They didn't. The "philosophy" and control laws they used were already tried and tested in military a/c (and the odd NASA vehicle I believe...). The behaviour you are referring to is apparent neutral speed stablity, and is (I think) a fairly fundamental characteristic of the c* ("c-star") control law. Airbus just applied existing FBW control knowledge to a civilian a/c.

Boeing, on the other hand, for the 777, invented a completely new control law, c*U, to give the effect/illusion of speed stability. As far as I know this was completely new control law/philosophy, never tried on any previous a/c, and arguably therefore much higher risk.

As to which turns out "better", I guess maybe we'll see as FBW spreads beyond A&B - already in Embraer I think, and filtering down to the bizjets. Have they chosen A-style or B-style ? [I don't know, haven't looked].

[There might be another complication - I vaguely recall Boeing having patented the whole c*U concept, in which case they might be able to keep it Boeing-only even if it is "better". Love the patent system...:(]

Well, there have been more than one deadly upset events with one system and up to today none with the other.

Tells a story, at least to me.

Take a look at this article from Flight Global about the Legacy 500 Bizjet: Boeing inspects 787s after aft fuselage composite delamination - FlightBlogger - Aviation News, Commentary and Analysis
(Ignore the statement about Boeing 787-it really is about Embraer.)

Linked sidesticks, C* law but with softer protections in some areas, an AOA limiter, back driven autothrottles, and a different way of doing some things. Looks to be a very pragmatic approach.

Iceman, I apologize for stepping on the toes of a real Airbus Pilot , however you must admit that in the light of subsequent discussion, your statements were not quite accurate.

I am a retired USN F-4 carrier pilot with ~5000 hrs of civil and military time-virtually all of that involved actual manipulation of the flight controls. I've been the Safety Officer in two squadrons, Maintenance officer in one. I've seen first hand a number of accidents, investigated a number of them, and read thousands of accident and incident reports. I've been studying your Airbus flight control system for the last 2 1/2 years and I think I understand many of the Engineering concepts behind it.

I came to the conclusion 2 years ago that AF447 stalled at altitude and descended in a stall. When I realized how little actual hands-on flying time you folks are getting presently, I knew it must have consequences.

I did not go with the airlines because I realized that in order to make a small fortune in aviation, you had to start with a large fortune.

Machinbird

Apology accepted from one naval aviator to another!

I feel the discussion has got way too "technical" and away from what an ordinary "line pilot" would or needs to understand. I too had 20 years of no A/P nor Flight Director where you had to "fly" the aircraft.
What we have here though is what appears to be a basic airmanship / piloting error, after a relatively minor emergency, as unpalatable as that may be. Blaming the aircraft or airbus for this failing is not good enough. Basic pilot skills of attitude and power would and should have been enough to solve the problem. There also appears to have been a failure in monitoring by the PM and by not taking control and commanding the situation.
The auto-trim did not cause the stall it was trimming the aircraft to assist the pilot, until it cut out. The upset training we carry out always puts emphasis on the fact that the aircraft may be out of trim and manual input may be required to assist in the recovery.
The effort required on the side-stick to achieve the pitch attitude and "zoom" climb was something that should never be applied to any "airliner" at FL350!
My very unscientific "experiment" achieved 13,000/min rod from 'stalling" at 38,000'. But with 15 degrees nose down, some trim, then thrust aircraft recovered by 29,000'.

Looks like the PPrune Administrators have fixed the problem by moving the thread into tech section. It is a natural progression for long term threads it seems.

Iceman, the problem as I see it is that the majority of new guys hitting the airlines do not have the same background and experience that we have acquired, nor will they be getting it in the typical airline flying environment. It is FUBAR, but that seems to be the way things are going. The AF447, and Colgan accidents are representative of this newly developing trend. Where it ends, who knows? In the meantime, the airframe manufacturers had better be looking for ways to protect this new type of pilot better. Whenever I see an area that needs improvement, I try to point it out.

I assume you would be perfectly comfortable, given a sufficiently maneuverable aircraft, to perform aerobatics by instruments at night by with no visible horizon. I get the feeling that the majority of the new airline input would look at you like you were crazy if you asked them to do that.

Different times with new economic factors at work since you and I entered aviation.

Machinbird,

Thanks for the link on the Embraer Legacy 500/450 FBW.
The audio briefing is very interesting.

You mention "Linked sidesticks" but that's not what it is.
Sidesticks are still independants (Airbus way)
Audio describes it between minutes 13 and 21.

As you mention, on the autothrottle side, Embraer did not follow Airbus, and even mentions 2 airbus accidents to justify their choice.

As I understand it also, autotrim under manual flying.

Thank you for the correction CONF iture. I see I must have misinterpreted the following statement's meaning:O:

What story does the string of 737 LOCs tell ?

What types are you comparing with what and with what normalisation (flight hours, departures, none(!)) ?

330 vs 777 is about equal on hull losses and hull loss rate, and both have had non-fatal in-flight LOC due to ADIRU failure. 777 has been a lot luckier on fatalities (especially BA38 - engines choked a few seconds earlier and that would have been a lot nastier however good the crew).

340 ? Luckier still on fatalities but more hull losses. Comparable to 777 - not sure appropraite - should twins be compared to quads, should it compare to 747 ?

320/21/18/19 - much different type and usage to 777, compare to 737 against which competes, and hull loss rate is about equal - provided you exclude the older 737s.


The big story is comparing 320 onwards against the old conventional controlled A300/310, and older 737s etc. - the type represents a massive improvement in safety, close to an order of magnitude lower hull loss rates.

If there's a major safety problem with Airbus FBW, it just doesn't show up in the stats - if there is a signal it's hidden by other much larger factors.

inrequentflyer789:
The story of an previously unknown abnormal state of the rudder actuator under specific temperature conditions? Is this comparable?

-Member of The Gretchenfrage Fan Club

Thanks for that link - very informative. Does indeed look like somewhere between A & B approach, proper moving throttles, but bus-style sidesticks. Ugly chimera or best of both - time will tell. Someone in marketing probably needs firing for naming it "Legacy" mind you...

This has me a bit confused:I think that must refer to failure / direct mode, as the videos look to me to show it controlled like a path-stable airbus (not like 777).

VERY different events IF789 :

• For the 777 the AP then the pilot followed unrealistic FD commands.
• For the 330 there was nothing at the time the pilot could have done to prevent the protections to mess up.

upset
 

Hi everybody,
i agree with most of you, is the lackness of training and too much complacency on the automatism that create a false sense of security among the pilots.
I currently fly one of these A 330 and i see people been slaves of the machine.
Even simple tasks like disengage the ATHR is considered a strange thing.
My background is a solid stick and rudder flying but with time everybody become lazy and don't want to give away the luxory of the automatism.
Yes if you don't touch anything probably the plane still fly by itself but is not easy to not become emotional and nervous in a situation like that

I was thinking of the 737 LOCs that were actually "lost control", or failure to take control (after automatics fail) of otherwise flyable a/c, not "controls failed".

Say: VP-BKO, TC-JGE, ET-ANB, PK-KKW, OB-1809P, SU-ZCF, VT-EGD

I think I've eliminated cfit and mechanical failure as direct cause from that list.


But the point is not the list, the point is that pilots lose control of 737s as well as A320s, in neither case does it necessarily mean the plane has a bad control system.
Pilots have been losing control of planes for as long as there have been planes, and probably will carry on doing so for as long as there are pilots. And then autopilots will lose control instead (no, I'm not a fan of pilotless either).

Comparing accident rates type to type, competing or old to new, simply does not show fbw as making things worse.

I don't think it shows it making it any better either (newer a/c tend to have better record, fbw or not) - despite all the protections. I don't think that shows anything more than that as a species we're really good at finding new ways to screw up when old ones are closed off...

Before I make my unsolicited comments please keep in mind that the only aircraft I've ever "flown", or piloted, are the type which one uses a radio transmitter with i.e., I know nothing about that which goes on in the cockpit.

After reading this thread since day 1 I've come to the conclusion that many of the comments here, taken together, are a damning indictment of the cockpit crew of this ill-fated flight. After poring over the data, reading the transcripts as provided by BEA, and reading all of the pertinent BEA-provided reports I suppose some of this damning indictment is appropriate.

But I'd like to take this one step further. Isn't the apparent lack of training, as far as the PF is concerned, the lack of CRM, the lack of communication between the flight crew members a further indictment of our modern world? Is not this one specific flight crew a microcosm of our greater world? Is not the underlying reason for the PF's apparent lack of real training yet further proof that "the bottom line" is really the precursor for all that comes after?

I'm involved in the I.T. field and have been for well over 25 years.
I fondly remember "the old days" when the latest innovation in technology was a "big thing". Nowadays I am astounded, shocked and dismayed at that which is coming out of our schools these days. It seems to me the "younger generation" really doesn't have a clue nor do they seem to want one. Apathy, lack of motivation to learn, inattentiveness, self-centeredness etc. These dubious traits seem ingrained in so many these days. Is not what happened with AF447, unfortunately, not another "small" example of all of the above?

rgbrock1
 

"Nowadays I am astounded, shocked and dismayed at that which is coming out of our schools these days. It seems to me the "younger generation" really doesn't have a clue nor do they seem to want one. Apathy, lack of motivation to learn, inattentiveness, self-centeredness"

Brock, I guess you're referring to American and European kids. Say thay to millions of families in Asia and they won't understand what you're talking about. Have no doubt, there's a transfer of power going on from west to east.

Naval confidence
 

Hi Maching Bird,

I liked very much your Naval confidence with Iceman50 (you are the winner).:rolleyes:


You said "Whenever I see an area that needs improvement, I try to point it out". That is a noble and generous project.:D

It seems Iceman50 wants only to avoid a unfavorable AF447 trial in France : This requires that the FRENCH Court has different possibilities :
1. several persons are charged and not only one (for instance AIRBUS, pilots and AF, certification autority, etc. are possible guilty persons) ,
2. The Court is not sure the charged is guilty,
3. The alone charged person did not knew nore understood the danger of his actions, and wanted to kill the passengers and crew from the AF447.:suspect:

The first way is the easiest with the famous fatal chain, used since years and years by the french lawyers.

The Blame Game
 

rgbrock1, I fully realize that you went on to clarify and amend the above comment, but I see a number of "damning conclusions," at least the ones that I myself make, from the comfort of my armchair. ;)

1. Air France, for not changing the substandard pitot tubes fast enough
2. The regulatory agencies responsible for rule-making in reference to the above.
3. Air France training
4. The entire industry, again for the failure to provide proper upset training
5. The pilots themselves, both their, uh, flying skills, and their poor decision to fly into that storm
6. Deficiencies in the Airbus design in terms of lack of tactile feedback in the controls (all of the defifiencies posited by contributors to this thread).
7. Deficiencies in the Airbus instrumentation and control design in terms of stall warning behavior, angle-of-attack display, auto-trim, and the list is long.....

As someone who has NO industry axe to grind, I strongly believe that ALL of the above were dangerous holes in the Swiss cheese, and if any one of the above factors had been not present, the crash might not have happened. I love posting opinions that cannot be proved wrong! :eek:

Some additions maybe:

1.1 AF for (possibly) failing to evaluate adequacy of training and procedure for a known risk (pitot fail) they chose to run as a result of (1). Allegedly when they did in the sim post-447, their crews crashed.

4.1 regulatory authorities (or entire industry) promoting negative (wrong) upset training - one opinion here: The big stall recovery debate - Learmount. To quote:...and opinions:

6.
Mmm, no fbw should ever be certified then - back to the regulators who approved it. Fbw means only artificial feedback possible, and then can't have proper control feedback in the absence of airspeed info, since proper feedback is dependent on airspeed.

7.
AoA: separate display is designed and implemented by airbus, blame the airlines (inc. AF) who ask for it not to be fitted. And the regulators who have ignored (several) past accident reports recommending it be made mandatory.

SW: same logic might turn out to be common throughout the industry - comments on these threads have alleged that it is there on boeing at least. Just that no one on other types has managed to get measured airspeed below the threshold whilst stalled. Based on info posted here, I think the Airbus BUSS option already "fixes" this issue - unless you get the speed so low the AoA vanes physically stop working. AF of course decided that their crews didn't need no backup speed tape...

Hi,

Click on for be readable

http://i.imgur.com/O9SEzl.jpg

Organfreak,
In what respect were the pitots 'substandard'?

jcjeant,
AF447 did not encounter the icing conditions (SLD) the NTSB recommendation is addressing.

Yet another way of looking at it is;

With so many barriers present (layers of cheese) it would be extremely improbable that all, or enough of them would fail resulting in an accident.

Obviously some of these were not actual barriers but intead were latent failures already in place (holes drilled in a line though multiple layers of cheese) because nobody had bothered to validate some of these barriers as being effective much of the time.

My strong convictions are that we can't expect all layers to work all the time, but that we must demand that controls and evaluations are in place to at least qualify that a layer of defense is active and not a dream on a piece of paper assuming that the other guy has it covered.

That's why we have regulators, to make sure that the process is active

Hi,

And
Maybe you're right about that (SLD) .. however it shows the lack of responsiveness of the European regulatory body
1996 -2010 ... 14 years of inaction ...

jcjeant,
14 years of inaction? You obviously have no idea of the work that needed to be done and has been done.

Hi,

No I have no idea
HazelNuts39 can you explain all this work done during those 14 years (and more) by EASA about this problem as obviously you have this information

Roulishollandais, you may have mis-perceived this interchange.

Basically When Iceman50 stated he was a Naval Aviator, I knew exactly what his training background was before his airline career, and he knew mine.

It means that if you were flying hundreds of miles from your ship in lousy weather and your navigation system packs up followed by your communications and some of your critical systems, you can still fly his wing back to your ship or a divert field after you pass your HEFOE signals, even if he is a brand new Ensign.

He was saying that his Airbus is easy to fly (for him) which I believe fully. It also points at inappropriate fundamental ATPL training standards as an additional causative factor for AF447.

Not to bad mouth a dead man but the first mistake I see in this entire tragedy was when the Captain pushed back from the gate. Mistake two was when he chose to go to the back and go to sleep. IMHO :sad:

Maybe reading NPA 2011-03 will give you some idea.

HN39, thaks for the link.
I haven't read all the 77 pages yet.
But... at first sight it deals with 'low-level" icing and the ATR-72 accidents, not with the problem of the certification problems of pitot tubes in hi-alt icing conditions (which are not the same).
So far, in all the discussions, I've not yet seen a mention of even a 'proposal for rule making' to replace the obsolete certification rules for pitot tubes.

BTW, I'm ignoring jcjeant's waffle and interference - he's obviously got his own troll agenda.

Keep reading.

Ahhhhh.....humans. If I could change one thing, and one thing only, it would be the cessation of the stall warning at whatever values of AOA and airspeed were inconceivable to the engineers that designed the system. Those poor bastards might have been on the verge of figuring it all out when an appropriate control input triggered a warning.

Hi Maching Bird,

Quote:

---

Originally Posted by MachinBird

Quote:

---

Originally Posted by Originally Posted by roulishollandais

I liked very much your “air" combat with Iceman50

---

---

If I unvolontary have hurted your high sense of egality and your humility, I apologize .
To hope some pardon could I edit my post, replacing "air"combat by "Naval egality" or "Naval confidence" ?

At reading the some agressive Iceman50's post, I felt he was throwing you a curious challenge...

Quote:

---

Originally Posted by iceman50

Very condescending!

---

And I felt equity, regularity, in your answer, asset on asset. Probably I was not alone to feel that .

You also showed us your surprise and your happiness to find a brother in arms, and brothers in very accurate, exigent and engaged flying.
I liked that from both of you.

Your description from coming back safely after nav and com systems failure in bad weather far away from your ship is very adequate in this forum about the AF447 LOC in the ocean, and full of rich humanity of our pilot community, Naval and not Naval, whose Home is in the sky...

Best regards, with regret to have been an "insider" in your interchange..

ChristiaanJ
Extract (pages 20-21)
Furthermore, service experience indicates that flight crews have experienced temporary loss of or erroneous
airspeed indications, malfunctioning and/or damage to temperature probes in severe icing conditions (in
areas of deep convection). The main suspected cause is ice crystals in high concentration.
The on-going investigation of an Airbus A330 accident
7
(flight AF447, 01 June 2009, Atlantic Ocean) has
established in the interim report No 2 that several (twenty-four) maintenance messages were transmitted
by the ACARS system and that these messages show an
7
Refer to the Interim Report No 2 dated 17 December 2009 available on the BEA France Website.
Please use the following link: FLIGHT AF 447.
inconsistency in the measured airspeeds. A meteorological analysis shows strong condensation towards
AF447’s flight level probably associated with convection phenomena. The aircraft Pitot probes potentially may
have encountered severe icing conditions including ice crystals and mixed phase. However, as of today the
root cause of the accident has not been established, therefore it is not possible to determine whether or not
the mentioned airspeed measurements inconsistencies have played a role among the causal factors.

infrequent flyer 789

I am as yet unable to locate my quote (sic) from the source that mentions the a/c did not immediately respond to Bonin's initial N/U command. As he took control, you will remember, he input Up elevator and RollLEFT ailerons.

It makes sense from the fbw protection philosophy, as the STALLSTALL just after a/p drop indicates an AoA approaching aerodynamic STALL.

I continue to hold to the uncommanded Pitch Up theory of cause, especially when the pilot's inputs appear to have been additive, should the a/c autoclimb theory prove out. If the aircraft was not responding to PF's initial NU, I consider that to be evidence of the corollary to "Uncommanded", ie: "Not Responding".

Wind` shear can easily account for sufficient immediate excursions in IAS of 100 knots and more, an immediate explanation of loss of autopilot and acquisition of both STALL and OverSpeed (in that order) protections, IMHO.

I do not agree with the "PIO" idea of Roll dynamic. The evidence points to a chronic (controls neutral) roll to the right. An obvious reason would be damage to controls, perhaps a partially unstowed spoiler; these were intially deployed, then retracted, by Bonin.

regards

Aside. A possible explanation of the lack of concern of the PF re: STALLSTALL
would be the appearance (or not) of the STALL activation threshold on the speed
tape. This cue is not visible after ADR is rejected, leading to the possibility that it was visible, showed an obvious reason for the STALL warn, and hence his lack of comment. It would have remained only if the ADR's were still in, showing that the a/c, at least at this point, remained in NORMALLAW. It would support an airspeed corrupted by wind shear instead of ICE, for all that may mean to Thales, in their defense.
edit. pp29 of 3rd Interim report shows a graphic of steady PITCH @1 degree whilst PF is inputting his supposed "mayonnaise" climb.:)

It does not make sense from any philosophy. In any case the sidestick was pulled back before the first stall warning.

Nowhere does it appear to have been additive. The pitch behavior was consistent with sidestick movement.

The aircraft was in alternate law 1 second after the AP disconnected. There is no overspeed protection in alternate law.

The spoilers/speedbrakes were not extended until 2 minutes after AP disconnection.

I cannot make heads or tails of this last paragraph, but suffice to say normal law ended when the AP disconnected.

Loss of Autopilot does not drive an immediate degradation to Alternate Law, does it?

I woud be interested in your patient reply, with explanations. My small child can say "NO" endlessly.

EG: Does the Autopilot not make use of the spoilers?

er, and the Rudder?

Lyman asked:
It certainly did in this case. Without speed data, there can be no autopilot.

When did you notice the loss of SPEED Data? The pilots noticed it by reference eleven seconds AFTER the Cavalry Charge, yes? (Robert: "Loss of speeds, Alternate Law)".

Somebody please correct me if I'm wrong, but the A/P disconnected because of the bad speed data! This hamster wheel is makin' me dizzy!
:yuk:

jcjeant,
You obviously totally missed my point.....
Aircraft and aircraft equipment are designed, certified and manufactured in accordance with airworthiness requirements.
This also applies to pitot probes.

The unfortunate problem in this case is that the certification requirements date back to the stone age...
Both the Thales and Goodrich probes met those ancient requirements, or they would never have been fitted on an aircraft, but the testing conditions no longer match the environment they are used in.

What I was saying is that, so far, I've not yet seen any reports of substantial research into the 'unusual' pitot icing leading to UAS events, or any 'proposals for rule making' based on such research.

Yelling: "design a better mousetrap" is a bit pointless, if you can't state what that 'better mousetrap' is supposed to achieve.....
Ah.... it should be able to trap 'zombie wombats'. Fine, give me a spec describing the size and behaviour of those zombie wombats, and I'll try and design something that might work.....

As an engineer, I design to specs.