Gysbreght

I must admit that I'm not expert on this, but I believe system failures are addressed in FAR/JAR 25.1309. The general principle is that there must be an inverse relationship between the probability of a failure condition and its effect on 'continued safe flight and landing' of the airplane. This implies that an assessment must be made of the ability of the pilot to deal with the effects. The FAA and EASA have both published extensive guidance material on how this principle is to be applied.

Owain Glyndwr

@Gysbreght

That is essentially what I am saying - handling qualities must be matched to probability and seriousness of failure consequences

RetiredF4

That is a vital question. I think it is a mix of not sufficient knowledge about general aerodynamic matters and missing energy awareness. In all our discussions there surfaced another obvious fact, that the normal documentation does not contribute to the knowledge level. The manuals are comparable to instruction manuals without comunicating the basic knowledge layer behind those instructions.
I do have no knowledge if the quality of the manuals for pilots, instructors and maintenance is part of the certification process, but there should be at least a minimum standard established for those documents and i would wish that this standard is above that one which it seems to be now.

The manuals i was used to in the military were gold compared to the dust of the AB manuals i could get my hands on.

CONF iture

He is aware of maintaining some back pressure on the stick, information positively known only by him due to the SIDESTICK concept by Airbus.
At no time it is getting tougher for him to maintain back pressure due the auto operation of the THS.
That THS operation has been a main actor in the fatal ending of AF447.

DozyWannabe

They're probably just trying to be thorough - obviously in this case the stall was preceded by distinct NU input on the stick, but they're probably asking for some clarity on what should happen if that isn't the case, and how the special regs conditions deal with that.

Regarding Direct Law, as it essentially bypasses all the rate demands and the automatic trim, would it not be the case that the jet does have longitudinal static stability around the current trim setting, as would any other airliner in that configuration?

You can only reasonably make that claim if there was a distinct and obvious effort to recover from the stall and the THS position prevented it - this wasn't the case. Furthermore, the THS was in that position because the pilot put it there, knowingly or not. Robert might not have been able to see Bonin's SS, but if he'd looked across, he'd definitely have seen the trim wheel rolling backwards.

Owain Glyndwr

@RetiredF4
There is a problem in that the manuals are not issued to the crew by Airbus.
In that “Counter Experts” report you will find the following (my apologies for leaving it in French, but I don’t want to get into any hassles over mistranslation)


So there you have it – there was a more complete guide to the situation, but AF management declined to send it on to the people who might benefit from it.

@Dozy

Quote:

Regarding Direct Law, as it essentially bypasses all the rate demands and the automatic trim, would it not be the case that the jet does have longitudinal static stability around the current trim setting, as would any other airliner in that configuration?

Reread the first and third paragraphs of my posting #752. I think you will find the answer there.

gums

Thanks, OG and Retired.

The thing I have been trying to get across since I was accepted by this august body is the underlying control algorithms and their implementation are not clearly explained in the FCOM's that I have ( flare mode one exception, maybe).

I still see pilots talking about the pitch attitude implementation as if the control law was only a rate command. As Winner and Palmer pointed out, the jet uses both rate and gee to get the nose moving. Once you let off the stick, the jet adjusts the rate/gee command "corrected for pitch attitude" . You get what seems to be attitude hold, and until you have several failures you stay at the last pitch before releasing the stick. Would seem to me that a pitch rate command of zero would apply when not pulling or pushing the stick. You would still get actual movement of the elevator, and then the THS to maintain zero rate, however.

That surface movement was what I demonstrated to all my students. Our rate function was minimal at normal speeds, so it was almost all the gee. Since we could trim for a gee, we would move the manual trim wheel all the way back for about 3.5 gees command while holding forward stick. At a realistic speed we neutralized the stick and the jet would do a beautiful loop. As speed slowed, AoA went up and stopped at the limit. We then rode the AoA limiter over the top and down the back side until 3.5 gees. All the while you could look back and see the horizontal tail "trimming" until it reached the mechanical stop, then going back to its original position as speed increased on the back side.

So that's what happened to the 'bus, Doze. The THS would have kept moving even with a neutral stick ( pitch) to reduce elevator position. I don't feel the elevator would have to be "maxed out", but the system would try to reduce its position just like we used to do with manual trim wheels and actual mechanical "feel" to our yoke/wheel/stick.

The insidious effect of the implementation and the lack of speed/AoA stability until in direct law was a very big player. And no AoA protection!!! So the jet flew right into the stall once at the pitch attitude, and the pilot's pitch inputs might only have been important if nose down for several seconds ( as we have seen by those here in the sims).

Glad to see enough here understand the longitudinal stability issues.

BOAC

So, we are back to having something saying "That's enough mose-up THS, boys and girls - if you really want more push this over-ride button" - what is wrong with the concept?

EDIT For 'mose' read 'nose'

DozyWannabe

Only below 1.3Vs, if I understand OG correctly. The pitch/power settings in the supplementary UAS procedure would have kept them above that, had AF thought to provide that information.

Regardless of this factor, the DFDR trace clearly indicates that the THS position change was neither gradual nor "insidious" - it was a direct response to repeated NU demands from the SS, including the slam against the back stop.

@BOAC - the A320 has a trim limit in Alternate, and according to OG it will also halt further NU trim in the event of a Stall Warning. Why this wasn't carried over to the A330/340 is a question for Airbus.

Owain Glyndwr

No Dozy, the THS will move to reduce sustained elevator deflection at any speed

DozyWannabe

Right, but isn't a neutral stick effectively zero elevator demand? I'm probably missing something here...

Owain Glyndwr

No its not. If the pilot makes a change of state that requires elevator deflection and then releases the stick the elevator will remain at the deflection required to maintain the new flight condition until the integral term causes the THS to move and replace the elevator with THS deflection. Neutral sidestick would be zero elevator in long term but not transiently

BOAC

- perfect! So, 2 things will happen

1) The crew will know they have reached a practical, 'sensible' limit

2) 447 would have eventually run out of nose up elevator and the nose would have dropped.

Bingo!

BOAC

- are we talking deflection or effect? I have no knowledge of AB THS angles, but what would full deflection at 2ANU achieve? I see mention of THS 13 in AF447 - is that 'sensible'? Presumably CofG and other factors could be used to compute a sensible limit? Remember I am arguing for a 'wake-up call' to the driver, and would full back stick still produce the expected response if the THS had been 'stopped'? There apparently needs to be some way of alerting a crew that auto-trim has gone beyond reasonable bounds and that shirley summat must be wrong?

PS I now have a line in spring-loaded boxing gloves if you want any.

DozyWannabe

@Winnerhofer - Boeing's "Synthetic Airspeed" is essentially the same deal as Airbus's BUSS system, presumably with the same or similar limitations.

Bpalmer

Yes! That's a point that is often missed and underemphasized in learning to fly the bus. The airplane is essentially point-and-go. If you pull back on the stick for a few seconds and let go, the airplane pitch will stay there. If you're in cruise flight (that's a lot of forward energy), you can easily be going up at 6,000 feet per minute quite quickly. (this happened in another very interesting—yet non-fatal incident between an A330 & A340 in October 2000 where " the aircraft climbed rapidly (reaching a peak rate of about 6,000 ft/min) due to the increase in lift created by the flight control system's capture of alpha prot. " Now, normally the protections will prevent you from getting into trouble as a result, (with rare exceptions like the above where it was arguably the cause of the trouble, but if the protections are degraded reduced (to stabilities) or absent (versions of alternate 2: AF447), then it's all up to the pilot to establish a reasonable pitch attitude. The natural aerodynamic stabilities will not do it.

Direct law is much more like a 767. But the stick forces are light - they are always the same with the same displacement from center as there is no artificial feel system. Flying in direct law is a matter of using the trim to avoid having to hold the sidestick out of the neutral detent. It's like some cars that have the power steering set so it has virtually no feel and takes no effort on the wheel. It handles fine, you just use the trim wheel manually so that you can have the stick neutral as you fly. (though still primarily controlling pitch with sidestick)



It's also interesting note that in alternate law the stabilities are actually the airplane's natural stabilities. The pitch is in essentially direct law. The effect is subtle—due to the absence of an artificial feel system requiring tens of pounds of force one might call them "weak"—but at least there is a tendency to pitch to the trimmed airspeed—which would theoretically be higher than the airspeed that results in trouble—but who can guarantee the crazy scenarios one might conjure up. But, the activation of those stabilities is based on airspeed as a trigger. So, when airspeed is thrown out of the equation as it was in AF447 (due to disagree and a change rate that was deemed invalid), the stabilities aren't allowed to become part of the flight control law and stall prevention and recovery is all up to the pilot again. G-load demand is a crappy flight control law to be in for stall recover. When the airplane starts to fall (accelerate downward G <1.0) the airplane's reaction to maintain a neutral-sidestick command of 1.0g is up elevator, followed by nose-up stabilizer. ooops.

gums

Awwwwww, man! After all these years I read from an esteemed contributor

In these cosmic FBW jets, the control surfaces move wherever they have to in order for the jet to achieve the "trimmed" or "desireable" gee. Not attitude or bank angle or speed, but a mix of rates, gee and maybe AoA.

Okie has tried to explain a lot of this with a mix of pilot lingo and engineer stuff ( and I think he had some Bk 10 Viper experience). I tried from first seeing the wreckage photos and then data recorder traces. Looked like a classic deep stall impact. The c.g. charts I then got showed it was damned near impossible to get into a "deep stall". "deeply stalled" was possible when in reversion modes such as AF447.

So I abandoned the "deep stall" line of reasoning and tried to see how to stall the jet that many believed could not be stalled. Turned out to be the same techniue as with the Viper. Achieve a decent pitch attitude, run outta airspeed faster than the manual or "protected" laws and control surface deflections can work to achieve the desired gee/rate and voila!

P.S. @Winner synthetic speed and use of AoA indications prolly shoul be on thread of its own.

g109

Excellent article.
..... The airplane was in the control of the pilots, and if they had done nothing, they would have done all they needed to do.....

Should Airplanes Be Flying Themselves? | Vanity Fair

CONF iture

That would be still pitch under alternate law but without any THS involvement.
What would be sensible is pitch under simple direct law where the pilot has to force to keep the nose up and any release of the stick would let the nose drop.

Very big indeed

jcjeant

g109
Except Derosay (Pilot Airbus Industrie) .. what is the pilot that will touch nothing ... when the autopilot turns off .. the pilot finds by instruments that the plane suddenly lost altitude and that he quickly takes a tilt
At Airbus we already had Ziegler as saying that the plane could be flown by his concierge .. and now we have Derozay who said in a documentary .. the pilot must touch nothing ...
I hope that the pilots are not going to learn from these two ridiculous comments