMHO one of the main reasons, why there is autotrim present and necessary in normal and alternate law, as in those laws SS deflection in pitch results in a change of loadfactor, which is kept stable when SS is back to neutral (as long as protections do not intervene). If i understand the system, the elevator is positioned by the flight control computer upon ss input, feedback from the response of the aircraft is fed back to the flightcontrol computer (not the SS) and elevator position is again adjusted by the flight control computer. With some delay the THS trim will zero out the elevator deflection. During this process (except the initial SS input) the SS is in neutral position. How should a pilot zero the elevator demand with manual trim without tactile feedback? Look on a gauge with elevator position? On behalf of the aircraft reaction? Therefore the need of autotrim.

That also poses problems to disable the trim in special flight conditions in normal or alternate law, where to start and where to end the input of autotrim? It would solve one problem and cause some others.

In direct Law there is a direct relationship between SS deflection and elevator deflection (some other factors like CG and configuration and ??? play a roll though), but neither tactile feedback to SS nor to the flightcomputers. Only the feedback from the behaviour of the aircraft is present. But this feedback is dependent on how fast the aircraft reacts to the inputs (as in this stall very slow to react to ND inputs), therefore trimming process (trim, check, trim again, check, trim some more, uups that was too much..) would be slower and a process needing more attention.

Unfortunately (in my oppinion) AF447 didn´t go into direct law, otherwise the aircraft behavior and the trim behavior could have changed the outcome of this accident.

Hope that clarifies my statement somewhat, no intention to post urban legends or to dsicredit airbus or to join any anti airbus comunities. I´m a german tax payer and lot of german tax money is working with EADS.

I´m just trying to understand the system in its detail like i was used in the aircraft i´ve flown, but in looking on many posts with different understandings on two threads and reading nearly every available documentation from AB over a period of 2 years still keeps me mainly in the dark concerning lots of subjects.

And i´ve the impression that the crew of AF447 found themselves in a worse situation, because all the time they´ve got where a few minutes.

Lyman

Is it possible then that the autopilot was not sufficiently rapid to keep control? (447) What happens when 330 Auto can't be so quick as your Viper?

What happens if roll rate for gums is too slow by half when one slat fouls?

What happens to 447 when auto is too lazy? airplane is an airplane?

F4 For two years, no progress has been made on either side. You and smilin have nailed the problem each time, and each time it is ignored by "The a/c did what was directed."

It also did what was directed, apparently, by NOT trimming whilst the a/c was in an aspect commanding certain g. So if the Autotrim is disabled when the airframe is under maneuvering load, whty does it return "automatucally" (at the top of the gd climb, of all places), to seal the fate of the pilot who probably couldn';t have recovered anyway, even knowing the Auto trim was active?

Don't be sly, messrs. AB. Whether the pilot knew or did not know the Autotrim was active may well have been irrelevant. "Red Herring"? A "helpful" Gremlin in the tail cone, with a prod to mystify the crew? Changing the handling of this airframe in ways not EVER practiced by the humans?

gums

My point about force versus deflection is that the stick felt more like what we were used to. We trimmed off pressure not stick deflection. the thing only moved about 1/8 th inch.

Unless the 'bus springs are weird, it should take more "force" to move it further, so manually trimming off "pressure" should not be a big deal.

In my case, the jet commanded a left roll to counter the loss of right wing lifties and roll due to yaw. That sucker had lottsa drag. So I had to add more control surface deflection, and post-flight analysis indicated I had about a pound or two force to play with. In short, I basically held full left stick for about 10 or 15 minutes.

As other pilots point out, the ideal stick would require increased pressure/movement according to dynamic pressure. Just like old days when you could feel the forces on the control surfaces due to those air molecules. Go fast and stiff stick, slow down and stick gets mushy, as does the plane, duh.....

If the force feedback system fails, then you have the basic spring/torsion tube/whatever built into the thing. No big deal.

The problem with the 'bus is that a small continuous input from neutral eventually causes the HS to move ( ours didn't do this - it always wanted to get back to the manually set trim gee or trimmed AoA). The autotrim is neat when maintaining level flight or a climb/descent, as you can gradually relax pressure.movement to get the desired attitude or AoA.

However, in this case it helped the plane to achieve an ever-increasing AoA due to speed decay and the "one gee" default trimmed gee condition. Had the pilot used only roll and a neutral back stick, the THS would not have continued to command nose up.

Once zooming thru all the AoA protections with back stick, the plane entered uncharted territory and only a few folks would have understood the exact aerodynamic condition and need for manual down trim and forward pressure. Despite some folks' thots here, I don't think pure elevator can overcome the pitch moment that the HS can exert when fully deflected one way or the other when flying slow. Mach effects could have been a player initially, but once slow enough, they probably were not a significant factor.

Lyman

"I basically held full left stick for about fifteen minutes" That sounds eerily familiar to 447. Drag? Right wing drop (Chronic and/or acute?)

Don't forget though, that PF could not "forget about aft stick", he had a bobbing nose to lift and drop to maintain "level" and avoid a SPIN. He heard the initial Warning, and if he was a pilot, he knew instinctively if he hears the horn, he gaddam better keep the wings level.

He "started" his "climb" WITHOUT AUTOTRIM, for once he exceeded 1.25 g, the slab inhibits (g prot). After his load (g) diminished, the Autotrim started up, at the top of his "climb", did he know? Could he have compensated if he did know?

The entry to manual reversion is not addressed at all well. Someone take a breath and discuss what may have happened if roll rate demand was beyond AutoPilot, and did the "Upset" (at handover) corrupt the Speedata?

Honest question. Was Pf "playing the g prot" his inputs were not held initially, he appears to have been "bumping" the elevators. How does he lockout Auto Trim? His full back and held inputs come later, AFTER the THS has planted on its Stops? Is mayonnaise a technique to lockout the slab by pilot device, and not in the FCOM?