Henri737 and wallybird7,

There has been a back and forth conversation on this forum about whether they flew into a Cb.

I think perhaps the minimum level of agreement by parties engaged in this conversation is that AF 447, at night with the moon aft, flew into a cloud that had sufficient ice crystals at FL 350 to quickly clog the pitots. There is no consensus on the degree and duration of any turbulence encountered, nor of the existence of updrafts, etc. I believe there is consensus there was no lightning. I don't believe there is agreement on what kind of cloud in the ITCZ might produce ice crystals at FL 350.

Presumably, the next BEA report will soon enlighten us all.

No Evidence of Thunderstorms???
 

SATURNV

No evidence of thunderstorms?????

Please read "Air France Flt 447: A Detailed Meteorological Analysis"
By Tim Vasquez

Every scrap of information on this flight shows a line of thunderstorms directly in line of their flight path.

Every advisory from NASA, to the FAA, advises avoidance by a large margin.

Every professional is aware of thunderstorms and the hazards they pose.

Weather Avoidance Radar is required on every airliner. Why?

What chances did these guys have if they were trained to "understand the aircraft will never stall" and they never knew the stall warning would silence itself under 60kts.

Plummeting towards the ocean with no stall warning might lead someone to think they were heading "nosedown" into oblivion, especially in the middle of the ocean at night.

Quoting One Outsider:-
I didn't say they should or would stop flying the aircraft, One Outsider. That's why there are two of them! But it is the duty of the PNF to read off the ECAM messages, one by one, cancelling each once the PF has absorbed it and then moving on to quote the next.

There's a video here that shows that procedure being carried out. Only a 'sterilised' mockup in the simulator, but it will give you the idea.......


I hope (without much hope) that you will now withdraw the word 'nonsense' .........

Wallybird7,

I have read Vasquez, both his 2009 and 2011 analyses. I have read the BEA Meteo France analysis in the first BEA report (an appendix in French only). I have read NASA's re-analysis of the Wx. I have noted AF dispatch's message to the crew alerting them that satellite imagery indicated Cb along their route in the vicinity of TASIL.

In the earlier 'conversation' in this forum on whether they flew into a Cb or not, I believe there is agreement that:
> nobody publicly knows what the gain and tilt of their radar was set at, and what they likely saw.
> nobody can yet accurately characterize the nature of the turbulence experienced earlier, other than the cockpit indicating to the cabin crew that the turbulence they were anticipating at 0208-0210 would probably be greater than the previous episode.
> they were using their radar, and made a slight deviation off the track before the pitots iced.
> nobody publicly knows what the CVR transcript may say about any conversation in the cockpit about the weather.
> that the weather in the ITCZ is dynamic, with conditions that can change rapidly.

As I noted earlier, there is agreement that they flew into a cloud with sufficient ice crystals to clog the pitots.

SKYbrary - High Level Ice Crystal Icing

cloud vs thunderstorms
 

SaturnV

"There has been a back and forth conversation on this forum about whether they flew into a Cb.

I think perhaps the minimum level of agreement by parties engaged in this conversation is that AF 447, at night with the moon aft, flew into a cloud that had sufficient ice crystals at FL 350 to quickly clog the pitots."

A "cloud" that rises to 50,000 is not merely a cloud -- it is a thunderstorm.
It can contain severe up and down drafts and severe turbulence.
The FAA advises avoiding it.
When AF 447 entered this area a multitude of failures occurred.

And yet you are saying it was merely a coincidence. It had nothing to do with all that followed.
I think you need to do some more research on the subject of thunderstorm flying and the hazard it represents.

RWA:

Is the "first duty" nowadays to listen to the systems or is the "first duty" still to fly the aircraft and use the systems as tools to achieve that end? Hasn't the latter been the general philosophy since the first instrument was added to an aircraft? I can't say with any authority.

SatrunV

Lets not confuse cause and consequences.

what radar system was installed? not seen an answer yet.

Henri 737.

The radar system in use was identified much earlier in one of the previous, now-closed threads on this accident. I am not going to search for the posts describing it. It was the same radar system used by AF459, which followed AF447 along the same track by about 35 minutes, but which after ORARO, deviated to the left by about 20 NM and then to the right by 70-80 NM, and did so after adjusting the tilt and gain on the radar.

Wallybird7,

Regarding the 'conversation' earlier in this thread, I do not believe anyone is asserting that the crew deliberately flew into a Cb whose top was subsequently described as being at 520 (Meteo France) or 560 (Vasquez). AF459 began adjusting its radar after unexpected turbulence at 350 in the vicinity of NATAL, in weather described as cumulus congestus. The crew of AF447 may not have been prudent, but I don't believe there is any indication from the CVR excerpts that they were reckless.

As we don't yet have a full CVR transcript, we don't know what, if anything, was said about what they saw on their radar (except for a reference to the anticipated turbulence and a reference to deviating a bit off the track). We don't know if they adjusted the settings on their radar.

As there was no lightning, and the moon was behind them, there was little illumination of the clouds ahead. We do know that a Lufthansa 744 flying 20 minutes ahead at 325 reported it was in the clouds at the time it deviated. The only weather-related comment by the AF 447 crew in the CVR excerpts released by the BEA references the outside air temperature.

No one disputes that the crews of the LH ahead, and the IB and AF459 following, deviated based on what they saw on their radars. The unanswered question is what did AF447 see on its radar that led them to continue on the track, and only attempt a slight deviation when they were in the middle of the 'cloud'.

SaturnV
"No one disputes that the crews of the LH ahead, and the IB and AF459 following, deviated based on what they saw on their radars. The unanswered question is what did AF447 see on its radar that led them to continue on the track, and only attempt a slight deviation when they were in the middle of the 'cloud'."

No one knows what they saw or what they understood it to mean.
My concern is why they didn't deviate sooner, and raises the question to me, that there is sort of a cultural issue regarding minimizing the potential hazard a thunderstorm poses. It exists also on this web site whereby the automation and it's "Protections" will save the day.
If a pilot is forbidden to "hand-fly" the plane at altitude in clear weather conditions, and then in a turbulent circumstance is forced to do it, with very limited resources, no airspeed etc, I think it is literally impossible.
This airplane is very fragile having 32 plus adventures with loss of pitot info and subsequent auto failures, and to venture anywhere near major storms is irresponsible.

AF447 Weather Radar
 

SaturnV,

Here is a post from Thread 2 in the Tech Forum by Graybeard. It notes the onboard radar type. Hope this helps your discussion with wallybird7.

Thank you Turbine D.

Wallybird7, nearly three weeks ago there was a rather robust discussion in this forum on what they flew into. Rather than my continuing to try to synopsize and synthesize the various views expressed, perhaps the better reply would be to give you links to many of the relevant posts.

http://www.pprune.org/6537291-post1849.html

http://www.pprune.org/6538283-post1865.html

http://www.pprune.org/6538679-post1870.html

http://www.pprune.org/6538743-post1871.html

http://www.pprune.org/6538901-post1875.html

http://www.pprune.org/6538911-post1876.html

http://www.pprune.org/6539033-post1878.html

http://www.pprune.org/6539051-post1880.html

http://www.pprune.org/6539466-post1885.html

http://www.pprune.org/6539692-post1889.html

You may view the failure of the crew in not deviating sooner to be irresponsible, but I think most/all of the earlier posters on the subject of a track through an ITCZ meso-convective system would agree there still is far too little information to form judgments with regard to culpability.

As Vasquez summarized last month,

Hi,

This is what is called "Russian roulette"
A game can not be a winner ..you always lose .. it's just a matter of time.
To be sure not to lose (or sure to win) should not play "Russian roulette" (do not go in these areas)

Next report of BEA is planned to be published end this month... I´m sure we all are planning to do our best! Fly safe!

I hope the final report comes out soon from BEA so we don't have to listen to Russian Roulette stories of how they iced up.

This leads to the same circular argument as volcanic ash. The air we fly through always has some degree of ash, water and turbulence. You can't just say don't fly in them because it's like how high is high.
The pilot can be given guidance but can not be given orders that are black and white without the means to measure. Thus the investigation needs to ask about the guidance and the means as well as the performance of the crew. As far as I know the investigation has not made a conclusion in this regard so why should we?
Weather is dynamic so what another flight did a half hour before or after is only a hint, towards their guidance and means of measurment and not a conclusive finding for the accident flight

Quoting DL-EDI:-
The professionals on here can advise you better, DL-EDI - but my understanding is that in this sort of situation (autopilot/autothrust 'sign off,' 'unreliable speed indications,' etc.) the PF should 'fly pitch and power' (basically keep the pitch attitude at an appropriate level to maintain altitude and (given that one of the few things we can be certain of is that the speed indications had gone haywire) manually apply the correct throttle setting to maintain flying speed). The duty of the PNF, as shown above, is to work through the warnings/messages so as to inform the PF as thoroughly as possible on what has gone wrong. So it's a mixture, with the emphasis on keeping the aircraft at a correct attitude and at a safe speed until (hopefully) the instruments come right (in this case, first and foremost, that the ice in the pitots and ports hopefully melted).

That's the way the AF447 pilots reacted; and it looks as if the PF did in fact manage to keep the aeroplane level and maintain airspeed for some 18 seconds; in other words, 'fly pitch and power' in accordance with the recommended procedure. I've no doubt that, if and when we ever get details of the CVR recording, the two pilots will have been going through the procedure illustrated above - checking each message/warning in turn, trying to find out what had happened/was happening.

What we DON'T know, of course, is what other instruments may have been misbehaving, in addition to the Air Speed Indicators.

Then came the (so far unexplained) zoom-climb, with no evidence of any sidestick input from the PF. Again, the PF appears to have 'done the right thing,' countered the zoom-climb with forward stick and pretty well levelled the aeroplane out again. Further, at some point during this period (the BEA 'neglects' to inform us when this process started and finished) the "Trimmable Horizontal Stabiliser' (what used to be called the 'tailplane' in my day) unaccountably went to 'full up.'

Finally, the stall warning sounded - and once again the PF appears to have 'gone by the book' as it applied at the time, carrying out the recommended procedure at the time - full Take- Off/Go Around power and try to maintain level flight.

At some point during this phase the "Trimmable Horizontal Stabiliser' (what used to be called the 'tailplane' in my day) went to 'full-up'; which would have made it just about impossible to maintain level flight. There have been several accidents due to that happening; Perpignan was one (though pilot error was definitely involved in that accident as well). Another was an MD80, Alaska Airlines Flight 261 - this accident was put down to lack of maintenance, the un-lubricated jackscrew operating the THS first of all jammed with the THS at a 'nosedown' setting, and then broke off altogether. For the record, the Alaska pilots had both to haul their columns back with all their strength just to maintain some sort of level flight, just from the 'down' THS attitude, even before the THS actually broke).

Alaska Airlines Flight 261 - Wikipedia, the free encyclopedia

TV dramatisation here too if you fancy watching it - though I warn you, it's long.........


Trouble is, due to what I will politely describe as the BEA's 'reticence' on the contents of the FDR and the CVR, we can form no opinions on what caused the zoom-climb. That brings us to the main difference between modern airliners (particularly Airbus, though Boeing are moving in the same direction); that older aircraft types always left the pilot with the final word; whereas modern 'systems' are designed to overrule them if the systems 'conclude' that the pilots' actions are in some way endangering the airframe or doing other things that (in the opinion of the designers) are dangerous.

Both Boeing and Airbus modified their 'stall avoidance' procedures soon after AF447 went in; the drill now is 'adjust power but don't necessarily use full power, concentrate first on getting the nose down.'

Hope some of that helps.......

According to the note more like 11 seconds. Anyway, is that supposed to be a good thing? Orville Wright managed 12 seconds on his first try...

At this point we have nothing indicating that they lost anything other than the airspeed derived data either.

While I agree that we don't know how strong it was, it's somewhat hard to ignore the "left nose-up input" made just before the climb...

Whatever book they were going by, I'd hazard a guess nose-up inputs resulting in 16 deg pitch attitude weren't in it.

It went up because the pilot wanted nose up (that that was probably for a good reason from his perspective is a different story, we will have to wait and see...).

I'm not sure where you see the relevance of that incident, but as you note there the pilots fought with all their strength against the failed THS, which from what we know so far seems not to have been the case at all here...

So, general anti FBW ranting aside, at which point do you suppose the FCS overrode the pilot's input, causing or contributing to the accident?

That's quite an extensive reply to my simple question, RWA. However, I still don't see that nowadays "listening to the systems" is the crews' "first duty" over flying the aircraft.

As a mere armchair enthusiast, I'm also curious about the answer to this recurring question.

As the A/C was supposedly in Alt Law (ADR Disagree) there would have been NO FCS inputs to cause the pitch up, they are inhibited.

RWA

The manufacturers and regulators re-emphasized the Stall Recovery Procedure not because of AF 447 but because of various other stalls on both Airbus / Boeing and other types. Too much emphasis had been put on minimum height loss with some "instructors"/ schools actually saying use "back" stick to reduce height loss!!!! The UK CAA have just issued a Safety Notice to that effect on the 13th July 2011.

BEA reportHowever, if you read the BEA report quoted, the inputs made were then nose up and more than likely caused the THS to move, then when the AOA increased above 30 degrees or the speed dropped below 60 kts - no more autotrim.

Hi,

As we have many technical documentations about the A330 systems and how they are working in the case of the AF447 event .. and pilots actions (thank's to takata and al and BEA interim reports and note) can some ones with same knowledge about the Boeing 777 .. propose a scenario about how will work the B777 systems if confronted at similar event ?

RWA,

I do realize that as a professional pilot I have become outnumbered here by instant experts who can tell us 'how it is' by watching youtube videos and quoting wikipedia and what not. I note that you also think you should 'give me the idea' of what I do for a living. Should I say thank you or just laugh?

What slipped past your eagle eye is that the A320 in your video is level in cruise with AP and ATS engaged and under control. It is a video of set piece like a free kick at the edge of the penalty box. It tells you nothing about how to get the ball there in the first place or even how to play at all.

I think I will stick with nonsense, thank you very much.

Hi One Outsider,

As a non-professional pilot, and therefore one of the multitude who now out-number you, there is still one burning question in my mind. I'm not being cheeky towards your post, it's just the most recent one I can reply.

Why the zoom?

I have flown before, in fact I trained in the Aussie Air Force. Ultimately I was unsuccessful on that course because I incorrectly applied good 'airmanship' on several occassions. Those fo-pars that ultimately failed my endeavours were far less serious than failing to apply known procedures to a system failure.

I've said it before, and I'll keep saying it, unless there was some monumental failure of things we haven't heard about yet, the pilots appear ill-trained / equipped to deal with what should be a simple emergency.

Is it actually is possible (is it? I really do not know) for the system to have the last word? Be honest please, think, is it possible for the system with A particular set of circumstances to override the pilots? I know it is like asking what if a wing falls off so please humor me with this seemingly naive question. I am just a old DC8/9 guy and the only glass I have flown is on a helicopter.

Thunderstorms - speculation
 

Lomapaseo

All the other aircraft in the area deviated. Why not this one?

Two posts in a row state as facts about the aircraft flight into a defined thunderstorm without deviation.

I challenge those facts

Where's the investigative data that so states?

To continually speculate for a simple explanation does nothing to furthur the understanding of a complex accident.

lomapaseo

Check out " Air France Flt 447: A meteorological analysis "

Not speculation!!

BEFORE LANDING CHECK LIST

I'm not fully aware of the A330 specifics -- but it's not so much having the last word, as it is -- doing something the pilots were not aware of.

And my take is with the A/P and A/T clicked off, and in turbulence, without airspeed guidance, and the pilots not allowed to hand-fly an airplane at altitude, and a overly sensitive control system -- it seems to me to be very easy to over control and not be aware of it and also unaware of the side stick inputs they are putting in.

Otherwise why a unusually high pitch attitude? The trim just did what it was told to do.

So... the big question I was leading up to; WHO (Man or machine) told it to do what and why?

before landing checklist,
I think the point is - wouldn't you have been just as dead when applying the same inputs to a DC-8?
If anything, the bus gave them a bigger shovel (autotrim), but it was their decision to start digging. How they'd arrived at that decision seems to be the interesting part.

wallybird7, you are suggesting that they deliberately flew into a defined thunderstorm? Defined by who? and when?

From a few pages back in this thread, mm43's corrected image:
http://oi53.tinypic.com/2u7n4vs.jpg

IMO, neither the Lufthansa nor the Iberia flights deviated enough to take them out of the bright green. (I concede I am using a static image to describe a dynamic situation, but the BEA has chosen not to release the precise path of the deviations by those two flights so one can't map against different satellite imagery.) AF459 deviated enough to the right to possibly avoid the bright green.
_______________

Do you have knowledge of the crew of AF447 adjusting the tilt and gain of their radar as AF 459 did, and still proceeding on the track?

wally:
A course deviation of 12 degrees was recorded before this upset event. It's in the BEA report. Was the deviation big enough? Different question.

I'll echo Saturn V's other point: defined thunderstorm conclusion is based upon what criterion? That they knew they were flying into unstable air seems to be confirmed by the brief CVS discussions released to date, and the apparent decision to reduce speed to turbulent air penetration speed before they hit what they expected to be a patch of the rough stuff.

Pre-flight briefing of AF447 will shed light on whether the storm cluster was expected and to what extent.
Keep in mind that the image above does not use the same color coding as the one we see in on radar. So the AF447 radar could have seen anything from nothing to a wall of red. From what I have read here the use of the gain adjust is critical when traversing the ITCZ.

In all my years I have never done a preflight briefing of weather enroute. We look at what is shown and take off knowing what to expect. They must have encountered some icing and their pitot tubes froze up. That alone apparently took the airplane down for some reason. There is no evidence of their deviation 12 miles left of course not being adequate.

Using the fact they crashed is silly. If they iced up 40 miles off course probably wouldn't have made any difference. Hopefully the final report will clarify everything but right now we have zilch info.

Assume UAS. Without Airdata, the a/c cannot be autoflown. A new Flight Law is entered, and the PF is the only way to get home (if but temporarily).

Fine. This aircraft has a reputation for being docile, easy fly, and hand flying no problemo. So why no auto flight? Again, no reliable Airdata.

So on the one hand, UAS is not a 'big' deal, but big enough that the a/c cannot do it. Just fly Pitch and Power, one hears.

Fine. Isn't Pitch and Power the easiest (basic) way to fly? So why no auto P/P?

At altitude, in those conditions, PJ2's post describes the hazards of flight re: control damping and authority.

Fine. With turbulence, the a/p has the same airmass to deal with as the Pilots, subject to overcontrol, sensitivity, g and PITCH. So the a/p can do these, but not something so simple as PnP? PnP is what it has been doing since ORARO, with flourishes, and a nice ride?

It is the fundamental duty of autoflight to keep the a/c stable and consistent. What about the challenge of UAS has nothing to do with what an a/p needs to do every day, all day long?

Yet the PF has no access to the accelerometers (except by display), and has to deal with natural conditions that don't affect the a/p one bit?
IE Dark, no constant updata, distraction, and ECAMs?

For once and all, can someone address why it is impossible to consider that the autoflight couldn't (keep up) even with reliable a/s, and UAS may have been the result of insufficient response in the auto regime? (turbulent air mass, discrepant instant baddata?)

someone, splain.

UAS vs autopilot
 

Because, in foreseeble future AP must be kept simple. AP and similar software (it's now software thing) is, compared to many other software systems, quite simple. And if it's going to be certified to fly 400+ pax in crowded skies over densely populated areas it must be kept simple. The key feature of such life critical systems (like AP, FBW controls, dangerous industrial processing controls, like chemical or atomic) is traceability. I.E. each action of the hardware must be clearly traceable to particular instructions in software source code. Then the thing must be proven (often formally, mathematically) that it faightfully represents it's specification.

You want P'n'P of that software, but now how to solve such things like P'n'P while climbing, while descending, while keeping altitude (esp. in reduced vertical separation space), etc. Then if there is UAS AP must be sure it is isolated and is not a sign of some other bigger problem. The human interface would also get more complicated and hard to grasp.


And there is one thing to all that -- the software is deterministic (and in transport planes it will for some significant time into the future). It's meant to always behave the same given the same situation. It has no second thoughts. It has no thoughts at all to begin with. It is preprogrammed for particular behaviours in particular situtations. If situation is not fully understood beforehand the best action is to disconnect the thing. Otherwise it could turn into garbage in - garbage out and you don't want garbage out while moving at Mach 0.8 at FL350.
But there are those two guys (or gals) in front. They're not there to just monitor all that machinery -- with current technology they could better monitor it from some warm place on the ground (as it happens with spacecraft, for example). They're there to handle all the situations which could develop.

Yes, because those guys and gals are not preprogrammed. AP has no thoughts no understanding. None, nada, nil. All additional conditions like ECAMs etc would add exponentially to complexity of preprogrammed AP.


UAS could not have been the result of insufficient response. Air which would be 60kts on one side and 180 on the other would simple tear airplane apart. UAS was due to pitot fail not due to turbulence causing reliable but discrepant measurements.

Originally posted by Bubbers44 : Congratulations "Bubbers", at last one sensible post on this subject :
1) I can't, either, remember of one pre-flight route modification due to en-route weather ;
2) as far as we know, the cause of the crash is not "flying straight into an active cb" but pitot icing.
Icing conditions can be found anywhere : in, near or far away from convective cells.

Thanks. I see your point(s). You have focused on the crux. Obviously, autoflight is well-equipped to fly the a/c. It does so virtually all the time.

How well equipped is the Pilot(s)? The irony is that the auto is sensitive, sophisticated, and connected directly to sensors that are processed instantly, allowing for smooth flight. Hand flying cannot produce the ride in most circumstances that auto can.

Two circumstances:

1. The autoPilot cannot keep up, drops out, and Normal Law remains.

2. Due UAS, the autoPilot drops out, and a/c reverts to Alternate Law.

Both conditions require hand flight immediately ("I have the controls").

It would seem to me that under condition (1), the a/c would revert to Alternate Law, the a/c is having control issues.

It would also seem to me that under condition (2), nothing has changed except an instrument is U/A. An important one, but one that hasn't to do with control, at least initially.

So Condition '1' wants (needs) immediate hands, and condition '2' is recommended "patience", monitor, do nothing, don't maneuver.

Protections would be better retained in condition 2, yes? Especially so since the a/s is not reliable, and ham handedness is more likely?

Protections also for condition 1, since the a/p has indicated control challenges for the PF at handoff?

This should be rephrased as: if Normal Law remains, auto pilot CAN keep up, there is no reason it drops out (none, zero, zlich, nada, rien, niente, whalou).
Consequently, only your case 2. seems true.
Beside, very good post from Sebaska (above).

As an SEP can I please ask you very experienced pilots a question?
If you are flying an (any) aircraft with no outside vis. and suspect instruments and your body and altimeter gave every indication that you were hurtling downwards at a high ft/sec. would it not be a 'normal' reaction to pull back on the stick irrespective of any flight instrumentation. Could this therefore be a case where old fashioned 'seat of pants' reactions were the wrong ones and less 'flying' training and more 'flight' training might, in this case, have yielded a better outcome.

takata noted.

funfly At 10kfpm down, ND creates even more fpm down. Counterintuitive to want to speed up. It also increases the 'g' when one pulls "out" of the ensuing dive (I think 'g' protections won't let him pull out anyway).. PF has surpassed all his flight training when the a/c STALLS. A guessing game for him, and instruments that, even if accurate, would be hard to suss?

Protections or automatics ?

Protections, I'd say definitely not. bad things can (and have) happen when protections kick in on bad data (or on good data with design rules that turn out to be not so good). To engage protections based on known bad data would be poor design.

The protections are a line of defense at the edge of the envelope where you (auto or real pilot) shouldn't be going in normal flight anyway. It shouldn't matter if they aren't there if the plane is flown properly. Yes, that might be more of a problem with UAS + IMC + Turb etc., but it will be a lot more of a problem if the plane starts protecting you based on invalid data. Some UAS events have reported more or less simultaneous stall and overspeed warnings...


On the other hand, having a degraded A/P that held pitch and power (correct for altitude - assuming that's not dud data as well) and probably wings-level might actually be useful in some situations.