Hi,

This is a Airbus issued procedure.
Was this one used by AF before AF447 problem ?

Dozy, it's more of a cost benefit issue. For newer aircraft right off the production line fancier CVRs might come automatically. But retrofitting them would cost way more than the benefit when you consider the probability of the accident happening again given the record of the plane's reliability. Retrofitting for a fancy experiment in sound is "right out."

Hi,
What am I supposed to say about that?
Better go in court for asking them to publish their manuals, drills and control procedures. But I'm quite sure that the lawyers won't refrain from doing it. Hence, you'll know about their findings and, not before, you will then be able to make educated comments because they would be based on something.
But until then, what would be the point for speculating about what happened in circumstances we are still not even aware of: like how the deck was manned by whom, what they did once in trouble, what procedure was applied, was it sound or not, etc..

John Tulla said:

My memory of the F-16 development (not hands on) is rather hazy as well. But, I seem to remember that the change to the initial fixed-stick force-control only was to incorporate a 1/16 inch backlash movement and include a breakout force measured in mere ounces...just enough feedback to let the pilot know that he had made a control input. Any more would have caused major problems in formation flying and sight corrections during weapons delivery.

I can't remember the stickforce per g for the F-16 (but I am sure someone will jump in). However, the force used in typical fighters at the time such as the F-18, was one lbf per additional g (with also one inch aft stick movement for the F-18) in pitch. The F-18 break-out force was very low.

With regards the Airbus series with sidestick control I seem to remember a 3 lbf breakout force being required to cause the AP to disengage if the instinctive disengage button was not used. The breakout force in manual flight is tiny but noticeable without being objectionable.

Again with regards the Airbus series, in normal flight, the fore and aft movement of the stick in pitch is a g command system (not pitch angle or rate)...ie, back stick commanded an increase in g above 1.0g and vice versa for a forward stick command.

If impending overspeed was sensed, the flight controls would automatically pitch the aircraft up at up the +1g additional (to reduce pitch angle and speed) but stayed in the g command mode.

On the other hand, if impending stall was sensed, the flight controls would go into the alpha mode where fore and aft stick controlled AOA (not g) until such time that the actual AOA reduced below a set value.

These control modes described rely on air data from the ADRs. Any wrong data from the ADRs is quickly detected but, in the meantime the erroneous data can cause the flight controls to revert to degraded flight modes and trip FCS computers off line.

I think that one of the most insidious effects of the degraded modes is when the THS "freezes". In this case the THS MUST be controlled manually using the pitch trim wheel.

I have seen many students completely caught off guard in the simulator when this happens as they try to control the aircraft in pitch using sidestick alone. Of course the THS completely overpowers the sidestick which is only controlling the elevators in this case.

There is an appropriate warning, but in the excitement this seems to be easily overlooked. Manually reset the THS to zero trim and all is OK.

Simulator training is essential to instill the importance of manual THS control and should be conducted regularly during "Jet Upset" sessions.

Didn't work too well in QF72's case, where one dud ADIRU brought the whole show to it's knees and completely out of the control of the pilots for some time...

Have AF made any change to their procedure for crew rostering/rest since the accident (ie who should be in LHS at ITCZ ) ?

gums, I am starting to wonder if the leaks are from BEA or people reading this stuff and thinking, "Oh, that sounds like it will sell newspapers!"

Dunno, that DS article seems plausible (why steer into Cb - didn't, one puzzle down, no turbulence -CC not seated, second puzzle gone, Capt could be on rest, confirmed). The Capt return to FD could also be the basis of the insinuation about who was on FD.

Stick force per gee
 

Salute!

Sorry, Flex, but the old mil-specs required approx 4 pounds per gee. The Viper approximates that very well due to the computers. Older jets had springs, dashpots, bobweights and bellows to "stiffen" the stick.

Actual control surface movement for most of the older fighters were directly related to control stick displacement as you have stated. The "artificial" feel techniques varied by aircraft, but the gee doofer seemed the best, as a bobweight exerted more opposing force the more gees you pulled. Planes like the F-4 had audio beepers to let you know about AoA, but I don't think gee - ask 'bird.

Mr Optimistic;

Too plausible perhaps? Who would look at a pitch attitude trace, and not look at sidestick positions?

HN39, hmmm, true, but I was thinking it was based only on CVR.

I think you are reading too much into the press article.

It didn't say there was no bad weather it said they tried to plot a smooth path through. (they failed)

So we did have turbulence. At the moment the puzzle still indicates an aircraft flying where it probably shouldn't have (still need to know why).

The rest is unfortunately the normal consequences to be expected when you hit the type of weather reported in that area. You know, the weather the others avoided. There is no need to make this unnecessarily complicated just because the truth is uncomfortable.

And please do not take that as blaming the crew. It is just a statement that we need to know why they didn't properly avoid that storm. There may reasons other than wrong call.

Pitot Static
 

Thank you, Takata for the diagrams and pages of info.

The info as presented tries to pack too much into a page, mixing pitot and static pressures together. AFAIK, all the prior events were airspeed errors alone, and did not have altitude errors. Airspeed and altitude are separate and unique functions within the ADR, except at low speeds, and their output data bus words are separate and unique. An ADR may flag or put out erroneous airspeed without affecting altitude output.

The FCOM Bulletin #009 page 3 gives the case for total pitot obstruction, not just drain hole obstruction, although it is known that Thales pitot were prone to drain hole obstruction due to roughness in the holes. Drain hole obstruction without impact air obstruction will give an erroneous higher than correct airspeed.

Total pitot obstruction, per page 3, puts the plane into an unstable situation. Increase of altitude cause increase of IAS. The A/P will try to increase pitch to reduce the IAS, causing even higher erroneous IAS. This is in accordance with Der Spiegel story.

where do you people get your imagination from?

Try engaging an autopilot with those fault messages. Oh ye. it won't.

And speed and altitude functions are not unique at all. Yes if the pitot probes only are affected then you will only suffer speed problems however if your static ports have a problem you will witness both altitude and speed issues.

All BEA are trying to do now is ensure a professional release of the information because it has been established that the a/c was technically fine. Now that doesn't leave many options.

Until we know why the crew didn't avoid that weather front we won't know why this happened. We already know what happened.

Safety Concerns: Do we know how long the pitots were iced and in what manner before the A/P disconnected?

Pitot total
 

I've followed this complete tread and have nothing to add really but now I was wondering the following: I've had many discussions about the result of a blocked pitot and all people I've spoken to think the same so I gave up!
But I'll give it one more go!

Although with an open pitot head the effect is minimum but I was thinking that with a blocked pitothead the buildup press inside will slowly "leak" away through the drainhole???

The result off this would be a slowly decreasing speed reading staying at fltlevel and same real ias???

Pls. You proffesional idea about this?

I have been a lurker on this board and have to say that I really appreciate the comments here by professionals. It has helped me realize how much it actually takes to fly such an aircraft. Something one usually does not think about when traveling as a passenger (as am I).

jcjeant:
Having said the above, I am unsure where your comments are coming from. In the interest of full disclosure, so I can make an informed opinion: are you a pilot, passenger, lawyer, associated with one of the family on AF447? Or just somebody with some hatred? What exactly is your expertise?

Thank you

JD-EE
 

I concur, see #1820.

takata & Graybeard,

Lets assume for a minute (I say assume), the scenario that Graybeard has proposed is correct, e.g., the AP is attempting to correct for an erroneous reading of higher speed by pitch up. Would, at the same time, the AT functions begin to roll back thrust (N1)? I assume the ECU or FADEC would be receiving the same erroneous speed data and react accordingly. Once out of AP & AT and without correct knowledge of airspeed, one would have to reset both pitch and engine speed manually using the tables for a 35K altitude, wherever those table might be in the cockpit at the time. So then, I wonder how long one has, without correction, to reach a stall point from which recovery becomes difficult or problematic.

Just some curious questions. takata, thanks for all the charts and data you posted.

Safety Concerns

"...All BEA are trying to do now is ensure a professional release of the information because it has been established that the a/c was technically fine. Now that doesn't leave many options.

Until we know why the crew didn't avoid that weather front we won't know why this happened. We already know what happened...."

Not sure how you in fact know what "All BEA are trying to do"

How do you know they were in weather?

You already know what happened?

I think you are basing conclusions on the Leaks.

At this point, "technically fine" is in play. The "Leaks" almost certainly derived from BEA data, though the "provider" identity is in doubt.

If BEA spill the beans this quickly, it will be surprising. If they do, they will have made a rapid fire finding of fact(s). Indicative of an unusually clear chain of failure, the data must have been remarkably straightforward

I am still intrigued by the method by which the a/c stalled. Was it AoA creep, was it inattention, the fact remains it occurred prior to, or consequent to a/p, a/thr loss. How steep was the incline Captain DuBois had to climb towards the Flight Deck?

If his voice is on the CVR, shouting, then one suspects loud ambient noise, which he felt he had to overcome, or did he take exception to the crew's lack of response?

All in all, even to this day, nothing new, whether airframe or crew related. I've been yelled at, but I deserved it.

look there is a lot at stake here (financially). Nobody wants the blame for this accident.

Airbus have nothing to recommend to its operators following readouts of the CVR and FDR.

It is well known and documented that there was some pretty awful weather that night and ALL other aircraft around that time went around it. That is a pretty powerful comment.

You have 3 systems on this aircraft and we are aware of the warnings that occurred up till a certain point. None of this supports AOA creep or one system speed issues. If a system creeps enough to endanger the aircraft there must have been something seriously wrong with the onboard data comparators.

This was very quick as if someone flew directly into a storm front.

Hi Graybeard,
Whilch prior events? Those before Dec. 1999?
Altitude errors are also recorded with pitot issues: variations are not in the same range but still, erroneous altitude will be displayed.

Each ADIRU module is separated in two: ADR+IR. If you are turning OFF the faulty ADR part because of unreliable airspeed, you will also reject all the associated static and AoA probes. Hence, if all three are rejected, you are only left with your standby instruments.

What is your source for this so-called defective "Thales" probes? Thales C16195AA (know as SEXTANT Av.) were first issued in order to replace the previously "known" clearly defective Rosemount 0851GR (now Goodrich) which was mounted on the first batches of A330. Goodrich developped the 0851HL for fixing the same issue and Thales the C16195BA for improving the C16195AA due to water ingestion (not ice). At the end, this Goodrich 0851HL is now considered to be the more robust to severe icing conditions.

Except the fact that those airspeed discrepancies were obviously correctly monitored by the system, and that would be not very likely otherwise. In your scenario, no ADR DISAGREE would be triggered - the system being "fooled" by undetected probes issues. In AF 447 case, it appears that the system was not "fooled" and what happened was 3 ADRs finally rejected for displaying unreliable airspeed. Get it?

Yes, CB13, impact pressure at the Air Data Module will drop to static level, zero airspeed, if the probe is blocked while the drain is open.

The drain hole is calibrated size with respect to impact hole. If the drain hole clogs, impact pressure at the ADM will increase.

Hi,

I'm nothing particular you describe above.
I'm just a person with common sense .. which followed several trials with the means at its disposal and which thus draws conclusions and evidence.
They may be less visible for some.
Follow the remaining events and I am sure everything will be conducted in the same vein of what happened before in several trials
This is so far my "expertise"

PR team working for "professional release"
 

Originally posted by Safety Concerns

And BEA since last weekend as showed by assertive content of "LF leak" delivered very soon

Your feeling IMO is based in a complex "Pattern recognition"

Trimming PR work

Der Spiegel today
 

Pilot war in kritischer Flugphase nicht im Cockpit

"What happened on board the Air France plane that crashed into the Atlantic off Brazil? According to SPIEGEL information to evaluate the flight data recorder is now providing new insights: Marc Dubois pilot was obviously not in the cockpit when the accident took its course".

First, the Der Spiegel article still fits into the category of rumor at this point, so this is comment assuming the rumor has a basis in truth.

This is beginning to look like the X-31 scenario. Once actual airspeed is far enough from system accepted airspeed, the control system gains become set inappropriately and the control system becomes unstable.

The Captain might have had a clue from deck angle when the carts started to roll, or he may just have heard the "Cavalry Charge" through the cockpit door. And I suppose that if I was in the seat, trying to sort out this scenario and my "boss" came in on the middle of it, you can bet I'd mentally pass the situation to him to sort out and then do what he said, whether or not he really understood the situation yet. In this case, the arrival of the Captain in the middle might have worked against this crew.

The swept wing pitch up characteristic works against recovery if a deep stall is achieved from the outset.

My question of a few days ago regarding whether activation of manual THS trim (as during a stall) while in Alternate law, would subsequently be overridden by the system was not fully answered by TAKATA.
Unless manual trim mode involves physically moving the THS trim wheel in some ?Oubtoard? position to achieve manual trim, the more likely method of operation is a force override detection switch to let the computer know that it has been overriden. But then the question is "How does the system know that the crew is done making manual changes?"
If it is a force override detection system, and the system no longer detects a manual override, wouldn't the system return to putting the THS where it "thought" it belonged? Then the next question would be, "Where would the system want to move the THS trim?"

Once in a deep stall, an A330 would likely behave in a manner similar to the F-16 deep stall. The deeper into the stall you get, the more wings behave like flat plates, and less like finely designed airfoils.

Many have assumed that the aircraft would begin to autorotate after a stall, but this general assumption may not prove correct with a FBW aircraft provided sufficient control authority exists to prevent it.

The final question is, whether or not a variation of the F-16 deep stall recovery would work on the Airbus?

In flat plate, the idea is to "spill" flow off one 'side' of the a/c so a 'rolling' moment is instigated. If one can start slow enough, and cleverly instill a "rock" to the a/c, well, if I remember my Boyd, back to flying on bogie's six.

The F-16 cheated, it had rock and roll elevators.

gums ?

Even CRM "inside the loop"
 

Highly complex, volatile and extremely dangerous scenario

Good point. If true, would mean a breakdown of CRM. OTOH, how could the capt. enter through the secured cockpit door unless the situation took minutes to unfold. On the whole, I'll pass on this rumor/leak. Thank you.

How hard would it be to have a crude reaction control system for spin emergencies? Just enough to cancel the spinning tendency and get the nose down.

In Test flight, a spin chute can be fitted. That is absolutely not necessary in a transport category Aircraft. All a/c can stall, and all systems are subject to failure.

447 was lost due influences absolutely designed for. Whether a Training syllabus blindspot, or a glitch in elec/mech, it happened, and BEA know how.


"Why" is different, notably, from how. It will take a judge to determine (jury?).

Why involves the human factor, and as such, we consequently live in a fail/possible realm...

And even then, not all will be satisfied.

Please explain?
If I understand you right, your question is in the same category as "Why don't all the passengers have parachutes?".

Academically, just for the sake of discussion, let's assume the A330 has a deep stall mode in ALT 2 with a CG near the aft limit. (This has not been determined to my knowledge.)

With unreliable inputs in what would probably be a fairly "locked" stall situation, would not ALT 2 logic position the elevator to cope with what it thought was occurring? In a deep stall with not much aerodynamic change taking place could not the elevator effectively be "frozen" in one position with fore or aft SS inputs possibly having little or no effect?

The A330 does not have a pitch override switch (ala the F-16) to allow the pilot to gain full movement of the elevator and use the SS to "rock" the aircraft.

Manual THS input in the nose down direction might result in additional nose up elevator to preserve the same dynamics. (BTW the THS manual trim wheel will remain where it is positioned by the pilot in ALT 2)

Could not flap extension provide a required pitching moment?

(As 'Hoot' Gibson said, "Gear down.")

If this has been discussed before, my apologies.

edit: by the pilot

Very funny :) No, I mean something like hydrogen peroxide thrusters on the outer reaches of the wing. Just enough to get the airplane out of the stalled configuration, hopefully with the nose down.

John the Mod;
In manual flight, the A330 sidestick has a breakout force. It is enough to prevent small motions. However, the experience I've had in heavy turbulence is that, even when the elbow and lower arm are planted firmly on the armrest and the sidestick is moved only through the wrist or more likely through just the fingers, it is difficult to not "stir the pot" or more importantly to achieve consistent, steady in puts in one direction, (subtle or large inputs).

The inertia of the arm/hand responds as one would expect in heavy turbulence, and if the stick is gripped firmly instead of being ridden loosely, (while trying to achieve steady inputs in one general direction), the stick inputs will follow the movements of the hand/arm.

In the meantime, the flight controls, even in ALT2, will all be trying to satisfy all movements of the stick (within the limits of ALT2) while trying to satisfy the commands of the flight control laws in keeping the aircraft at its last-commanded attitude. gums mentioned this phenomenon earlier but it didn't get much play. In FBW, the flight controls are not in a steady, fixed state...they are always moving, attempting to keep the aircraft in its last commanded attitude. In manual flight, these inputs are blended with the commands from the sidesticks, to produce what, in flight data work, looks like a mashup instead of steady state control.

Stabilators and flaperons and rudder
 

Salute!

Bear is right about the Viper stabilators - they moved independently and were used for both pitch and roll. Funny, and most folks don't realize this, but the flaperons and stabilators were interchangeable WRT left-right. made maintenance easy and reduced procurement costs, ya think?

The flaperons were the ailerons that also moved independently but could be cranked down as "flaps". In that case, a roll command would raise one while the other remained down. Of course, the tail surfaces helped and we had a good aileron-rudder-interconnect until WOW.

In our deep stall, both stabilators were full leading edge up as they tried to reduce our AoA. Unfortunately, as as you can see from that graph I posted, our pitch moment at 50 - 60 degrees AoA was zippo for nose down. Above 29 degrees AoA, our rudder was taken away from us - GASP!!! It countered any yaw we had, according to the designers, and guess what? It worked as advertised. That's why first deep stall was hard to recognize by the test pilot. As he recalled, smooth as silk except altimeter was unwinding at max display value!!! Other than that, and no pitch control, a Sunday drive.

While we wait, I would point out that an aspect of the 'bus control laws in one of the very deepest reversions seems to place more emphasis upon speed than upon AoA. This puzzles me. Maybe TK can explain.

From the time I first flew a Luscombe ( Earth was still cooling and I fought raptors during pre-flight), I was taught and learned that AoA is what causes a stall, not speed. I fully understand a design consideration for mach, as a high mach can reduce or even reverse aileron and spoiler inputs. But pure CAS seems a poor input for major attitude commands by "otto".Considering that previous 'bus incidents had erroneous airspeed as a major contributor to the ends results, why would Airbus not consider AoA the primary consideration in an "upset"? I wonder....

p.S. Maybe TK can find the pitch moment graph for the 'bus like the one I posted for the Viper. We ain't proud, heh heh, and our discovery led to better FBW control schemes down the road.

For the techies on this thread, I've managed to dig up this report from my old Software Engineering/Reliability professor, Peter Mellor. In it, he details the visit he made to Airbus Industrie in January 1993:

Peter Mellor's visit to Airbus.

Interesting snippets include (emphasis mine):


Showing, as syseng68k pointed out, that while later generation processors were used in later models, they were still a few generations earlier than the state-of-the-art (which in 1993 would have been the Intel i486, with the first Pentium P5 coming out that year).

It should be pointed out that judging by his posts on RISKS going back to the late '80s, Mellor was definitely willing to be sceptical about the use of computers in aircraft, but it would appear that the more he learned about how it was done, the more comfortable he became with the concept.

Airbus FBW Sidestick
 

Quote from Garage Years:
"my understanding of the side-stick is that the Airbus stick is quite a different design to that of the F-16. The F-16 stick only moves a very small physical deflection and is more of a force sensor (I worked F-16 simulators about 17 years ago!), while the 'bus stick is a position sensor. That in itself though is merely interesting."

Quote from Smilin Ed:
My instructor, Nello Infanti, asked which I preferred. I preferred minimal stick motion with aircraft response determined by stick forces. Nello informed me that the vast majority of pilots also preferred that configuration. I would think that, when finding it necessary to fly current FBW aircraft using the stick, that precise control would be more difficult using stick deflection rather than stick forces. Is Garage Years correct in his characterization of the AB side stick?

You, gums, Machinbird, Flexible Response, and others may find some thoughts of mine on the use of the Airbus sidestick (from experience on the A320) of some interest:

http://www.pprune.org/tech-log/31609...ml#post3979423

From a layman's point of view I was surprised that there appears to be no 'split cuff' or similar to hold the forearm in place. Isn't there anything to provide a positive anchor for the wrist -must be a reason why not but curious.