markjmd

iTrader: (11)

Nothing wrong with the link, plenty wrong with your inability to understand anything it said.

SirFozzalot

Not sure if anyone has read this:

BEA Update on Investigation

Gives quite a detailed report of the initial findings, including a 3D view of the last 5 minutes of flight.

billythekid

Very odd... if you ask me.

The pilot went for a nap, Wx was poor, very poor. Various issues start shortly after.. A/C went into a nose up AOA after some big bumps. Thats normal, but the IAS went from 270kt to 60kt.. thats very rare. Air speed cant be trusted could have been a fault.

She then climbed, stalled out at 38 grand, then boogied all the way to the deck in a nose up stall with the throttles at idle. Nasty, nasty stuff. God knows what the artificial horizon said... sounds like the pilot pushed the stick forward, got the stall horn, pulled back, stall horn stopped.... all due to alternate law - where the stall warning wont sound when air speed is outside of flying envelope.... which could well be a big part of this crash.

I bet this issue will be flown on the simulator back in France quite a few times!

Leslie

I have flown the 747-400 simulator which I accept is an older aircraft, but I was surprised to find how well harmonised the flying controls were and with perfectly well balanced forces making it particularly easy to fly. Powered flying controls do resist feedback from the control forces and the artificial feel in the system is controlled by the Indicated Air Speed. If the IAS is varying by large amounts the artificial feel in the control system will change but should not cause difficulties.

I see no reason why the aircraft should not be perfectly controllable under those circumstances assuming that the flying controls were functioning normally. I have had to cope with quite violent clear air turbulence on the edge of a powerful jetstream over the States. The mach number changed from 0.96M to over Mach 1 faster than the eye could follow. Despite the fact that was well beyong the max allowable Mach number and it was necessary to use full nose up elevon to control the aircraft it was still just possible to retain control.

If there was airframe damage then I would expect that it would be very difficult to control the aircraft since full control movement may not have been available.

Even in heavy turbulence, with loss of indicated airspeed, and a standard power setting, and the aircraft attitude maintained withing reasonable limits in the turbulence, it should have been physically possible to retain control of the aircraft. It would be a pretty uncomfortable ride though.

Restricted control authority in a powerful storm like that is a different matter and is the sort of thing people wake up screaming in the night about!

It is perfectly possible to experience G forces well beyond the aircraft limitations while the aircraft is travelling through violent vertical air currents in an enormous cu nimbus cloud or close to a jetstream. It feels like there is a very large jack hammer hitting the aircraft, strong enough to physically cause you to catch your breath. It can feel as though the wings might part company with the fuselage at any moment and is a very good reason to avoid flying through such storms.

Les

Petem95

Very interesting to read your opinions here Les.

I don't know if the full cockpit audio has been released along with this information, but from the audio they have released there is nowhere that the pilots state they are having issues with the controls. They must have been fully aware of the rapid descent, as one of the co-pilots advises that the altitude is 10,000ft when they are at that altitude.

It just looks like they stall the plane at 38k ft, and then fail to correct this, and the plane just falls out the sky, hitting the water with a vertical speed of just over 120mph.

The final incident report released in a few months will be interesting to read, but surely pilot error is looking most likely at this moment in time?

Leslie

If that is how it went, those pilots should never been allowed near the controls of an aircraft. A First Officer should surely be up to a minimum level of ability in order to deal with any such occurrence.
It is possible however that the mach number increased rapidly due to a sudden large temperature change to a high value which meant that there was insufficient up elevator left to keep the nose at the proper attitude. This would cause the nose to drop as the centre of lift moved backwards and the speed to continue to increase which is an unstable condition until they reached a lower and more dense altitude with a lower mach number. This is what I described happened to us above. They might well have had full up elevator applied as the aircraft was descending but as the mach number decreased the requirement for a large application of up elevator decreases very rapidly and the resultant rapid nose up movement may well have overstressed the aircraft and damaged it severely thus leading to the eventual crash. This did in fact happen just like that once and the aircraft experienced some 10G nose up! Luckily it held together and they were able to land, but that aircraft was written off!

This situation could also lead to damage to the airframe and further loss of control if they were too slow with the necessary corrective control actions in the first place. They could have wound up at the lower height with virtual complete loss of flying controls left to them, or a very high nose up attitude and no way to correct it. So much room for speculation of course.


Les

markjmd

iTrader: (11)

The two co-pilots had 8,000 hours flying experience between them, from what I've read (compared with the captain's 10,000).

Leslie

I read that one had approx 3,900 hours and the other first officer had over 6000 hours. More than enough to be able to cope initially under the circumstances. They are pretty experienced with those totals.

I think they must have had a catastrophic failure due to the violent storm which meant they were unable to control the aircraft, more than just airspeed indicator loss.

The big mistake was to attempt to fly through an area with such dangerous conditions, they must have known of the possibility of those storms from the weather briefing before flight and the captain should not have left the cockpit at the time. Everyone who flies in that area knows of the enormous towering Cu Nimbus clouds which can go up to 50K feet which are to be found in the combining zone at those latitudes and how dangerous they are.

I find it hard to believe that simple airspeed indicator failure should lead to that crash with pilots of that experience at the controls unless there were further and more serious problems.

Les.

billythekid

I agree, I wonder if they had serious damage to the rear of the a/c - elevators perhaps.. they are tested on rigs to cope with serious loads but I just wonder if there was some kind of failure which meant they could not bring the nose back down.

Not good if you ask me.

Leslie

I certainly suspect some kind of catastrophic control failure because of the severe air turbulence.

Les

Funkii Munkii

Excellent posting and insights as usual Les

Thoroughly enjoy reading your aviation stories and knowledge

markjmd

iTrader: (11)

Seconded, far better than trying to pick through the speculation and waffle from who knows who on the news.

If I ever fly to Brazil though, I think I'll make a point of doing a 2 or 3 stopper with connecting flights somewhere in the US. Seems like the safer option, in the light of storm problems in the part of the Atlantic under discussion here.

Petem95

Storms or not, the most dangerous part of flying is take-off and landing, so technically doing a 2-3 stopper instead is going to involve more chance of an incident!

FlightMan

Daily flight Brazil to Heathrow. Whilst that area of the Atlantic is redoubtably challenging, it's by no means unflyable. Go direct.

Funkii Munkii

I did AF to GIG about 8 years ago, cheapest ticket I could get at the time, they were running the A340 on the route at that time.

Coincidentally the last flight I ever smoked on, they had a bar and smokers area at the tail of the aircraft.

Gauloises and Brandy ahoy

DaveD

iTrader: (2)

True, the highest percentage of recent accidents have been runway over-runs.

Regarding the AF flight, it is odd that the PIC keeps the nose up, despite the stall. However, in the reversionary mode the aircraft was in, it apparently removes any trim from the rear stabilisers, which the pilots would usually expect. Because the air speed data was duff, the stall warning was deactivated.
Also of note, is that the tail section was found reasonably complete, and apparently some distance from where the aircraft came down.

Leslie

Wonder if it parted company with the aircraft before it entered the sea.

Les

Jamie

Me too makes scobbynet worth looking at

tony de wonderful

I think they can tell from the blackbox data.

For a start I think a ripped off tail would cause all sorts of failures flagged 'cos of the electrical and hydraulic systems being parted.

Also would make the aircraft perform all sorts of crazy aerobatics and you would see that on the telemetry. As far as I have seen AF/Airbus say the aircraft descended reasonably 'level' ( a little nose up) i.e consistent with a stall not wild aerobatics.

Leslie

You would think so TDW, but why was it some distance from the rest of the aircraft?

Could be even more convoluted than we think.

I don't know the handling of that aircraft of course, but it is a general rule that as the mach no. increases with a swept wing aircraft, the centre of lift moves back which induces a nose down trim. With a violent increase of the mach number which could happen under those circumstances with large horizontal temperature changes, the mach number could increase beyond the limits of elevator control which of course is an unstable state since the nose down attitude would maintain the high mach number as the aircraft was descending until it got to a lower level where the mach number would then decrease. It would be normal of course for the pilot(s) to be applying full up elevator to try to raise the nose. As the aircraft reached the lower level the mach number would decrease rapidly and the full up elevator would then bite as the centre of lift moved forwards again. If the pilot was slow to reduce the up elevator the aircraft would be liable to very high G forces beyond the aircraft limits and the aircraft would pitch up to what could be an extreme amount. The forces involved could seriously damage the aircraft.

This happened to me once in the Vulcan once in the USA due to an extreme horizontal temperature change and we managed to control the aircraft with no further troubles, but it was a real frightener!

It happened to another Vulcan too and they were less fortunate in that it dived out of control to 20K feet from 40K and when the elevator control bit it climbed back up to a high level vertically! You had to be ready for the elevator to become fully effective again and if you were too slow to take the control deflection off rapidly the aircraft would pitch up violently. It was a strong machine and it held together for the landing but it never flew again. They reckoned it had experienced some 10G which is enormous and says a lot for the aircraft.

This is a possibility for the airliner of course-try WIKI on "Jet Upset". This is just another posiibility and could even have been a combination of several circumstances.

How much the black box would reveal is anyone's guess, and there may well also be commercial reasons for not admitting everything as one might expect.

Les

tony de wonderful

I think that is partially what I meant about controlling a big widebody with a manual stick.

I'm not a pilot of course just an interweb expert but don't they 'wallow' a lot. Add in factors such as turbulance, horizontal temperature changes etc, and black ocean/black sky, and of course faulty airspeed readings, and I imagine it would be very tough to fly an A330 manually in such conditions....that is to say maintain stable flight...impossible to recover from an upset?

fast bloke

Les - if I read this correctly, you say that extreme turbulence can cause the Mach reading to increase quickly. Closest I've been to a cockpit is row c, so the rest is guessing. If you got a Mach reading of 1+ would it be correct to 'nose up' so you could would lose some speed? When the crash happened, there were reports saying that endless warnings were being flagged.

So if they got a 1+ Mach reading and a stall warning at the same time, with incorrect info on airspeed, what would conventional training tell them to do?

fast bloke

Not sure how pitot tubes work either, but I have a theory on what happened. Not sure if I can explain it though.

The pitot tube measures airspeed by measuring how much air is passing through. If it ices up around the intake, it will be getting less air in, so it will report a lower airspeed. If it ices up inside, the intake will be normal, but the measuring device will be measuring air that has been squeezed through a tiny hole. Sort of like putting your finger on the end of a hose. You are getting the same amount of water, but it is travelling much faster.

So the pilots get a warning that they are over Mach1 caused by the turbulence. The pitot tubes report an artificially increased airspeed due to internal icing. Everything else is reporting something incorrect and the stall warning is sounding. You have to make a decision based on the readings, and the only two that are reporting the same thing are the Mach meter and the pitot tubes. You can fly based on the stall warnings and nose down, or fly based on the Mach meter and pitot tube readings, so nose up. In this case, there is no middle ground. Believe the wrong sensor and you are going to the bottom of the ocean. The sensors are giving you a 2-1 chance of surviving.... what do you do?

ALi-B

iTrader: (1)

Fire up the Tom Tom and see what that says

(although with the time it takes for mine to boot up and find the satelites, I'd have crashed into the ocean long before it gave a GPS fix on my speed )

Seriously though, it does make me wonder. Blocked pitot tubes (or static ports) have been blamed on a number of aircraft crashes over the years, (from memory: one blocked, another covered during cleaning).

DaveD

iTrader: (2)

Les, the BEA report will be completely impartial, and will not skirt information due to any commercial issues.
Fast Bloke - the pitot probe is basically a tube with two pressure sensors - one orientated head-on into the mouth of the tube, the other mounted on the side wall. These measure total (dynamic) and static pressure respectively. With increasing air speed, the total pressure reads relatively higher than the static (dead air) pressure. The pilot probe has heaters in it to prevent it icing up, but in very humid conditions it could be possible for the mouth of the tube to ice over. If that happens, the air in the tube will become static, as if the aircraft were stood still.

tony de wonderful

Can't have both.

billythekid

Dont forget - whilst this was happening the a/c was sending fault warnings back to Airbus, which is giving an indication of physical damage..

Leslie

Yes I would not expect it to be anything different, but I wonder how much of the true story will be made public. That is what I actually meant.

The air does not flow through the pitot tube as Fast Bloke said, it enters it and comes to a dead stop which increases its pressure. That pressure increase is an indication of the airspeed and is fed to the airspeed indicator in the cockpit. As you go higher, the air becomes less dense and the ambient air pressure decreases. This will cause the pressure in the pitot tube to decrease and this shows as a lesser airspeed in the cockpit. This is called "indicated airspeed-IAS" The actual speed of the air as you go higher is greater than the indicated airspeed. The actual speed is known as the "true airspeed-TAS" At 40K feet the TAS is roughly double the IAS or more. The stalling speed would relate to the IAS. The Mach Number will relate to the TAS. The IAS is also known as the pressure airspeed.

The static pressure is also taken at the pitot head and is fed to the static pressure instrumentation such as the altimeter.

If the pitot head becomes iced over, which it should not providing the electrical pitot pitot heater is on as it should always be, then the IAS will remain at the last reading and of course will not alter with change of aircraft speed. The altimeter will also remain at the last reading before the icing problem and will not alter with change of height.

For TDW, I presume by manual control you are referring to the control column. That will not be connected directly to the control surfaces mechanically,but to a powered flight control unit (PFCU) which moves the control for you as you demand via the control column. There is no feedback from the control surface but they artificial feel will provide resistance to moving the control column according to the IAS of the aircraft. This gives a sense of the flying speed to the pilot and also helps to prevent overcontrolling. There will be one PFCU for each flying control surface, and probably doubled up for the rudder in case of single unit failure. For a fly by wire aircraft, the PFCU's will be computer operated as in the Airbus or the Tornado.

The controls of the aircraft will be easy to use and well harmonised and even in severe turbulence it should be possible to fly the aircraft by maintaining a flight attitude on the attitude direction indicator which is gyro operated. Even with loss of airspeed indications the normal procedure is to set a standard power setting and maintain the flight attitude. The aircraft speed should automatically settle to an expected value.

The aircraft will of course possess a fair degree of inertia and the pilots should be familiar with that and that should not mess up the ability to control the machine. It might even have a bit of a "wallow" under certain conditions but that should not lead to any loss of control.

Having said all the above, with pilots who each had a high level of experience as all three of them had, I consider that the loss of control of the aircraft was due to more than the possible icing of the Pitot head. Yes it would not be pleasant to fly under conditions that you would find in such a powerful storm such as they doubtless experienced, but I feel that the loss of control was due to more than the reasons which have been mentioned so far.

Flying an aircraft by use of the control column is the natural way to do it TDW, and an autopilot would rarely be used in preference to that in such adverse conditions especially with large variations in temperature and pressure which the autopilot relies on. Autopilots are liable to disconnect themselves anyway under such conditions.

Les

fast bloke

Cheers Les - makes a lot more sense now.... The thing is... it sounds like they made the totally wrong decision. Any other suggestion what incorrect readings might encourage them to do that?

Leslie

Yes I agree about wrong decisions.

I was amazed that they should have continued on track into those vicious storms which were on their planned track. Not only that but the dispatchers should have taken that into account and routed them around them in the first place. That was the first major error. Everyone who flies in that combining zone knows about the likelihood of towering cu nimbus clouds and on that day the situation was very much worse than usual. We used to operate through that area regularly on the way to Ascension Island and back and always carried enough fuel for possible diversion.

The ensuing problems they encountered after hitting the dangeroius turbulence and heavy weather would have been exacerbated had they lost flight information such as loss of the ASI etc. The altimeter would not work without accurate static pressure fed to it either. Not very helpful if you lose altitude information and also the vertical speed indicator which also relies on static pressure information. None of that would be available if the pitot head was frozen solid.You need all that information to be able to fly level. The attitude direction indicator working on its gyro would at least help to maintain a sensible flight attitude but you still need the altitude information to fly an accurate height. The turn and slip indications are also gyro generated. Those are the indications you need for blind flying if you cannot see the horizon. Loss of any of those make things that much more difficult.

Couple all that with the extra difficulty in flying in very turbulent air and although not impossible, it would not have been easy to cope with the aircraft. If there was another factor such as serious damage to the flying controls, then I would not give much for anyone's chances. No one could predict the flight path except that it would be pretty certain to be downwards.

The big mistake was the first one-ie flying into the big storms. If the aircraft remained intact in its instrumentation and its airframe and flying controls then it would have been perfectly possible to fly it safely. A combination of the above problems would be a very different story.

Les