What do you mean by that, Bearfoil, in response to gums? (See also Smilin' Ed's comment on power/attitude flying).

I do not understand the A330 cockpit and AFCS degraded modes, but there are some basics that seem to me applicable regardless of aircraft model.

As I have read this thread, I kept getting the idea in the back of my head that, with AirSpeed signal lost due to iced pitot tubes (let's call that a necessary assumption for the moment) one could maintain Straight and level within the air mass with power (roughly what you'd had set just a moment ago) attitude (nose and wing) , and a reference to VSI (basically, a partial panel scan/crosscheck).

Granted, VSI is a static instrument, so if static ports are also iced up my idea is at least partly dead in the water, particularly if you ride the downdraft inside a turbulent airmass all the way down to the ocean ...

But why did you imply that it would not work?

I have understood that were they at stall already, unwittingly, that maintaining power and attitude would not do much to get out of a stall.

EDIT: sorry, that was some weird cut and paste for quoting, have tried to clean it up.

Lonewolf 50

Hello. The Airbus in the conditions imagined for 447 can be a handful. Automatics and FBW are two distinct subjects, suffice to say that Airbus trains reliance on the a/c and systems such that in the commercial cockpit, it morphs into dependence, and not all the Pilots fault, certainly not here. ACARS is not a flight tool, it is a maintenance convenience to save money (and very valuable in that regard). We are reading delayed messages meant for mechanics, not for flight support.

I personally believe that this dependence on auto played a very real part in this tragic accident. Even after AP dropout, the ECAM prompts a/c solutions for the computers, not avoidance of, or recovery from, upset. The manual must be consulted before action? Evidently, since Pitch and Power as a solution appears on the fourth page. Dark, bumpy cockpit, Captain most likely absent or due to relieve, there was no time for sequencing, prioritizing, or effecting control by hand.

The Airbus, (this one) has no Artificial Horizon, and Throttle lever position doesn't communicate truthfully the power status. At the cusp of upset, the computer and its trained for approach to flight does NOT entertain heroic manual efforts, the concept is foreign to the designer and flight support. What it does instead is steadily degrade pursuant to its own timetable and program.

gums is dead on, but his old fashion style (mine also) has no place in this modern cockpit. There is an Artificial Horizon (gyro) available for this model, it is an option and here it was not selected.

I don't accept necessarily that the pitots malfunctioned. In a cell, airspeed can realistically get bizarre; it is in some cases actual airspeed, but not accepted by the flying pilot because it doesn't make sense. Neither is a three way discrepancy unheard of in wide bodies, each ias being actual, subject to turbulence. Why is the windshear alarm not attracting more attention?

Neither is it known what RTLU fail meant. BEA assures us it meant only that the Rudder was captured by limits provided for >272knots.

regds,Bear

Bearfoil

First off, thanks for the reply. :)
Understood.
Roger.
This got my attention. :eek: It is also foreign to my understanding of flying any sort of aircraft.

Could happen to anyone, it's why airline pilots make the big bucks. :ok: Got to fly the bird as best you can.
:eek::eek::eek:

While my gut instinct is to recoil at this concept, my brain tells me that the pilot of the A330 must know his robot, and its habits, inside and out, or the chance of a robot and a human brain working at cross purposes poses some scary ( to me ) problems.
Not selected for purchase, or not selected as an option by the crew from a menu of choices? I think you mean the former, did I guess correctly?
Roger. One reason (among many) to avoid cells wherever one can.
Again, thanks for the insight. This thread has been an eye opener.

Change of sub topic for a moment?

For anyone still intersted in search, and a tool referenced a few pages back, MAD.

I don't see MAD as of any use. For airborne use, slant range too great given presumed depth of target, and mass of target.

For a submersible, order of magnitude too slow, and target mass too low, to get a useful signature.

MAD working versus a submarine (consider how much steel/iron is in even a small one) has a few orders of magnitude more signal to try and achieve, but it is the rate of change in mag field that gives useful indications, which is a function of airpseed (or water speed?) of MAD and the size of the target. (Orientation as well, but let's not get too far into the weeds).

Lots of words to say: MAD, not what will find what you are looking for.

I tried to find an F-16 (the assumption was maybe the engine would register if we flew over it) with MAD off the coast of Turkey, in coastal/shallow water, with a reasonably well established datum. Nothing doing. Not a sniff. Was eventually found by other means.

The Falcon is mostly Resin, Aluminum, Stainless, and Titanium, all non-magnetic.


MAD is expected to not work here?

Is Mass density important, or just ferric structures? As in Magnetic Resonance Imaging, where iron is a nono.

bear

OT The Airbus is a triangle, not a duality. The Triangle is the strongest stucture in engineering; in aviation, it must be FBW, Pilot, And Automatics. All three must be interdependent, and easily understood, one and the other TWO. This tragic accident may be partially explained by hubris, ego, and politics, among other more direct causes.

bear

What MAD is good at picking up is a change in the density/amount of iron/steel material due to its magnetic characteristics. The amount of iron/steel plays a big part in making a big enough mag field change to pick up: signal to noise ratio issue. (I'd need to look up a few bits from long, deep memory, but I do recall there being degaussing ranges our subs used to run through to reduce mag signature ... memory very fuzzy). HY 80 and HY100 steels were what a lot of US subs were built with, but that's very old news. I think they have more exotic alloys in the newer stuff.

Russian Alpha class, all Titanium, was not something MAD would pick up ... did find a Tango once, though. Good day for us.

As to A330, I presume the engines have lots of steel alloys in them, which is what I was hoping for with that F-16. Again, more than an order of magnitude less steel/iron than a standard submarine ...

Just out of curiousity, has A330 got titanium in the wing boxes?

bearfoil, Lonewolf_50,

I thought MAD (in this context...) stood for "Magnetic Anomaly Detection".
I thought it 'could' work for bl**dy great subs with a lot of steel at a few hundred meters depth.

I can't possibly see how it could work for a debris field of largely non-ferrous (non-magnetic) material at 3000+ meters....

But I suppose this already has been mentioned in this thread before.

CJ

Bearfoil,
You seem to be implying things which are aimed at misleading.
In this Version of the 'Bus the Artficial Horizon is provided by the IR Part of the ADIRU, shown on the PFD's plus the Horizon provided by ISIS.

Yes it doesn't have a mechanical one but three electronic ones and none of the ACARS messages suggests they were u/s.
Ans a mechanical one is much more prone to tumbling in turbulence or upset than the Ring Laser gyros.

I still have some difficulties understanding what you are trying to achieve here ?!

I am trying to contribute to this thread and keep current. It is extremely important to keep this accident in public view; as one who is familiar with the mission and method of investigative bodies ennabled by Public Authority, I can provide an insight into how they work. I have been a member, a Director, and an investigator for many different bodies, though all in my native tongue, and Politic.

My memory of ACARS is that both PFDs crapped out, and IRU is dependent on AD, and that at a critical time in the sequence, the pilots had not a clue about Pitch, nor did they about Power, and this at a time when all automatics had left. In leaving, they ignorantly displayed their mischief on the panel, taking up crucial time for the pilots to catch up with real behaviour.of their aircraft. If the auto is inop and unavailable, I frankly do not give a fig about troubleshooting the system ahead of maintaining (or recovering) controlled flight with pilotage relative to a manually controlled aircraft.

That is my background, and my motive, along with trying to defend this flight crew against well financed and powerful story tellers who have taken advantage of dead pilots before. What is your purpose here?

bear

Henra - "In this Version of the 'Bus the Artficial Horizon is provided by the IR Part of the ADIRU, shown on the PFD's plus the Horizon provided by ISIS."

But what happens if the electrics fail ?

What do you expect to happen in a fly-by-wire aircraft if the electrics fail?

Providing the a/c is still in one piece, the Essential Systems Service buss (ESS) will have at its disposal AC and DC power required for flight control, navigation and communication. If primary sources, e.g. main engine generators fail, the APU is available, then the Emergency generator (RAT) and then Battery banks for DC buss and an Inverter for essential AC buss powered equipment. There are alternative AC and DC buss supplies also available.

Electrical supply is not an issue.

mm43

Hi,

How much time (start sequence etc ..) for have the APU connected (providing electric power) to the buss bars ?

Well, I wouldn't expect the ACARS transmissions to continue for a start...

As I have said before, and still believe, the event started with a sudden, unprepared and unexpected upset which promplt escalated in uncontrolled flight agravated by time as the flight continued. Most likely, all acars were a spin off and a result of the initial event, not a contributor to the actual cause. I agree, electric had nothing to do with it.
ww

Christiaan: yes, about MAD not being useful in this case, I was trying to explain why. Size of metal/iron object not big enough, nor speed of search vehicle, nor slant range within parameters. Because of how MAD works, or doesn't, this scenario is not MAD suitable for detecting desired object: engines.

Gentlemen, please forgive an old pilot ... I am unclear on the AH/AI/Attitude Gyro equipment on the A330.

First off, I understand it is part of the usual glass cockpit, not the old ball in a box I first flew with back when dirt was new ... thanks for the laser ring gyro point.

When flying the A330 in standard configuration, does the flying pilot normally have on his display the pitch and bank indications that an attitude gyro/artificial horizon usually provide in aircraft as I was used to? I would assume "yes," and be surprised if the answer is "no."

I was led to believe, though I may have read incorrectly, that there was no AH either selected (or even equipped ??? makes no sense now) in the A330 model being flown by Air France flight 447.

If I have misunderstood, please help me.

I think I understand one estimate being that AH was for X time unreliable, but I do not grasp what in the ACARS pointed to that estimate or conclusion.

Thanks for any enlightenment you can shed. Perhaps the acronyms, and my unfamiliarity with them, hold the answers that are right in front of my face.

That really got my attention. I heard a NASA/JPL robotics-expert lecturer define a robot as a machine that can make autonomous decisions, and the A330 and its kin are certainly that - surely the largest and likely most 'intelligent' robots man has built so far. I wonder if it helps the cockpit to think of it that way.

poorjohn, it helps me, and has been an aid in better understanding what I have learned from some superb PPrune commentary on the Hudson River landing, this lost flight, and other mishap discussions. The explanation of normal law, degraded modes and Alt Law, and the sorts of decisions IF/THEN rules result in -- actions by the software/hardware package (robot ;) ) -- makes the state of the art autopilot a hell of a lot more complex than the old George who'd keep airspeed and altitude for you at cruise conditions.

The good thing about robots, so far, is that you can know what programming parameters are, and limits, as robots built to date tend to be rule-driven machines. But you do have to know in order to get the most out of it, and not expect more out of it than it can deliver ... like most machines.

FWIW, I understand that learning machines with AI that is beyond simple rule structure have been in development for some time, but am not current enough on that matter to comment further.

Bearfoil
If you read back into the old Thread plus BEA's analysis the common understanding seems to be that there is no inidcation the PFD's were u/s.
Also the IR part is not dependent on the AD part.
Only the other way round, if there is no IR, then AD is blanked out on the PFD's as well.
None of the ACARS messages indicated that attitude information was lost.

BTW. thanks for the explanation of your motivation.
I can follow that logic and it's OK for me unless it would be mainly used to spin the public opinion in the wake of a lawsuit.
Keeping the public interest up and keeping the spotlight on this accident should be in the ineterst of all involved in aviation.
But I would be grateful if you could avoid oversimplifications which could imply things which are not meant to be.

Thanks for your kind reply, henra. I have revisited the Preliminary, and take note of pp51 forward to 02:13. Initially the PFD's (display) are denied FD (director), it was selected but unavailable. Next comes the failure of the FPV (bird), and all orientation by PFD is eliminated. This implies to me that all attitude (and airspeed) values are not transmitted to the screen; the discrepant pitots are minimally reacting to a decrease in a/s of thirty knots from the medianed number, so none can be seen. BUSS (back up speed system) is an expensive option, and was not chosen for this aircraft. I am lacking the perception of an available value for attitude, of any kind, airspeed likewise. Add to this the need to know power, either by throttle lever position, or a displayed value by N1 (or epr). What am I missing? Lay it on, I have thick skin.

thanks again

bear

From the point of view of information presentation in real-time, I'd rather have a bunch of round-dials and scan-scan-scan as I was taught....

Should something unforseen happen rather fast, the thought of having to page through assorted displays to see the info that I need is disturbing.

Maybe going half-way back to a mix of the two would be a thought.

:cool:

My initial comment on the loss of electrics lacked clarity and thanks to all for correcting me.

What I’m really getting at is pulling the plug on something that provides all your necessary flight information and the cause might be behind a panel.

Something that didn’t come across in the calm cockpit shown in the recent BBC documentary was just what it’s like to fly in a cocktail shaker conditions AF447 may have been flying through.

The initial AF perhaps knee-jerk statement was – it may have been struck by lightning, yes possible but more important where, it may have taken out some important wiring.
Connectors shouldn’t but can come adrift, wiring can chafe and short, circuit boards can fracture.
I presume that a PFD failure requires a complete circuit to generate a failure message.

Basically with PFD’s all you eggs are in one basket and I’m going to stick with the opinion that this flight crew lost more than just their airspeed indication

The PFD's may have been red flagged (FPV, SPD LIM, FD) but the interim report states: "no message present in the CFR indicates the loss of displays or of inertial information (attitudes)" (page 41). And no loss of attitude (purely inertial, not based on any air data) has been reported during the past cases of Pitot freezing events.
Jeff

Mack,
I do software that is touch screen based for broadcast stations. I note that there are companies making "industrial", as in ruggedized, touch screens as large as 1920x1200 pixels and typically about 20.5" by 12.75" viewing area. A pair of such beasties in front of each seat might make a display you didn't have to page through.

I get the chills every time I think of a poor fellow in a cockpit with the important display of the moment up when something pickles. "Er, let me see, press exit, exit, exit, frimble, wubble, mixup, pickle-solution." By the time he's done that he has a missile up his tailpipe. (I have a background doing DoD work in my bad old days.)

The feeling persists. A basic set of critical gauges and displays must always be present. I have a suspicion that larger display real-estate might make a difference. I am relatively certain that the 'surprise' solution, automatically flipping the display to the important one, is not a good one. The brief disorientation could also be fatal. But, just maybe, it might be less fatal.

I observe that when the brown stuff starts to shred in the impeller more problems than major problem often appear. Picking the one true problem might not work automatically. If not, a large display could mitigate the problem a little.

Am I on to something here?

JD-EE,

Do you read French?
You might enjoy "Concorde, Essais et Batailles" by André Turcat.
He was the Flight Test Director at Aérospatiale at the start of the Concorde programme, and also the first one to fly the aircraft.

There's almost an entire chapter about flight instruments, and presentation formats, etc. that's worth reading.
This was the age of the "clocks", although some were tape instruments, and the three-hand altimeter was becoming a thing of the past.

Reading your post, I was suddenly wondering how much of the present-day glass cockpits are still designed by pilots, and how much by geeks accustomed to flipping through browser pages, clicking on icons, and using 'Google'.

CJ

flying in the center of a such big cell at those latitudes doesn't give you any chance to fly out of it alive...no matter if all the instruments are working or not ..why they flew in there?looking on the radar tracks all the other aircrafts avoided that big cell...

FPV
 

Hyperveloce, could you help with this:

You quote Interim Report No. 2 p. 41. stating that "No message present in the CFR indicates the loss of displays or of inertial information (attitudes)."

Same report on p.36 states: "This message indicates that the flight path vector (FPV) function is selected but unavailable."

From reading other forums, it appears that FPV is useful in unreliable airspeed situations.

Just trying to understand why this is not an inconsistency---nothing else.

Bird
 

Well, it also appears that the FPV (elaborated from the inertial/barometric parameters) is not available in a UAS (and for the AF447 crew), when the 3 ADR are disagreeing ? the basic procedure is to set the pich (not the FPV slope) and thrust.
Besides, in such a situation, usefull versus misleading parameters are given page 54 of the interim report:
Jeff
PS) does the BEA formulation "There was no search for display of an attitude of 5°" refer to the application of the pich & thrust procedure ? (p 52 of the interim report #2, analysis of past cases of Pitot freezing events).

Such Big Cell ... ?
 

Which one ?
Were you there butterfly68 ?

The BEA choose to not represent the tracks of two flights using the same airway that AF447 used. It appears that those two aircrafts have been through the same big red area that you like to call "that big cell" but managed to proceed to destination.

Do you have any other comment ?

it was just a consideration..the other 2 aircrafts ( I didnt know about them because they are not shown) maybe overflew the core of the cell, you know .. 2000 feet only or more can make huge big difference and the cell was really wide..

always keep in mind that the big systems seen by satellites don't necessarily show up like that on an aircraft's weather radar.
Yes, that's what it refers to.

HN39

At 350 no chance to survive the monster but at 370 nothing to worry about ... I'm not too sure to follow here your logic butterfly68 ?

Maybe there is a question to be asked by the so called journalists to the so called BEA ...

Note that is not known definitely what altitude AF447 was at. The flight plan was to be at FL370, from

Air France 447 - AFR447 - A detailed meteorological analysis - Satellite and weather data

http://www.weathergraphics.com/tim/af447/flightplan.gif


but it has been asserted before that they would not have followed the plan and climbed from FL350 without ATC approval.

If you look at the 10min ACARS positions graphically plotted here

Flight Paths of Flight AF 447 and of the flights that crossed the zone around the same time

and in detail for AF447

http://www.bea.aero/fr/enquetes/vol....data/AF447.txt

and assume exact times there is a slightly shorter sector between 01:50 and 02:00 which could indicate a change? Could also be due to a 15 second shorter time interval/ change in head winds etc. While FL370 may have been helpful to avoid the "SALPO storm" it would not be much use for the "Main MCS cluster"

http://www.weathergraphics.com/tim/a...47-profile.jpg

from the very early analysis by Tim Vasquez
Air France 447 - AFR447 - A detailed meteorological analysis - Satellite and weather data

Later 'official' versions available are by BEA/ France Meteo


The locked rudder limit is stated by BEA to be M0.80, but you have to guess temperature profile inside the CB to get altitude estimate at approx -40 degrees - it seems to have been very hot over the equator.

thank you Sensor for suggesting these data and webpages.. helpfull to understand more that weather situation..:)

butterfly68, the two flights were a Lufthansa 747 (preceding AF447) and an Iberia A340 (following AF447). Their deviations from the track are described in the first interim BEA report. However, the second interim report, for reasons unknown, does not include the plot of their deviations. Both flights did deviate from the track, one to the left and one to the right.

@sensor

The graphic depicted shows estimate that 447 penetrated thunderstorm in upper third, with tops at 50K and above. :eek:

AIM 7-1-28 and 7-1-29 come to mind.

An old rule of thumb that was once taught: if going through a T Storm, penetrate in bottom third. Is this still conventional wisdom, prevailing attitude in re thunderstorms (with avoidance being the normal default position ... )

Insofar as considering "going over" a T Storm, a course of action perhaps considered in this case, can the A 330 fly above 50K? :confused:

The wiki page entry shows a service ceiling of 41,100 ft. Is that correct?

41,100 ft is the upper limit of the approved envelope per Operating Limitations, ref. FCOM 3.01.20. The service ceiling at M=0.8 for the weight and temperature would have been approx. FL370.

Thanks, Hazel, no way they could have gone over that second build up.

Hazelnuts39

Per your service ceiling for 447 (actual), one concludes this flight was bumping at trouble, saved only by extra mach (.82). With "extra" speed to "burn", will this autopilot "hunt"? A gentle rolly coaster at 37,000 could get critical if the PF allows automatic hunting to maintain a/s/altitude, no? If the upsy downsy causes a trip out, isn't the PF in a crapshoot to ascertain where and when to input controls? A PIO could follow an APIO, yes? Isn't it up to chance (possibly) whether it's STALL or overspeed at the beginning (continuation) of upset?

much respect,

bear

Are there a subset of failures where ACARS would stop but all flight critical systems continue to function.

Here are a few thoughts:

1. Lets give credit to the men on the flight deck that they were able to establish pitch/power to take manual control of the flight

2. Lets give credit to the machine and its designers that the failure of pitots (albeit all three of them) will not prompt the airplane to simply fall out of the sky.

3. Lets give credit to the men and women of the search operation.

4. Finally, lets give credit to BEA for its analysis.

If we do the above, we have an important piece of evidence we cannot overlook. The aircraft is not in the area we looked in and so, she continued flying for far longer than we are expecting.

Lets say the pilots established pitch/power. What next? Is it possible to re-engage automatics in the degraded law once they are out of the woods. Or, were they left with flying pitch/power without KIAS on the night flight all the way to Paris. Under such circumstances, would they have opted to turn back and fly to the nearest land mass. Would they have tried to ditch?

Why would the pilots ditch if the engines were still working and there was fuel aboard? :confused: