HazelNuts39

Maybe slightly but, IMHO, with full nose-up elevator maintained, not enough to unstall the airplane. I don't expect the airplane to have the straight-wing characteristic of a nose drop that can't be arrested.

Lyman

Who would know re: neutral airframe, with the Autotrim busting a gut to keep the flight Path, ?

I don't think the A330 was tested at STALL with full nose up trim? Who would have the balls to do that? And if the a/c can get to STALL by having the flight path seeking it, why isn't the a/c in DIRECT at this point?

If AUTO STALL is available, let's test it in the certs. And at least consider that some additional cue might be helpful...

Lonewolf_50

Lyman, I'd like to revisit a fundamental task drilled into pilots from formative days, formative hours. This is in reference to the "mayday call" idea and what raises or lowers the nose.

The fundamental? Aviate. Navigate. Communicate. Given the trouble with the first, any pretense that the last -- calling Mayday or whatever -- has any importance is incorrect.

Aviate.

Pilots are (or should) control the aircraft. If it does X, and X isn't what should be happening, the pilot, using the systems at his disposal, makes it do something else. That is the premise of "controlled flight" versus "uncontrolled flight" or even "passenger in the very front seat."

While the THS certainly helps find a pitch attitude commanded by the system, there is a certain monotony to repeating again and again that the pilots inputs are part of the system. Based on what the FDR data found, the input from the SS was not cut out leaving the pilot at the mercy of the robot.

Why the pilot at the controls kept the nose up is in part (in my mind) explained by the idea that he latched onto the FD as a primary scan item, rather than flying a more primative scan based upon attitude indicator (artificial horizon if you like) compared to other instruments, which takes us back to opinions rendered here on PPRuNe while the BEA were still trying to find the wreckage site deep in the Atlantic Ocean: the Pitch and Power chorus will once again seranade us, since pitch and power comprise fundamental concerns of the fundamental number 1 noted above.

Aviate. Why that broke down seems to have a far heavier HF component (culture, training, and more) than mechanical component, though the complaints on how nuanced the control laws can become in the variety of degraded modes seems to have some merit.

Lyman

At four thousand feet, descending at 120 knots, I might consider a last will and testament, though a short one, an aural one, with a keyed mic.

Lonewolf. From the outset of the report of the loss of this a/c, I suspected the pilots were betrayed by the aircraft. I did not necessarily conclude they were without blame, but something did not add up. One of my hundred or so theories early on was that the crew followed the 'bird' into the drink. If ADR can fail, my mindset would be to fly, and trouble shoot. ANC is swell, but sometimes they get rearranged, and there was no teamwork here. I frankly did not suspect any, Teamwork is disincentivised when troubleshooting is trained to be snail paced, and recovering the flight path is URGENT.

Get out the freaking BOOK?

."what raises or lowers the nose."

You forget, the NOSE is secondary. What raises and lowers the tail is the deal.
My instructor told me to be aware of the cause, not the result. Looking at the nose instead of thinking about the tail?

DozyWannabe

Agreed.

Which was the point I was making. Autotrim was trying to compensate for the simultaneous fall-off in airspeed *and* the commanded nose-up attitude, both of which were directly attributable to the PF's backstick commands.

@franzl - If I had to guess, I'd say that the autotrim limitation might have been dropped for the widebodies due to criticism from some pilots over not having full authority through the PFCs in Alternate Law.

Lyman

Quote:
Once PF had commanded a pitch attitude which resulted in a real (not just spuriously indicated) speed decay - even if he didn't touch the side stick again - the FBW computers would attempt to maintain that trajectory by moving the elevators and hence the stab trim.

Even if the pilot had let go the stick following his first input of NU, the a/c would have flown into STALL. This means the pilot is constantly responsible for Pitch, but cannot relax the stick either, setting up a lethal and terminal misunderstanding of flight path in an upset a/c. Peachy.

Alternate Law in upset is absurd. And deadly.

Machinbird

Lyman
Come on. Which end of the teeter totter is going up? Completely irrelevant point IMO.

Lyman

Bear with me Commander, my teacher had an angle, and it served me well. Step back, and see if you do not agree. Airbus and aviation are dumbing down the pilot pool.

Jim G was a Naval Aviator. Corsairs. The ones with propellers. He taught me to constantly assess what the aircraft was doing, and the controls part in it. He may have said something like, " if you think something is wrong, it is. Otherwise you don't understand your airplane." He taught feel. If he hadn't flown his Baron into that building, he'd be all over this case.

(get home- itis.)

gums

Here's the snip from the FCOM I have been given from one of the pilots here:



I can't find any clear reference to "auto-trim" until "Direct Law".

Note that the THS seems to follow pilot inputs except when AoA "protection" is active, and in our case HAL decided AoA was irrelevant. GASP!

Maybe A33Z can help here.

So I'll throw my vote in with several here that the THS auto-trim feature did not help the situation, and may have contributed to the stall entry and remaining in the stall unless the pilot commanded nose down/lower gee than "one".

I continue to iterate that the 'bus system is not an "attitude" command but a gee command. Because HAL corrects for pitch attitude, the basic one gee command is "corrected". So you think the jet is trying to maintain an attitude, but that's just a result of the corrected gee comand. Additionally, without auto-throttle, the only speed stability is when drag equals thrust, and it is not what we old dinosaurs were used to when the plane tried to achieve the trimmed AoA ( and resulting drag versus lift properties).

So way I read things, the plane would have trimmed for one gee corrected for 10 or 12 degrees pitch even if the pilot let go of the stick. The THS eventually reaches max and.........

Clandestino

Issue with BEA investigators is they had expert help at hand, they correctly understand how Airbus FBW works and are fully aware what "acceptable by certification authorities" means.

PPRuNers have no such resources readily available, therefore I find it unsurprising that some only slightly wrong theories about how Airbus FBW works resulted in quite flawed assumptions being passed as facts, leading to another oozlumistic round of suspicious glances being cast towards BEA, DGAC and Airbus.

Let's get to basics of aircraft stability: conventional certification rules state that no passenger transport aircraft will get certified unless she displays adequate static and dynamic stability in pitch. Static stability means aeroplane, when disturbed from trimmed speed, will return to it stick-free. Dynamic stability means it will do so in acceptable number of convergent oscillations. Stability is not just a matter of aerodynamics; control forces, controls architecture and power all affect it so it can be (and mostly is) significantly modified by FBW. FBW Airbi can not demonstrate hands-off stability in normal and alternate laws as due specific settings of their FBW they will not make an attempt to return to trimmed speed after the stick is released.

As for how Airbus FBW works, lets first start with how it doesn't. It was said it maintains 1g. It was said it maintains pitch. It was said it maintains alpha. In normal operation, none of this is true. It might maintain all three of it but not because it pursues them actively but only as a consequence of trying to maintain constant flightpath when stick is pitch neutral.

So how it works for the pilot? First we'll assume steady, level manual flight with ATHR in SPEED mode (to digress a bit: most of the world flies this way while my company was pretty adamant: manual flight - manual thrust, no matter if it is Boeing or Airbus or whatever). If there's moderate pull or push and then release to neutral, FBW will maintain new flightpath, autothrust will compensate with thrust change, aeroplane will feel pitch neutral. So far so good.

Now let's assume ATHR is off, aeroplane is flying straight&level and a couple of degrees of pitch up are introduced while thrust remains constant. Aeroplane tries to maintain the flightpath but as some power now has to be used to climb, drag causes the aeroplane to decelerate. As it decelerates, it needs more AoA to keep the fligtpath so FCS introduces a little more pitchup on its own. Aeroplane feels slightly unstable but if power and pitch are reasonable, equilibrium point where the new flightpath can be maintained without further increase in alpha is met. Still good.

If we take extreme case, where aeroplane is flying straight & level and TLs are moved to idle, aeroplane indeed starts to pitch up on her own as the speed is bled off. How long does it pitch-up? Until alpha prot activates in normal law, low speed stability kicks in in alternate1 and introduces pitch-down or it stalls in alternate2.

So why is it all so seemingly complicated? Simples. To make life easier for mister pilot and provide him with perfectly conventional control response within the envelope. There is absolutely no need to think about flightpath or trim. Stick forward moves nose down. Stick backward moves nose up. Stick left rolls the aeroplane left. No points for guessing what does stick right. Many a claim was made on this rumour network that oh-so-complicated-Airbus-took-the-controls-away-from-me-when-it-shouldn't-have. None of them were substantiated except St.Johns and Bilbao - that's fixed now.

While the nomenclature between SA and TA Airbi is different, principles of protections are pretty similar. We didn't call it ALT1 and ALT2 when I was on 319. Their rough equivalents are "alternate with reduced protections" and "alternate without protections" (nitpickers, welcome).

How do I know AF447 wasn't deep-stalled? Well, I started by paying attention in high school science classes, which helped me to understand aerodynamics lessons during my flight training. One of them involved deep stall and I learnt it has specific meaning: it is stall which cannot be recovered by conventional means such as actions on elevator, stabilizer or power. While AF447 never fully recovered, mere reduction of power or just moving the elevators towards the neutral (while never reaching it) abruptly decreased its AoA - which is incompatible with deep stall as we know it. Now if your goal is to be taken seriously in discussion, you won't go a long way if you take well established terms and start assigning them just your own definitions.

If it were true, then dr. Evil Genius who invented should be stripped off of his PhD! He has less than three percent success in catching the crews in his trap!

Seriously, these unsubstantiated claims are so repetitive I can't even find them humorous anymore.

Is it news for anyone else?

Hollywood's bravo sierra. In their minds, crew was fighting for their lives until the end, there was no time for pathetic farewells. That they completely misunderstood their enemy and eventually committed the suicide by incompetence is tragic, not stupid.

Harsh words, but if we assume they're true then so were others. Difference is they choose more appropriate response, even if it meant doing nothing at all while been confused about what is happening.

Using brain instead of balls tends to produce better outcomes. Like not even trying what was computed to be fairly suicidal.

Looking into that matter could turn out to be very expensive... not just in Euros but also in lives.

No. Reference, please.

Glad I work for abnormal outfit.

Panic.

When ECAM works as designed and is faced with simple faults, it covers everything you need to know and do. No more, no less. No point putting "pretty please with sugar on top" on ECAM lines.

Yup. Proper flightpath control was never achieved and that always takes precedence over ECAM actions.

I suppose it is even more appropriate than you assume it is.

Perhaps it is leftover in our DNA from the times our ancestors were preyed upon by cave hyenas and short-faced bears but there are two main types of panicky reaction: getting frozen (maybe the hunter won't spot you if you don't move) or running away in whatever direction (this might help confuse the predator or make it catch someone slower). Seems to me that it was very unfortunate set of events that both pilots simultaneously fell victim to panic. CM2 was trying to run away from the danger he sensed was coming from below and had no understanding for the aeroplane's energy state while CM1 got too scared to intervene. IMHO another important aspect of your story is that your comrades, while doubtlessly well trained, were simultaneously faced with realization that they are not in training anymore, that (unlike drill sergeants who just play-pretend) folks out are there are really bent on killing them and that not everything is going according to the plan. Sudden realization of one's own mortality can easily set the atavistic instincts in motion.

Lyman

Is it news for anyone else?

Clandestino. You have the luxury of your assumption that the pilot was a moron. Or that I am, or anyone else who thinks luxuries such as yours are Ill advised.

I meant that the pilot is responsible for constantly monitoring the stick as well as the plane. You ass ume he had full complement of displays, and was ignoring pitch, in favor of roll? These are easy peasy, not attested to by the report, and are frankly tiresome. Perhaps as tiresome as my insistence on reading more data?

In an upset, and without dependable instruments, my guess is your drawers would fill up. I truly appreciate your style, with some exceptions, but draw the line at your holier than thou nonsense.

Lyman

the aircraft's nose may have fallen through,

HazelNuts, I did not mean the Nose would fall and everything was ducks, I meant that as one of only two airframe cues for STALL, it's lowering against the stick would have some warning value to the pilots, as intended in a normal aircraft.

Bear in mind even BEA had to ask experts if the accels in the DFDR traces were buffet, suggesting the pilots missed that one remaining cue, perhaps missing it in the turbulence.

In the approach to Stall recovery that is trained, the nose does not drop, so a non dropping nose and no buffet may have meant: " I knew it, that stupid Stall Warning was bs..."

rudderrudderrat

I can't imagine how you were trained to recover from a stall warning without a briefing - but I'm glad you work for an abnormal outfit as well.

gums

Very good post by Cland, and a lot of agreement from this old FBW pioneer.

I was particularly impressed by this snippet:

I would only point out that "stability" in the FBW jets is still a basic aero function of cee gee, center of aero pressure, moment arms and capability of the control surfaces, and on and on and on....

As Cland pointed out, what we see is "apparent" stability. The FBW system works to modify the inherent aero stability (or lack of) in order to provide the basic pilot a plane that he/she can fly. I had a chance to fly the AFTI sim ( modified Viper) and see various control laws that were enabled with the software. The body rate law was very foreign, and was quickly passed by. The attitude law was same same. The pure AoA-biased law felt really great for any old pilot. But the best law was one based upon a blended AoA-gee. Sheesh! It was close to the one that GD had come up with ten years before. The AoA limits helped to prevent stalls and such. The gee limits kept us from ripping the wings off.

Due to our mission requirements ( Doze will jump on this, heh heh) and our immense speed envelope, we did not require "speed protection". In fairness, we had one troop die when diving down at a speed beyond the placard value, but the airframe was solid and the motor exploded. Vne was 800 knots CAS or so, and low-speed warning was about 120 knots CAS, so see what I mean?

IMHO, the 'bus designers learned from us, and their overall concepts seem very reasonable and directly related to the mission of the jet. Only thing I would have balked about was the AoA role and the plethora of reversion sequences.

Cland's description of what the pilot sees and senses is spot on. We didn't have the auto-throttle feature, so our workload with the throttle was much higher than holding an attitude. We stabilized speed when the drag equaled the thrust.

PLZ remember that I abandoned the "deep stall" theory after a few documents were made available to me. Don't know how old Cland is, but my high school aero and the stuff I could read at the library didn't approach this characteristic of some planes. Wasn't until the 727 came out that we saw a discussion. I then flew the VooDoo about 6 or 7 years later and we had "pitch-up", but it was complete LOC and the jet did not remain fully stalled in any fashion. Sucker tumbled in pitch with a super roll/yaw component.

So 15 years later I flew a jet that had the classic "deep stall". Funny, but some of the entry maneuver had FBW control system aspects. The maneuver was exactly the same as AF447. Climb at a pitch attitude and power setting until the jet ran outta the air molecules required to get the nose down with full elevator deflection ( no THS, as we had the all-moving horizontal stabilizers). Super directional stability like the 'bus seems to have, and no large pitch changes - just a reasonable ride while the altimeter is unwinding at 10,000 feet per minute. Fortunately, we had a positive pitch moment at that combination of cee gee and AoA. Unfortunately, HAL was commanding nose down and the jet couldn't get there. So our wizards gave us "direct law" in pitch via a switch that would not even work unless HAL senses AoA above 30 degrees. It worked and we didn't lose a lotta jets.

Thanks, Cland, for a great post.

PJ2

Agree...refreshing and well worth reading Clandestino, thanks.

Lonewolf_50

For Lyman:
The stick.

If that isn't what's happening, you have a flight control problem when you are sitting in the cockpit.

Your Corsair veteran IP would have agreed.

mary meagher

Going back a couple of pages, to the discussion of the "startle factor".... we were told it is not so simple to include genuine fear in sim pratice...

I'm sure somebody has mentioned Ernest K Gann and his experience of an instrument approach on a nasty evening, with the captain lighting one match after another under his nose. Gann managed to maintain his concentration despite this insane behavior, which made him better able to deal with difficult situations.

I have been startled on a couple of occasions; my response was to keep calm and do nothing; controls still OK, loud bang turned out to be unrequested descent of the undercarriage and not a mid-air. In another case, lap strap out the door began a rattatat on the fuselage. In the third case, the glider on tow suddenly appeared beside me, instead of behind; in all cases doing nothing turned out to be the correct response. But believe me, I was startled alright!

DozyWannabe

Not at all - speed protections would be inappropriate on a fighter, not to mention an absolute pig to implement!

I'm sure they did, along with everyone else who'd tried it!

If you'd looked across the pond at around the same time you'd have seen it become a very big deal in the UK (home of the T-tail) just before then - the 727 owing more than a little debt to Boeing's visit to De Havilland during Trident development. To this day I think it's the only large US-made airliner with a T-tail - and thus susceptible to the full-blown deep stall phenomenon. Deep stalls had claimed the lives of several prominent British test pilots before the characteristics were fully understood.

That said, the original design for the 757 had a T-tail, and ironically it was British Airways' demand for a conventional empennage which changed that.



PS. Lyman - he didn't call the pilot a moron, he said that the response to the situation was incompetent. The former is a personal insult whereas the latter is a tragic fact - it's important to be aware of the difference.

OK465

If you light a match in modern simulators, it will set off the fire suppression system.

jcjeant

Indeed ... it's politically or diplomatically correct ....
But what do you think (first thinkink) about people who are making incompetent decision ?
They are "moron" ... or "smart" people ?
So I understand that Lyman can use the word "moron"