Any updates on the Airbus lost in the Atlantic?

I've done 6000 fpm. G meter from that flight showed min of -3.5 and a max of 4.5. But those weren't under commercial conditions. If a pilot has his head down looking at the instruments and is confused, it's not going to end well.

 

Thanks Taz. You said it in a nicer with an excellent example. I'll bet those student pilots would have taken long odds that they were flying straight, level and constant speed. It's possible the AF crew couldn't overcome the descent, but I think what some aren't understanding is this descent took a long time (in relative terms) and it was very possible the crew hadn't the slightest idea they were descending.
Hitting a downdraft in a plane, cresting a hill in your car or the initial descent in a fast elevator have no relationship to what this crew was experiencing.

 

LJ- Yes, I was the instructor, and no, the aircraft came nowhere near departing. When the auto pilot could not hold level flight, it disengaged and the aircraft immediately started a descent, which wakes up the pilot. Stall warning on an F-111 was at 18 degs AOA or 14 degrees AOA plus pitch rate to exceed 18 degrees AOA. Actual departure was considerably higher than that. The autopilot disengaged at 10 degrees AOA.

You never intentionally departed an F-111. Because the sink rate was so phenomenal, there might be insufficient time and altitude for a recovery. On the only attempt to spin the F-111, the aircraft would not recover, so the spin chutes were deployed. They immediately ripped off. The aircraft was descending rapidly, and this is where the 50,000 fpm descent rate was generated. The crew were still trying to figure out what happened when the chase plane crew started yelling "Eject, Eject". The crew looked around and later said that part had never occurred to them until they heard the eject calls. They successfully ejected, the aircraft left a very compact hole in the desert, and nobody ever intentionally departed an F-111 again.

The difference between something like an F-111 departing and an A330 departing is the Airbus departs in a stable attitude and the F-111 departed violently with a huge, slicing yaw in one direction or the other due ot lack of keel surface and the rudder being blanked at high AOA. One really gets your attention, the other does not except for the horns going off.

Remember also, the A330 was departed at high altitude where it would be barely flying in the coffin corner under normal conditions. Heavy weight and barely enough thrust and lift. The sensation of loss of lift would have been much, much less than at lower altitude. One second she is barely flying and the next she is on a ballistic trajectory.


Taz
Terry Phillips

 

When I was doing my IFR training, working on unusual attitudes under the hood, my instructor would often crank the aircraft up into a zoom and then gently roll to a knife edge so that as you were given control the airplane was falling through the horizontal or near stall, or pointing down at low speed but aimed at the ground. If you looked at the gages you would see that you were pointing down and yank back but that would stall you because you had no airspeed.

I later asked him why he put the aircraft into a zoom while I was looking down. His comment was that pulling some g's earlier and then zooming was particularly disorenting. He said that often the aircraft could be near stall, or falling really fast and when the student went back to the gages he didn't know where he was.

 

Taz-- I think your use of the phrase "coffin corner" is incorrect, at least in this instance.

My understanding of the "coffin corner" has always been that it takes place at an altitude where the margin between low speed buffet and high speed buffet is minimal-- in extreme cases, as little as 20-30 knots (although I'm not aware of ANY certified airplane with limits so low-- I think that the Lear 20 series, at FL510, has something like 60-70 knots of margin, for instance).

The A330 has a maximum operating altitude of FL410, and an Mmo of 0.86. According to the attached buffet boundary chart (which I found on the internet), the low speed buffet at FL380 takes place around Mach 0.58, depending on CG position and at 1g. Of course, the low speed buffet boundary increases dramatically with g-loading, but that's true of any aircraft.

So the airplane was nowhere near any reasonable definition of "coffin corner." They had a margin of between M0.58 and M0.86 (or probably higher, since that's the Mmo) for buffet boundary.

That's a lot of margin.

I will call your attention to the "CAUTION" note on page 4:

http://www.google.com/url?sa=t&source=web&cd=7&ved=0CDgQFjAG&url=http%3A%2F%2Fwww.smartcockpit.com%2Fsite%2Fpdf%2Fdownload.php%3Ffile%3Dplane%2Fairbus%2FA330%2Fmisc%2FA_0_Limitations.pdf&rct=j&q=a330%20limitations&ei=5pniTbyCFYPhiAKkhJmSBg&usg=AFQjCNF80CSwxtLT53EH7mQtXScYc5xlOw

 

In the PPRuNe forum there is a lot of discussion as to what warnings were issued and when. Even some of the Airbus pilots have a concern that the stall warning was shut off after the aircraft stalled. In this case the stall warning went off, but then silenced itself after the airspeed dropped below 60 kts.

I don't quite understand why that is. If you are below 60 kts and have a big pitch attitude you are probably already stalled. The horn came back on when the pilot reduced pitch angle and the speed started to come back up (nearing recovery). If you were pushing forward and a warning horn went off (and yes it is probably a different horn than the mach horn), would you continue what you were doing (pushing forward stick) or would you pull back to shut the horn off (which is what they did)? The horn shutting off when they pulled back again, reinforced their actions as being correct, even though it was not.

The software should be smart enough to know that the aircraft is stalled. Big AOA, low airspeed high vertical speed are plenty to determine that the aircraft is probably stalled.

Most commercial aircraft have a stick pusher and, in a sense the Airbus does too. If you approach stall the aircraft pushes the nose back down to prevent stall. Problem is, when the airplane goes away from Normal Law (in any degraded mode) that feature goes away, so the aircraft can and in this case was stalled. In other aircraft the stick pusher system is a simple system that uses AOA and pushes the nose down if it goes out of limits. I don't see any reason why a simple stick pusher equivalent shouldn't be on these airplanes that prevents stall even in the degraded control modes.

 

Don- The coffin corner I was discussing is the one where the lift and thrust at altitude no longer allow the aircraft to maintain level flight. In this case the A330 was heavy weight and trying to fly above bad weather, so the margin between cruise and stall got slimmed down quite a bit. A little more altitude and you stall from lack of lift or lack of thrust or both. The aircrew raising the nose when they should have been lowering it both cut down lift and decreased thrust, pushing them beyond that coffin corner.

There is a huge difference in where some jet aircraft can fly depending on their weight. The A330 was not that far out of Rio when the trouble occurred and still very heavy, pushing the coffin corner down in altitude and higher in airspeed. If ATC allowed, cruise climb on many aircraft is the most efficient way to get from A to B if it is a long distance. As the fuel load lowers, higher and more efficient altitudes become possible while maintaing a control authority margin. I have flown many times at about as high an altitude as the aircraft would maintain without using afterburners. It often feels as if the aircraft is teetering around its center of gravity, and all control inputs have to be very gentle or you lose altitude. It is very difficult to fly formation at those altitudes.

Taz
Terry Phillips

 

solo

I tried cutting through the noise on that issue and could not. From my readings it sounds like the stall warning went off until the airspeed indicated < 30 kt (as opposed to 60) I'm not as experienced in as many aircraft as many here but I don't think that is uncommon.

Even still, if you're getting 3 stall warnings in a 90 second window you DO NOT hold the stick full up for 30 consecutive seconds. (which the PF did)

Here's my theory... (and I could be wrong)

================
I think the PF did not honor the threat of the stall warnings because he figured they were bogus since the pitot was iced. (which is not how it works on an airbus but that's all I can figure) I think it went down like this:

Turbulence
AF kicks off
PF figures he's try to climb to get over the weather
PF gets stall warning and ignores it figuring the pitot is iced and he's at TO/GA etc.
----From here I dunno what he is thinking unless Capt is also flying-----
Throttle gets kicked to 55%, stick goes down for a while etc.
The trim issue -- Then PF holds full up for 30 seconds. (still trying to get over weather?)

I'm at the point where I'm far more interested in the voice than the data.

We know HOW the plane went down, we don't know WHY. I keep obsessing on what the PF was thinking for one simple reason... We'll never know WHY the plane went down until we understand his actions. His actions are so irrational it is inconceivable a trained pilot would make them. (assuming any reasonable situational awareness)

One guy I read said the PF committed suicide.... Now I love to bash conspiracy theorists but even I gotta confess, that's not the most kooky theory I heard this week. At this snapshot in time, I can't give you a better explanation for his actions. (save my feeble attempt above)

I dunno, I get the hows, but the whys still are still a mystery. I'm hoping the voice explains what the guy was trying to do with the aircraft.

 

430 man

They had to be thinking that they were nose down and at high speed. If they thought they were stalled they would have pushed forward.

They got a stall warning but didn't stall, they then got a second stall warning and stalled. At just after that the airspeed went low enough for the computer to disable the stall warning (and according to some the autotrim also).

Put yourself in their shoes. You get a stall waning and then a few moments later, the warning stops. Have you recovered from the stall, or not? Well, if it was me and the horn shut off I'd assume that I wasn't stalled anymore. Every airplane I've ever flown had a horn that was blairing at me when it was near stall, and it didn't go off until I recovered it.

So now, you are falling like a rock and think you are nose down, you are pulling back like a gorilla and nothing is happening, and you think you are in a dive. You naturally keep pulling.

You think about it for a minute and the you say to yourself, "well maybe I'm stalled", so you release pressure and push the nose down some, and the warning horns start blairing at you again. So what do you do? You pushed forward and the warnings came on, so then you pull back again and the horns shut off. The warning was exactly backward. Was it a Mach horn or was it a stall horn? At this point you are really confused and the airplane is still losing altitude.

Easy to see how it could happen. They were in the dark, IFR, nothing outside the aircraft to see, only the HSI display for a reference and they didn't put it together that they were stalled. I think that at first they thought they were in a dive, a stall is such a rare thing in this automated aircraft that it maybe they didn't think it could happen. They tried to start a stall recovery and the warning system told him that was the wrong thing to do. Also, the auto trim shut itself off at the same time of the second stall warning, adding to the problems with recovery.

While the crew made a call out of alternate law they didn't indicate if the crew caught the fact that the fact that the flight control had changed modes.

 

No because you don't magically go from stall to overspeed in 10 seconds. There was no overspeed warnings of any kind.

Also as far as the stall warnings being backwards, you have a bunch of people saying a bunch of crazy stuff on prune which may or may not be true. (Unless you found a definitive post, I only found yahoos that had never flown a day in their lives pontificating about how horrible airbus's stall warnings where... gotta love that internet) -- I spent a few hours on the two main prune thread and the signal to noise ratio made it not worth the effort.

From reading the report (and it was not 100% clear) it sounded like the stall warning went off for 30 seconds and the guy was holding full back stick the whole time.

But EVEN IF your stall warning goes off and then stops... do you then hold the stick full back for 30 full seconds?

You said: "They had to be thinking that they were nose down and at high speed. If they thought they were stalled they would have pushed forward. "

What on earth would make them think that when 3 different stall warnings went off and no overspeed warnings that they were overspeed? Besides, like one of the Boeing captains said, if his plane is overspeed he damn well can hear that. I can not bring my myself to move that into the realm of plausible. (I give the suicide theory about a .001% but I put presumed overspeed almost the same place... if they thought that with no overspeed warning, no air speed indicator, after three stall warnings AND not hearing the speed, then they were REALLY confused.)

BTW please be careful repeating knuckleheads on prune unless you can fact check them...the whole 'the stall warnings were backwards and when the guy put nose down the stall warning sounded' is not based on the report, it is based on knuckleheads spreading FUD. You and I have been down that road, let's not go there again.

 

Taz-- these guys were not trying to outclimb the weather. They were level at FL350 when they lost airspeed, and the autopilot disconnected. In the confusion, the aircraft climbed to FL380, and stalled, but it doesn't look like that's what they were trying to do.

I don't know about tactical aircraft, but in my business jet experience, we would never climb at such a slow speed that low-speed buffet was ever a concern-- because, if we ever got that slow, we would not be able to accelerate to anything like a normal cruise speed.

While it's a little hard to understand why they reacted the way they did, I think you have to remember that they were in a thunderstorm, getting bounced around, confused, dealing with multiple conflicting data sources, and the airplane was not reacting in the way in which they expected.

 

From the BEA report

The report doesn't say the horn went back off after the pilot appled forward stick, but if the pilot pulled back again and if the speed went back down it would have.

So, going back to the stall event, the stall horn sounded, the pilot added power and maintained the nose up and about 15 seconds later the speed increased and about 25 seconds after that the stall warning stopped sounding.

If you did those actions and the horn stopped what would you think? When the horn stopped I would think that (and I believe the pilot figured too) that I had flown out of a near stall. All of this was happening at the top of a ballistic arc. The aircraft didn't have sudden drop and a break like a normal stall because of the way he entered the stall, at the top of a zoom climb. If he didn't feel a stall and the horn stopped, I would not believe I had stalled it. Why would the horn stop if you were still stalled? The pilot kept trying to climb and that wasn't happening so what was wrong.

The horn was not on for about a minute and then the pilot pushed forward on the stick and the stall warning was sounded again.

Put yourself in the pilots shoes. You think you flew out of a near stall, you have TO power on and you are still descending, and then you push forward and the stall horn starts up again??? WTF is the airplane trying to tell you??? You don't believe the ASI you only know you are running out of altitude.

Confusing to say the least, and I don't have any trouble thinking that they didn't believe they were stalled.

 

You are correct. I do not think they thought they had departed controlled flight and that is the crux of the accident.

Don- I do not think we are disagreeing any more. The aircraft was probably at the maximum cruise altitude for the gross weight they had. A bit higher and a bit slower, and she departed. Like 430man said, there seems to be no logical reason for the control inputs the flight recorder documented after she departed.

Taz
Terry Phillips

 

solo

First you are quoting an edited version of the report and you're missing some important information. I've seen several of the nutty prunners posting edited hunks of the report that (surprise) are all edited to make airbus look bad.... The real report I linked above is here: http://www.bea.aero/fr/enquetes/vol.af.447/point.enquete.af447.27mai2011.en.pdf

I will bold the main parts you missed and reformat it a bit for readability:

Starting at 10 min 50 *and after 2 previous stall warnings at ~38K feet in a heavy aircraft*:

+ 0:00 Stall waring sounds, PF maintains up input (huh?)
PF increased trim from 3 to 13
PF held *full up* for 30 straight seconds. (I reread it and YES while stall warning is sounding)

+ 0:50 "During the following seconds, all of the recorded speeds became invalid and the stall warning stopped."

Note: When the measured speeds are below 60 kt, the measured angle of attack values are considered invalid and are not taken into account by the systems. When they are below 30 kt, the speed values themselves are considered invalid.
----------------


Now your question is, what would I think? Well, I have no idea solo because I would not have increased trim and held the still full back for 30 seconds while the stall warning is sounding the whole time.

WOULD YOU? (please answer that one)

Your theory is based on the fact he got confused AFTER these events. I submit he was very, very confused before this event happened.

 

Would I assume that, unless the aircraft was fully enveloped in the clouds, that none of this would likely have happened if this had taken place during daylight?

 

It certainly increased the degree of difficulty, but then you can play what-if games all day long.

Most accidents are caused by a chain of events, and if you broke any link in the chain, the accident would not have taken place.

In this case, the chain went something like this:

1. First mistake-- flew into a thunderstorm. Certainly the fact that it was night contributed to this, as did the fact that they were over the ocean and most likely couldn't see what others have done. I'm willing to cut them some slack on this, as most of us who have done this sort of flying have made this mistake at least once.

2. Pitot tubes iced up. Were they defective, or was there so much ice that the heat couldn't overcome it? I don't know the answer to this.

3. Crew failed to maintain basic aircraft control, once they lost airspeed indication and some of their automation. While I hate to pile on to the crew, it's hard to argue that they didn't have sufficient information to properly control the airplane. They had reliable attitude indication and power the entire time, and that should have been enough.

Any time I criticize someone like this, I feel bad. After all, I wasn't there, and they clearly did their best. Would I have done better? I like to think so, but until I'm in a thunderstorm with conflicting information and various warnings going off, and the airplane not responding as I expect it to, I guess I'll never know. Hopefully I will never find out.

 

Are you Boeing guys saying that the Airbus is unsafe?

Dale

 

Does the sun rise in the east? lol

 

Deep Stall:

http://en.wikipedia.org/wiki/Stall_%28flight%29

Ok, Wiki says the turbulent, unstable air coming off the stalled wings of an aircraft hit the tail, and on "T" tails, this makes the elevators useless, or very degraded in response.

Could the autotrim on the stab been effected by the air and it "failed" in the nose up position, after the autopilot disconnected?

Would this alone put you into Alternate Law from the FCC? or shall we suspect the pitots are iced causing invalid data, triggering Alternate law?

Would this effect on the tail also cause the rudder over travel messages seen on ACARs?

Thanks for putting up with me.

 

I am lost in all this techno-stuff with the instrumentation and messages. I can relate and understand most of it but my old head keeps going back to the pitot static system failure as the root cause of this incident and all the computer functions seemed to work as designed albeit it blasted an overwhelming load of information at the crew. I still can't believe that the pitot masts and tubes weren't heated enough to prevent icing. Have any of you professional pilots out there ever heard of a Boeing airplane having this problem? I haven't and I wonder how it happened.

As far as stalls. Blanking the horizontal tail with stalled turbulent air can happen on a low mounted tail or a "T" tail. It depends on the angle of attack and the profile of the bad air. After the KC-135 incident Boeing devised the leading edge Kruger Flap to prevent the separation of flow at the leading edge. It was also effective on the 727 that had a horizontal tail blanking flow that could lock the airplane into a fatal deep stall. Again, it depends on the A O A.
Image Unavailable, Please Login

 

If I didn't think the aircraft was stalled I might well have applied aft stick to start to climb again. Remember that early in the stall the vertical speed wasn't great, the aircraft was stalled and starting to fall, but the pitch attitude was small (6 degrees nose up), looking at the HSI the pilot would have seen a slight nose up conditoin and likely never felt a stall break. The airplane wasn't in a high angle of attack stall yet, it was just stalled at a low load factor. Have any line pilots actually stalled an airbus? Not many, any stalls that are done in the simulator don't necessarily feel the same as they would in the real airplane. I'm willing to bet than none of these pilots had ever experienced a stall break in an Airbus at altitude. It is totally plausable that they didn't recoginzed they were stalled considering the turbulence that they were in. The pilot had appled T/O power and expected the airplane to fly out of it.

You are saying you would not increase nose up trim, nobody would and nobody did. The aircraft did that all on its own. While the stall warning was sounding the aircraft was in alternate law. This means that the auto trim was still fully functional. In response to the pilot nose up command the auto trim system set the trim to 13 degrees nose up. When the stall warning stopped sounding, the aircraft went into alternate law 2 and in that mode the auto trim is disabled. Simply put, no warning horn, no auto trim. The trim was disabled in a high nose up position. Did it get there because the pilot applied a nose up command? Absolutely, but he didn't change the trim, the airplane did it because that is what it is programmed to do.

Once the trim was set to the 13 degree nose up condition, unless it was reduced, the aircraft was probably not recoverable with the engines in TO power. The reason I say this is that when the ANZ Airbus crashed, the autotrim set the THS to a similar condition. In that case, as the airspeed built back up after the aircraft nearly stalled, it pitched back up and stalled again. With the fully loaded aircraft and the fuel in the trim tank at this point in the flight of AF447, the CG would would be further aft than it was in the ANZ crash.

You are saying that the pilots were very confused very early in the process. The loss of airspeed and the system warnings are plenty to set the stage for confusion. In my experience with how pilots cope with in flight system failures, lots of lights, annunciated warnings and horns can very quickly and easily lead to him taking exactly the wrong actions. Been there and seen that happen, and when we dissected it down to each action taken, they were wrong, but they were logical, just like this series of inputs were. I think I've explained a very plausable series of events, taking into consideration the limited knowledge we have at this point of what they did and when.

There were three pilots in the cockpit and none of them recoginzed the fact that they were stalled. If the two not flying had recognized it they would have been yelling to initiate a stall recovery. That didn't happen, so I'm thinking that none of them realized it.

We have the advantage of sitting here in front of a computer looking at the recorder data and saying these guys were idiots and of course you or I wouldn't have taken that action. What I'm saying is that the actions taken, in the context of what they saw and felt, weren't totally illogical. They didn't percieve what was happening and didn't take the right actions and paid for that mistake with their lives. But to assume that they were stupid or totally oblivious to what was happening is not the case. I am not so arrogant to assume that I would have not done it differently.

 

I don't think the airplane was really in a deep stall, although it may have looked like one. In a deep stall, you don't have elevator effectiveness, and thus full forward stick (and even power) won't get you out of it.

In this case, the flying pilot never released the back pressure on the stick. This would probably cause the airplane to mush forward in a nose high attitude. If he had released back pressure, the airplane would probably have flown out of it.

From reading the BEA report, it looks to me like they may even have had valid airspeed for most of the descent. The problem was that with the airplane in a sustained stall, mushing forward, the airspeed seemed unreasonably low to the crew-- even if it was reading accurately.

 

Remind me to never fly with you.... On second thought, I'd remember.

DING DING DING - I was wondering how long it would take for you to crack and you to start making stuff up. I *thought* after I caught you editing the BEA report you'd settle down.. guess not.

You're right solo, the pilots did everything right it was that damn Airbus again....

Dude, don't you ever get tired of being a tool?

 

Dale if two Boeings collided on a taxiway (and both pilots fail drug tests), SOME PEOPLE would post very wordy posts explaining how the Airbus computers took over the planes and crashed them together. .

It really gets tiresome.

 

awesome reply.

Thank you sir.