rudderrudderrat

Thanks jcjeant,

"The BEA is to establish the working group "Human factors" whose creation was announced at the publication of the third progress report of investigation into the accident flight from Rio to Paris on 1 June 2009.

This working group aims to analyze all aspects related to the conduct of the flight:

actions and reactions of the crew over the last three phases of flight described in the third progress report, especially vis-à-vis the stall warning;
ergonomics of the cockpit;
human-machine interfaces.
This working group is composed of seven experts:
three BEA investigators specializing in Human Factors;
a psychiatrist expert in risk analysis;
Aviation Human Factors consultant;
a qualified airline pilot A330;
a test driver.
The BEA may call occasionally to other experts and consult, as appropriate, Airbus and Air France.

The group's work will start very quickly and should be completed by the end of December 2011. His reflections will be fed by the work group "Operations" and group "aircraft systems."

All these works will be recorded in the final report which will establish the causes of the accident and whose publication is expected in the first half of 2012."

Thanks Google Translator.

Jig Peter

The team(s) "doing" the Airbus FBW system did not include "just engineers" but also pilots, some of whom also had high engineering qualifications.
Also, Airbus FBW wasn't a "one-off" programme, but the logical continuation of the control philosophies on preceding programmes, and many of the members of the FBW teams had been involved with those programmes.
Don't go along the "uncontrolled mad programmer/engineer" route - it doesn't wash !

ChristiaanJ

Thanks Peter....
I's vexing sometimes to see those comments from people that have never designed a flight control system, and that have no idea how such a system works....
No guys.... the 'mad' aviation 'engineers' mostly went out of fashion in the early 1900s.

Linktrained

Machinbird #790

About sixty years ago some high performance gliders were fitted with a " Total energy Variometer" which would show the relationship between speed lost and height gained (or vice versa). These would have been non-electric !

Ian W

There are no mad engineers, and there are many very dedicated people. However, at each stage right down at low levels in the system assumptions are made.
For example: "No aircraft will be airborne if its speed is less than 60 Kts as read from the pitot tubes". There will also be misunderstandings coded perfectly into the software. It seems from reports here that outside air temperature was considered to be a stable variable - yet it can vary hugely very rapidly as you fly through warm updrafts and into cold downdrafts - making Mach number calculations for Vmm and Vmo unreliable.

These 'features', quirks and problems, sometimes based on well thought out good intentions, appear in every major software system. They need to be looked at and the logic flaws carefully fixed (or you insert even better bugs). It does not pay to be defensive about these problems - they exist. Unfortunately, some of these 'features' if occurring in rare combination (back to the holes in the cheese) can be very misleading or surprising for the line flight crews: Even if the reasoning behind them still makes perfect sense to the careful design engineers and flight test crews.

xcitation

"Alternate law and protections lo"(st) was verbalize by PNF after a slight delay from its onset. PF did not verbally acknowledge this. PNF did not press the issue.
PF was not to be aware of his climb to max altitude due to his stick back.

DozyWannabe

To be fair, that's a hypothetical extrapolation you're making from the rules and there's no guarantee that was the logic behind that decision.

Ian W

It may be a 'hypothetical' but almost every other aircraft in the world would use the undercarriage 'squat switch'. Someone made the decision to not use that. I am sure that there is a squat switch to send the 'OFF' ACARS message.
\
But it was just an example of assumptions and decisions buried in complex systems that are defensibly logical but which can have unexpected consequences.

gums

@ Dozy, I am gonna go with Ian W on the weight-on-wheels design consideration.

If you want both < 60 knots and WOW, a simple "and gate" works. If you want either, then the "or gate" works.

OTOH, seems the 'bus and other heavies use WOW for spoilers and such after touchdown, or am I off base? I'll bet most pilots would like the spoilers to be active above 60 knots, as they wouldn't do much to shorten the landing roll below 60 knots, ya think?

I also question the alpha protect sub-law that attempts to fly the plane at the alpha protect AoA when the stick is released/neutral versus the basic one gee command.

And there are other puzzling features I have seen in the manuals regarding the reversion laws.

rudderrudderrat

Hi Gums,

I agree with you and Ian W - simple WOW logic for stall warning activation.
If I'm airborne - I want to know if the angle of attack is excessive no matter what my indicated air speed tells me. (There is no mention in FCOM about the >60kt IAS logic)

If only it was that simple!

This extract is from Airbus Safety First Magazine Feb 2010: (my bolding)

"3.2. Hard landings
Among hard landings, one specific category has been identified where by the hard landing occurred after a bounce. They fit to the following scenario (fig. 7):

No engine throttle reduction (retard) during the flare
No ground spoiler extension.
Bounce induced by a too high energy level and by the lack of lift destruction.
Engine throttle reduction performed during the bounce
Ground spoiler extension if the retard is performed within 3 seconds following the first touchdown.
Severe hard landing due to sudden loss of lift leading to a fall from a height of about 5ft to 15 ft.
It has been established that most of the hard landings occurring after a bounce are severe."

AlphaZuluRomeo

Already said, but anyway...

I've never seen an official (manufacturer's) document which stated :
If IAS < 60kt then we're on ground then stall warning should be off.

I fully concur with DozyWannabe : that's a hypothetical extrapolation you're making from the rules and there's no guarantee that was the logic behind that decision.
If you know of such a document, please let us know.


My guess about the logic stopping the stall warning:
- nothing to do with being on the ground : for that, as was said, re-said, re-re-said, test wheight on wheels.
- if IAS < 60kt, then assuming the AoA probe risks (too much to be acceptable) to output false/inaccurate data.
- then if IAS < 60kt, disregard AoA probes data as unreliable
- then you don't know your AoA, then you cannot have stall warning


I don't pretend this is really better (factually : it was not in AF447) but at least, before being sure, let's not assume Airbus folks just "forgot" WoW... It's as bad & unfair as saying that AF447 crew were idiots, and that all is their fault.

Clandestino

Humour is a matter of personal preference, yet I must state that I find your excellent example of straw man argument not particularly funny.

If definition of knee jerk includes "quick, large, fast, precise, with positive feedback loop", then knee jerk reactions are required while flying in extreme circumstances, such as when one is getting cornered and prompt resolution is required, e.g. stall warning, GPWS maneuver, windshear escape. If not, forget about it at all. Whether my proposition of knee-jerk definition stands or not, UAS wouldn't call for knee-jerk reaction even in your wildest dreams.

Can we agree that half back stick, progressing to full was unwarranted and maladjusted knee jerk reaction to UAS that untimely terminated the lifespans of AF447 crew and passengers or are we going to attack the BEA's findings, so far?

Do you have a slightest idea what effects your cure has, when you pronounce it worse than disease or are you disregarding them and are just into concentrating on side effects?

That the aeroplane with size and weight of A330, at 60 KIAS will be soaring through the air with grace and elegance of falling refrigerator and that therefore her airborne time at said speed will be most cruelly limited is not just an assumption. It's a fact.

He could have checked attitude and altitude. I am puzzled why he didn't.

Brief summary of flight controls effects in Alt 2B, provided the aeroplane stays within flight envelope:

STICK FORWARD ........................................ NOSE DOWN
STICK AFT ................................................ NOSE UP
STICK LEFT ............................................... ROLL LEFT
STICK RIGHT ............................................. ROLL RIGHT
PEDALS LEFT ............................................. YAW LEFT
PEDALS RIGHT ........................................... YAW RIGHT

Ladies & gentlemen, aren't you forgetting the basics of instrument flying? Namely that successful instrument flying absolutely requires: idea where is aeroplane, where we want it to be and strong positive feedback loop between pilots actions and aeroplane reactions? Aeroplane banks and I don't want that? Give opposite command via stick, yoke, ram horn yoke, whatever happens to be in your hand at the time. How much is enough? 5, 15 or 30° ? Who cares, effects on the aeroplane can be easily read on AH, outside horizon or deducted from turn & bank, if you're going on partial panel and that's what guides pilots hand. Or at least those pilots that can be truly considerd to be proficient in instrument flying.

Considering the matter of untimely activation of alpha prot in normal law, mentioned here and having no relevance on AF447 AFAIK: while I admit I would prefer having its activation clearly anounced, it's not a such big deal. It pitches the aeroplane up mighty quickly but it gets disconnected by pushing the stick forward - which is what you'd be doing anyway if you're unhappy with aeroplane climbing on her own. Sitting, watching the pitch and altimeter go up and comenting " Ehhhh... what eez eet doing now?" is not likely to help.

Retired F4 and his former tactical fighter pilots colleagues have repeatedly mentioned "unloading the wing". Civilian equivalent of this is "PUSH FOR CONTROL" and is sadly neglected in training. Idea behind this is that when faced with loss of control of the unknown origin, most probably it is stall induced, so reduce AoA.

rudderrudderrat

Hi Clandestino,
If you look at page 111 of the report, at time 02:10:15, the computed IAS was about 50 kts, the ISIS was 280 then fell rapidly to about 50 kts. The aircraft was still flying perfectly - and bore no resemblance to your refrigerator.

If I'm airborne and stalled - I want to know about it, especially with UAS.

The 60kt logic is normal with on ground logic to prevent nuisance warnings before take off.
There is no mention in FCOM of the stall warning being deactivated with IAS<60kts when airborne.

GarageYears

Pitot specification?

Has anyone access to the specification of the pitot system fitted to the A330?

Could it be that the difference between static and dynamic pressure at airspeed deltas of <60 knots is 'out of spec'? Somewhere in the past, I believe I read something related to pitot design and the issues of drain holes... or at least less accurate.

The presence of the drain hole effectively reduces the low speed sensitivity, since some of the sensed dynamic pressure is bled via the drain (or something like this). There is compensation for this effect, but at some point that gets iffy.

jcjeant

Hi,

TAM A320 Conghonas related .... no retard .. no spoilers .....
http://www.pprune.org/rumours-news/3...-released.html
Brazil charges 10 over TAM Conghonas fatal overrun

DozyWannabe

@gums, RRR, Ian W:

I don't think the <60kts in this case was intended to be tied in with ground mode - I suspect it was designed to cover another potentially false stall warning trigger. Remember that one of the main reasons pilots have ignored stall warnings in the past was because they were perceived as unreliable. If I was designing the system I'd have to weigh up the probability of a false stall warning triggering due to UAS causing an inappropriate response by the systems or the crew against the probability of the aircraft ever actually getting that slow in flight, which as logical people you have to admit is considerably higher in the case of the former than the latter. The problem we have now is that while the probability of the latter is low, it has in fact happened and now the design and logic path must be looked at from the ground up. As engineers will tell you, knee-jerk changes to design in order to react to a single failure mode are a bad move in general - one must be absolutely sure that the change has no knock-on effect.

@jcj - With all due respect, what does that have to do with this incident? The post you quote I don't think has to do with any logic change deriving from that accident (or the incident that preceded it), which in that case seemed to have been caused by a diligent attempt to slow down as much as possible in bad weather and use an old (more effective) procedure for engaging reversers - leading to the error the new (slightly less effective) procedure was designed to prevent.

Clandestino

1. I should have been more pedantic and write KEAS instead of KIAS, to exclude the case of speed indication loss. It would make it more understandable to those who are learning aerodynamics through Google search.

2. I should have been more pedantic and make clear that my response was referring to stalled phase of AF447 flight (at that time, indication below 60kt was realistic and not result of clogged pitots), as was the Ian W's claim to which I've responded.

3. Since you mention interim 3, it has some other pages, like 29. At 2:10:09 and 2:10:13, computed airspeed is still a mess, yet the stall warning goes off as the aeroplane is jerked into climb!
Where does that leave the theory that unreliable airspeed alone will rob A330 of stall warning?

Let me see.... you pull the stick and are shouted at that you are about to stall for 54 seconds, at least two (and probably three) airspeeds are agreeing and going down, eventually pitch is positive yet the altimeter is unwinding. Somehow you conclude that all those are not clues enough that you're stalled. I am at loss to explain this. Would you, please?

GreatBear

Greetings, old friends. I've been lurking in the daily background of this thread since the amazing recoveries from the sea floor; haven't had much to contribute, but I've been following the thinking nonetheless. Today I ran across this interesting cross-industry (marine) interpretation of Joan Lowy's August 30, 2011 AP article "Automation in the air dulls pilot skills." What happens when the electronics on your supertanker go dead and you realize you left your sextant at home? Perhaps of interest to those who have been developing this line of thought.

Maritime Professional

And for those who missed the Associated Press article, here it is complete:

OaklandTribune.com
Aug 30, 2:22 PM EDT

AP IMPACT: Automation in the air dulls pilot skill

By JOAN LOWY
Associated Press

WASHINGTON (AP) -- Pilots' "automation addiction" has eroded their flying skills to the point that they sometimes don't know how to recover from stalls and other mid-flight problems, say pilots and safety officials. The weakened skills have contributed to hundreds of deaths in airline crashes in the last five years.

Some 51 "loss of control" accidents occurred in which planes stalled in flight or got into unusual positions from which pilots were unable to recover, making it the most common type of airline accident, according to the International Air Transport Association.

"We're seeing a new breed of accident with these state-of-the art planes," said Rory Kay, an airline captain and co-chair of a Federal Aviation Administration advisory committee on pilot training. "We're forgetting how to fly."

Opportunities for airline pilots to maintain their flying proficiency by manually flying planes are increasingly limited, the FAA committee recently warned. Airlines and regulators discourage or even prohibit pilots from turning off the autopilot and flying planes themselves, the committee said.

Fatal airline accidents have decreased dramatically in the U.S. over the past decade. However, The Associated Press interviewed pilots, industry officials and aviation safety experts who expressed concern about the implications of decreased opportunities for manual flight, and reviewed more than a dozen loss-of-control accidents around the world.

Safety experts say they're seeing cases in which pilots who are suddenly confronted with a loss of computerized flight controls don't appear to know how to respond immediately, or they make errors - sometimes fatally so.

A draft FAA study found pilots sometimes "abdicate too much responsibility to automated systems." Because these systems are so integrated in today's planes, one malfunctioning piece of equipment or a single bad computer instruction can suddenly cascade into a series of other failures, unnerving pilots who have been trained to rely on the equipment.

The study examined 46 accidents and major incidents, 734 voluntary reports by pilots and others as well as data from more than 9,000 flights in which a safety official rides in the cockpit to observe pilots in action. It found that in more than 60 percent of accidents, and 30 percent of major incidents, pilots had trouble manually flying the plane or made mistakes with automated flight controls.

A typical mistake was not recognizing that either the autopilot or the auto-throttle - which controls power to the engines - had disconnected. Others failed to take the proper steps to recover from a stall in flight or to monitor and maintain airspeed.

The airline industry is suffering from "automation addiction," Kay said.

In the most recent fatal airline crash in the U.S., in 2009 near Buffalo, N.Y., the co-pilot of a regional airliner programmed incorrect information into the plane's computers, causing it to slow to an unsafe speed. That triggered a stall warning. The startled captain, who hadn't noticed the plane had slowed too much, responded by repeatedly pulling back on the control yoke, overriding two safety systems, when the correct procedure was to push forward.

An investigation later found there were no mechanical or structural problems that would have prevented the plane from flying if the captain had responded correctly. Instead, his actions caused an aerodynamic stall. The plane plummeted to earth, killing all 49 people aboard and one on the ground.

Two weeks after the New York accident, a Turkish Airlines Boeing 737 crashed into a field while trying to land in Amsterdam. Nine people were killed and 120 injured. An investigation found that one of the plane's altimeters, which measures altitude, had fed incorrect information to the plane's computers.

That, in turn, caused the auto-throttle to reduce speed to a dangerously slow level so that the plane lost lift and stalled. Dutch investigators described the flight's three pilots' "automation surprise" when they discovered the plane was about to stall. They hadn't been closely monitoring the airspeed.

Last month, French investigators recommended that all pilots get mandatory training in manual flying and handling a high-altitude stall. The recommendations were in response to the 2009 crash of an Air France jet flying from Brazil to Paris. All 228 people aboard were killed.

An investigation found that airspeed sensors fed bad information to the Airbus A330's computers. That caused the autopilot to disengage suddenly and a stall warning to activate.

The co-pilot at the controls struggled to save the plane, but because he kept pointing the plane's nose up, he actually caused the stall instead of preventing it, experts said. Despite the bad airspeed information, which lasted for less than a minute, there was nothing to prevent the plane from continuing to fly if the pilot had followed the correct procedure for such circumstances, which is to continue to fly levelly in the same direction at the same speed while trying to determine the nature of the problem, they said.

In such cases, the pilots and the technology are failing together, said former US Airways Capt. Chesley "Sully" Sullenberger, whose precision flying is credited with saving all 155 people aboard an Airbus A320 after it lost power in a collision with Canada geese shortly after takeoff from New York's LaGuardia Airport two years ago.

"If we only look at the pilots - the human factor - then we are ignoring other important factors," he said. "We have to look at how they work together."

The ability of pilots to respond to the unexpected loss or malfunction of automated aircraft systems "is the big issue that we can no longer hide from in aviation," said Bill Voss, president of the Flight Safety Foundation in Alexandria, Va. "We've been very slow to recognize the consequence of it and deal with it."

The foundation, which is industry supported, promotes aviation safety around the world.

Airlines are also seeing smaller incidents in which pilots waste precious time repeatedly trying to restart the autopilot or fix other automated systems when what they should be doing is "grasping the controls and flying the airplane," said Bob Coffman, another member of the FAA pilot training committee and an airline captain.

Paul Railsback, operations director at the Air Transport Association, which represents airlines, said, "We think the best way to handle this is through the policies and training of the airlines to ensure they stipulate that the pilots devote a fair amount of time to manually flying. We want to encourage pilots to do that and not rely 100 percent on the automation. I think many airlines are moving in that direction."

In May, the FAA proposed requiring airlines to train pilots on how to recover from a stall, as well as expose them to more realistic problem scenarios.

But other new regulations are going in the opposite direction. Today, pilots are required to use their autopilot when flying at altitudes above 24,000 feet, which is where airliners spend much of their time cruising. The required minimum vertical safety buffer between planes has been reduced from 2,000 feet to 1,000 feet. That means more planes flying closer together, necessitating the kind of precision flying more reliably produced by automation than human beings.

The same situation is increasingly common closer to the ground.

The FAA is moving from an air traffic control system based on radar technology to more precise GPS navigation. Instead of time-consuming, fuel-burning stair-step descents, planes will be able to glide in more steeply for landings with their engines idling. Aircraft will be able to land and take off closer together and more frequently, even in poor weather, because pilots will know the precise location of other aircraft and obstacles on the ground. Fewer planes will be diverted.

But the new landing procedures require pilots to cede even more control to automation.

"Those procedures have to be flown with the autopilot on," Voss said. "You can't afford a sneeze on those procedures."

Even when not using the new procedures, airlines direct their pilots to switch on the autopilot about a minute and a half after takeoff when the plane reaches about 1,000 feet, Coffman said. The autopilot generally doesn't come off until about a minute and a half before landing, he said.

Pilots still control the plane's flight path. But they are programming computers rather than flying with their hands.

Opportunities to fly manually are especially limited at commuter airlines, where pilots may fly with the autopilot off for about 80 seconds out of a typical two-hour flight, Coffman said.

But it is the less experienced first officers starting out at smaller carriers who most need manual flying experience. And, airline training programs are focused on training pilots to fly with the automation, rather than without it. Senior pilots, even if their manual flying skills are rusty, can at least draw on experience flying older generations of less automated planes.

Adding to concerns about an overreliance on automation is an expected pilot shortage in the U.S. and many other countries. U.S. airlines used to be able to draw on a pool of former military pilots with extensive manual flying experience. But more pilots now choose to stay in the armed forces, and corporate aviation competes for pilots with airlines, where salaries have dropped.

Changing training programs to include more manual flying won't be enough because pilots spend only a few days a year in training, Voss said. Airlines will have to rethink their operations fundamentally if they're going to give pilots realistic opportunities to keep their flying skills honed, he said.

rudderrudderrat

@Clandestino,
Page 76.
"Until the end of the flight, the angle of attack values became successively valid and invalid. Each time that at least one value became valid, the stall warning triggered, and each time that the angles of attack were invalid, the warning stopped. Several nose-up inputs caused a decrease in the pitch attitude and in the angle of attack whose values then became valid, so that a strong nose-down input led to the reactivation of the stall warning. It appears that the pilots then reacted by a nose-up input, whose consequences were an increase in the angle of attack, a decrease in measured speeds and, consequently, the cessation of the stall warning."

So you think stopping the stall warning when airborne and IAS<60kts clarified their situation?

DozyWannabe

You're poking at cross tortoises* with that statement and I think you know it. The fact is that you are both talking about very different phases of the accident sequence. You're referring to a later event, where the <60kts stall warning inhibition may have worked counterintuitively, and that's a fair appraisal and reasonable point. Clandestino is referring to the point just after the apogee where the stall warning is functioning and the partial IAS, altimeter and attitude indicator were all giving readings that when combined should logically indicate to an airman that their aircraft is either approaching stall or indeed stalled, and that this went on for nearly a minute without apparent corrective action by the crew before the <60kts inhibition even came into it - as such, he's also made a fair appraisal and has a good point.

* - Best Spoonerism ever IMO...