Air Asia Indonesia Lost Contact from Surabaya to Singapore [Archive] - Page 11

Ian W

25th Jan 2015, 16:23

I'm assuming you're not a pilot? Multi-engine aircraft are designed / certified such that above Vmca speed (which is lower than takeoff speed), there will be sufficient control authority even with the most critical engine out and the other engine set to full thrust. So at M0.67 losing an engine (by itself) should not cause a fatal accident.

Now, at high-angle nose up, if the airspeed decays rapidly while conversely the AoA increases rapidly (resulting in a stall), then a one-engine-out situation at that point becomes deadly.

Yes I am aware of the certification of twins and that many of them will reduce maximum available power on a good engine in asymmetric etc.,

But you then go on to agree with what I proposed.

As you are so experienced with A-320 you will know what the minimum safe operating speed is at FL360+. Give the aircraft being 15deg nose up and M0.67/ 205KIAS and autopilot out in AoA law (assuming similar behavior to the quoted report) as you say going asymmetric with a compressor stall one side and full thrust the other could be deadly. It's not only drag - there will be pitch up forces from the live engine and pitch down from the stalled engine. You could also consider what the 'safe control authority' is up in 'coffin corner' at greatly reduced airspeed and what happens if someone is too hasty applying 'full' control authority.

NigelOnDraft

25th Jan 2015, 16:25

TAS = IAS x sqrt( ISA Density / actual density)

TAS will remain largely over short term (KE) if W/V is unaltered.

Alter OAT alters density (Boyle IIRC?) so requires altered IAS.

Overspeed limits are in IAS terms, so if IAS changes due delta Temp, you can go into an "overspeed".

At altitude, you will be M limited, but again this is Temp dependant.

Clear as mud ;)

Ian W

25th Jan 2015, 16:27

Could somebody explain to a simple engineer how an air temperature "spike" would cause the aircraft to go into an overspeed protection? I've seen this theory posited to explain an automatic pull-up, but I can't see why the aircraft would, either correctly or erroneously, "think" it was overspeeding.

It is important in many aircraft to avoid stresses due to shockwaves forming on the aircraft surfaces due to approaching the speed of sound. The speed of sound is dependent on the air pressure _and_ the temperature see Mach Number (http://www.engineeringtoolbox.com/mach-number-d_581.html)

So a sudden change up in temperature can convince the automatics that the aircraft is too close to the speed of sound and damaging shockwaves may be forming on the airframe and control surfaces. The automatics then do their best to recover back to below the limiting Mach No. The quickest way to do that is to pull up.

Just as an addition- going too fast and getting shockwaves build on various surfaces can lead to other interesting effects such as control reversals which could lead to dangerous effects.

FlyerBabe

25th Jan 2015, 16:35

@md80fanatic there is a YouTube video shown from the perspective of on the boat. Just before the cable snaps the wreckage is high enough where the letter "C" is visible. This would make the visible portion of what was brought up as the forward fuselage and part of the right wing. No windows were visible also no idea as to the portion below the water. They will be using a crane in their third attempt.

HarryMann

25th Jan 2015, 16:46

Isn't a, the speed of sound pretty well proportional to the Sq. Root of Abs. Temp?
If so aircraft Mach No. would decrease if Temp suddenly increased?

Have I got my knickers in a twist or am I correct :)

HeavyMetallist

25th Jan 2015, 16:56

Exactly what I was thinking - an increase in OAT increases the speed of sound, so the aircraft would be even further away from a Mach limit than before. And IAS is dependent on pressure, not temperature, so there'd be no effect on an IAS limit. I'm struggling to see why the aircraft would see any increased airspeed to protect itself from.

Leightman 957

25th Jan 2015, 17:37

RF4 at 2502: it has been stated many times that his increase in FlL request was denied. It's highly unlikely that he would initiate a unauthorized climb to move up to optimum .

And Lost in S at 2515: He had been cleared to FL340 but there was never a response to that clearance. That tells me they never heard it. It also tells me they did not respond because they were already too busy dealing with the problem that caused the aircraft to climb unintentionally.

It is wrong to simply assume that 8501 did not receive the climb authorization, based on lack of acknowledgment. The clearance may have been received, and pilots therefore may have known that a climb was then allowed, but THEN got busy and were not able to acknowledge. The time bits are small and important.

Australopithecus

25th Jan 2015, 17:48

...can be the case when a third party overtransmits on the frequency or breaks in, preventing a timely response. Which happens quite a lot in that area due to frequency congestion and poor radio manners.

A0283

25th Jan 2015, 18:00

Very fuzzy video again ... in some recent (not only aero) investigations images were deliberately fuzzed ... wonder if the same is true here.

If they are using the Crest Onyx again. Crest has a beam (width) of 14m.

You might consider that in your analysis.

BG47

25th Jan 2015, 18:10

I have not seen any discussion about the cargo possibly breaking loose due to extreme turbulence (once they entered the level 4 or 5 Thunderstorm) moving the cargo tail aft position causing a nose high attitude = increased rate of climb.

FlyerBabe

25th Jan 2015, 18:12

@A0283 the YouTube video from the second attempt to raise the fuselage where you can see the letter "C" is from the Crest Onyx. I can't post the link but you can find it by Googling...AIR ASIA JATUH LAGI ! I also know of someone working on the recovery on the Crest.

Leightman 957

25th Jan 2015, 18:25

>@md80fanatic there is a YouTube video shown from the perspective of on the boat.

Video link please. I'm not finding it. I can't imagine there was any real confidence that given previous video of the fuselage that it could be raised intact as there was no apparent remaining structural integrity.

Also, all the short fuselage video I have seen is the same shot, showing the right wing generally in place, but the left wing is not merely by low visibility with a distended fuselage section where the wing could be expected. Is there any other video or photos that clearly show that the left wing is absent?

Given the buoyancy of the vertical stab, the tail could clearly have drifted some distance given the current strength. I have assumed for the present that the fdr/cvr would have been found close to the impact site if they were torn out on initial impact. I too am puzzled that the recorders were found "under the wing. I realize "the wing" could have been in initially misidentified, but the number of divers during the late fdr search and in proximity to the recovery of the fdr and cvr makes it less likely that the HStab would be mistaken for a wing. We had an initial plot of the debris field but I have not seen an update. The absence so far of the relative positions of the fdr "wing", center fuselage section, cockpit and any other major components is frustrating given the time since their discoveries. Any links to debris location plots would be appreciated.

RE BG47, has a cargo manifest been published?

ZeBedie

25th Jan 2015, 18:48

Spotted some slats there, so more likely the wing-fuse centre section.

For sure, yes.

md80fanatic

25th Jan 2015, 19:02

I can't imagine there was any real confidence that given previous video of the fuselage that it could be raised intact as there was no apparent remaining structural integrity.

There may have been, the first time around. That right wing looks nothing like the original video, which appeared to show the top surface mostly intact. The latest video shows nothing but internal structure. I wonder if one of the previous "recovery" attempts included rolling that wing over the ship's stern?

http://www.bbc.com/news/world-asia-30972235

clipped from above source ..

"Rear Admiral Widodo's comments came after a renewed attempt to raise the wreckage from seabed failed when it kept breaking into pieces."
"The wreckage includes parts of both of the wings of the Airbus."

From Wall Street Journal ...

http://www.wsj.com/articles/airasia-flight-8501-more-bodies-recovered-as-efforts-to-float-fuselage-continue-1422087284

On this page one can get a good idea of the challenging diving conditions, including swells of less that 1 foot and water clarity that really cannot be expressed as being similar to whale poop. Great video on that page as well.

Stabilo31

25th Jan 2015, 19:14

Definitely it's main fuselage and wings. In the view of the damage to the structural fuel tanks I suspect an "almost" horizontal splashdown similar to AF447 at around 10000ft/mn.

WanganuiLad

25th Jan 2015, 19:41

They seem intent on ripping it all to bits dragging it over the back of the ship. Dear Lord have they no cranes down there?

jcjeant

25th Jan 2015, 20:18

Indeed .. recovery operation turn into a farce !
Real shame and what must think the families of the victims seeing that ?

enola-gay

25th Jan 2015, 20:20

The vessel being used is an offshore oil industry anchor handling tug, which normally hauls or deploys a cast steel anchor over the stern with steel hawsers or chains as part of a semi-sub drilling rig deployment.

It is not a suitable vessel for an aircraft fuselage recovery. The outcome was predictable.

Coagie

25th Jan 2015, 20:29

Are they trying to tow the fuselage to shallower water maybe? It seems kind of large to put it on the deck. At first glance, I thought it was the HS, so it might have had enough room on the boat, but as a wing and fuselage, it looks as though they would be too unwieldy. Maybe they can tie them down, fold, or stack them, where they don't interfere with the boat's seaworthiness.

Leightman 957

25th Jan 2015, 20:42

Different pic of wings and fuselage top I've not seen before:
https://www.dropbox.com/s/q67265sv48qc4zt/wings%20as%20in%202.png?dl=0

Machinbird

25th Jan 2015, 21:15

A photo grab which I hope is appropriately sized. This is at the instant the lifting line parted.
http://home.comcast.net/~shademaker/CenterSection.png
We are seeing the center section (from the bottom side) with the slats still retracted and the skin immediately behind the slats missing/fragmented.

The wing&fuselage is canted to the left in this view which might imply more right wing under the water or alternatively, more air bags on the LH wing.

As for the choice of vessel, probably the best that they could lay hands on in the time available. Now that the black boxes have been recovered, their priority is on recovering the dead, not so much the wreckage.

Upon being given a priority grab list by the accident investigators they will work that into their plans, but you can anticipate that salt water has intruded almost everywhere. For example, on the THS position, all they will see is the final position of the THS, which can confirm the DFDR calibration of that parameter, but the actual flight history data is going to be on the DFDR.

I think that when all is said and done, this will be an aircraft flight law induced departure, possibly also related to a hidden failure which suddenly became unmasked.

One question for the Airbus pilots here. If you really really needed to, how would you get to Direct Law. Drop the gear in alternate law? But then suppose the aircraft had shifted into Abnormal Alternate Law due to an unusual attitude? Airbus Flight Control Laws (http://www.airbusdriver.net/airbus_fltlaws.htm)

WanganuiLad

25th Jan 2015, 21:19

"Well,Airplane-salvage ship may not be the most economical... I guess these Big "tugs" -designed for oil platform work are the best available for now. I guess everyone there are doing all they can,with what they have."

indeed... all credit to the effort in dangerous waters

henry_crun

25th Jan 2015, 21:36

That is most definitely not the way to raise an aircraft which contains human remains and vital evidence. Every effort should be made to enclose the parts within a load bearing frame which is lifted intact by a sea-going crane.

I have watched, in calm water, a sunken boat completely wrecked by inappropriate recovery attempts. This is simply a repeat on a larger scale.

Plumb Bob

25th Jan 2015, 21:41

@ Leightman 957, the link to the video filmed from the ship is http://youtu.be/qlk-eJoRLnc

Note at t=0m50s the white-on-red letter C of “Now Everyone Can Fly”, which on the RH side should be above a window above the leading edge at the RH wing root.

The aircraft concerned indeed had the large-size fuselage-long lettering version of that text, which is evident on the image http://si.wsj.net/public/resources/images/BN-GJ622_AirAsi_G_20150111220801.jpg of the retrieved short part of the fuselage sidewall ahead of the aft RH service door. On that image the large-size letter N of “Now [etc.]” is visible when it is being lifted from the ship onto the blue open trailer for transport from dockside to investigation site.

@ Machinbird: with the aid of the video taken from the ship, your image appears to be the wing center section top skin inside, as seen from the bottom up. The view of curve of the slat extenders suggests likewise to me.

mseyfang

25th Jan 2015, 21:43

We are seeing the center section (from the top side I believe) with the slats still retracted and the skin immediately behind the slats missing (probably removed by some hydraulic effect of water entry). The fuselage section seems to be missing the crown, and the break at the forward fuselage seems to angle up and aft into the center section.

I'm no expert on the A320, but that could also be the rest of the tail section with those "wings" actually being the horizontal stabilizers. Also looks like it rather fits that part of the empennage already recovered. I hope I'm right because if I'm not, they've badly mangled what from prior pictures looked to be a fairly intact section of the center fuselage extending forward and aft of each wing and at least a portion of each wing. Without scale, though, it's hard to tell with any certainty.

FlyerBabe

25th Jan 2015, 21:48

While I agree that it's not the best way to do things. I do feel badly for everyone involved. I am familiar with someone on the Onyx. He is very eager to be part of the team, and the whole team just wants to do the right thing. They are working long, hard hours and dealing with things I wouldn't even want to see in a horror movie.

enola-gay

25th Jan 2015, 21:53

I am sorry but that did not look like a professional salvage operation to me. There was no vertical lifting force on the fuselage once it met the stern of the anchor handling tug at a right angle, and the fuselage was not a flexible sausage nor a solid anchor which would roll over the stern

Instead the tension on the hauling ropes increased to breaking point at the critical position and they broke, consigning the fuselage to a second crash to the sea floor, probably distributing the victims over a wider area.

This act has made the operation far more difficult. They needed a floating crane of which there are many in SE Asia.

PT6Driver

25th Jan 2015, 22:04

In previous statements I believe the stated goal was the recovery of bodies. The divers reported that entry into the fuselage was too difficult hence the attempts to raise the fuselage to the surface so as to be able to search for bodies in a safer environment.
In terms of forensic analysis of the wreckage very little will be accomplished because of the destruction we have seen to the tail section and now the centre section during and after recovery. Some deliberate some accidental.
Sadly I suspect that CVR, FDR and whatever chips are recovered from other systems will be the only credible evidence left.
Whether this is a deliberate policy or just a by product of the "recover bodies as the highest priority" remains to be seen.
This is not intended as a criticism of the divers themselves who are working in a very challenging environment, rather that of the policy makers.

Leightman 957

25th Jan 2015, 22:50

I too think the video grab of the attempted hoist shows the underside of the wings. The pic I posted earlier appears to me to show the upper wing surfaces nearly intact, really quite amazing, while the larger stills of the right wing, lettering intact, really show only the tip broken off.

The largely intact wings really present a puzzle. When we think of wings being pulled off by aerodynamic loads, and how relatively easy that is, it is easy to wonder what kind of impact on water with a large vertical component might allow wings to stay attached. I'd welcome any conjecture about how that might occur.

Short of a small drydock that might have been lowered and the fuselage section raised just enough to load it, then being towed a long ways at slow speed to shallow, calm waters for eventual raising on the dd, I see no reasonable way the fuselage could have been raised. Even lifting under water would have caused some damage. I do hope that many more underwater videos and photos were taken in anticipation that recovery would cause a lot of damage, and that some of those are released. They may not be released at least initially due to implications for the passengers survivors, but a final report in this instance would have to contain quite a few. The still pic and video grabs of the fuselage to me suggests the fuselage as it sat on the ocean floor had lost some internal cabin height from the fuselage top indentation, and probably more height from the floor pressed up from below. That could have resulted in very little vertical cabin room, mostly taken up by a jumble of seats which both pics suggest largely remained inside except at fuselage breaks. Divers reports of wiring and piping would apply inside. Not surprising divers could not enter.

No further word on cockpit location or confirmation of that of other major parts?

bob1vt

25th Jan 2015, 23:37

One question for the Airbus pilots here. If you really really needed to, how would you get to Direct Law. Drop the gear in alternate law? But then suppose the aircraft had shifted into Abnormal Alternate Law due to an unusual attitude?

Machinbird, an excellent question! Although I am not an Airbus pilot, I am a CFI with 40+ years of experience and have a BS/MS EE with 50 years of experience programming and using computers. More importantly, I have a good friend who is a current and qualified A320 first officer.

A previous post really put this question in perspective. It stated that the computers fly the Airbus, the pilot merely gives suggestions! It is only when sensor or other hardware/software failures make it obvious to the computer that it doesn't have a clue, that the Airbus goes to Direct Law. There is no button that the pilot can push, that activates Direct Law. According to my friend (who works for a major US air carrier), he has never been trained how to take over the control of the aircraft. He believes that there is a combination of circuit breakers that can be pulled to shut down enough computers to force Direct Law. He also thinks there is a document in the cockpit that describes how to do this, and given a few minutes, he could make it happen. I (along with the rest of you?) firmly believe that the pilots of the A-320 had only seconds to react.

Leightman 957

26th Jan 2015, 02:01

Thanks for the video link Plumb Bob and FlyerBabe. There's not much there but its all we have. Slats appear retracted but they are there. I'm beginning to wonder what the point is in hauling the wreckage aboard. If they do get it aboard the interior will have been run through a mixer, its contents may largely be evulsed, interior having become exterior in the process, and a lot of lighter debris will escape like styro peanuts in a 100' deep wind with limited visibility, vastly spreading the debris field and virtually guaranteeing a lot of debris loss. Without getting more detailed, integrity of bodies will be almost completely lost so recovery of bodies does not seem like a good reason. Is there some other reason to hurry? Compared to raising it would appear to remain more stable and intact on the bottom.

Msunduzi

26th Jan 2015, 04:11

Machinbird

26th Jan 2015, 05:10

Assuming that the assumptions made about how the upset came about are near correct, and that there was no mechanical fault (not meaning sensors) or airframe damage.......

If a properly experienced pilot was hand flying in direct law, at the time of the problem, alert and hooked up, would he stand a better chance of maintaining or re-gaining control than the electronics (purposely broad) ? (including taking over when they kick out) The hazard in Alternate law, IMO, is that it cannot anticipate your trim needs properly in certain kinds of upset scenarios, particularly those involving high angle pitchups and rapidly changing airspeeds.

I used to fly an aircraft that entered an unrecoverable flat spin mode a relatively high percentage of the time if spun. Yet the job entailed flying at the limits of aircraft performance. When very nose high and running out of airspeed, the formula for spin avoidance was to neutralize the controls and fly zero g until the aircraft was flying again. This even worked when the aircraft backed down before reversing direction.

When flying with a Trimmable Horizontal Stabilizer (THS) equipped aircraft, holding your flight controls neutral is not sufficient if your THS is trimmed for a low airspeed. It needs to be trimmed for cruise range airspeeds or it may cause a secondary stall.

Your airliner can swap ends on you in as little as 3 seconds when very nose high (See the Interflug-Moscow loss of control reconstruction and count seconds). You need to accelerate through the point where your flight controls are very loose to the point where they become solid and the aircraft stabilizes before attempting recovery. You do not need the aircraft helping you in the wrong way.:eek:

Until we see the QZ8501 DFDR readouts we will not know if the aircraft systems interfered with the piloting task, but it is a distinct possibility.
We already know that a bedlam of noise in the cockpit must have made communication difficult.

FlyerBabe

26th Jan 2015, 05:49

Looks like the weather abruptly changed so attempt to raise the fuselage has been put on hold. Don't know if it will resume today.

Gretchenfrage

26th Jan 2015, 06:20

to bob1vt

He believes that there is a combination of circuit breakers that can be pulled to shut down enough computers to force Direct Law. He also thinks there is a document in the cockpit that describes how to do this, and given a few minutes, he could make it happen

It is no circuit breaker, but you need to push three of five buttons on the overhead panel for the Flight Control Computers (2 of 3 prims and one of 2 secs).
As you rightfully point out: A task not suitable for sudden extreme emergency situations.
Boeings FBW can be overridden (some sort of direct law) by applying force on the yoke. The Primary Flight Computer will then be bypassed and the signal directly sent to the Actuator Computers.
This is well suited for such situations and is one of the main criticism i voiced, mainly by asking for a "red" or "hot" button on the Airbus-stick to get to Direct Law in adequate time.

to Machinbird

If a properly experienced pilot was hand flying in direct law, at the time of the problem, alert and hooked up, would he stand a better chance of maintaining or re-gaining control than the electronics (purposely broad) ?

Most certainly. Today most training departments and the manufacturers suggest in extreme upsets to initially steer between 0 and 2 degrees nose up and some 80% of N1.
This is however only possible IF the system allows the pilot to do so.
The main difference between the two philosophies is that Airbus will only do give you this full authority (when the protections feel otherwise) when in Direct Law. This state is however rarely automatically achieved when in extreme upsets and to achieve it: See above.

humbleppl

26th Jan 2015, 07:04

Last week I expressed my humble opinion about the "inability" of professional pilots flying advanced aircraft (Bus/ Boeing) to recover from a stall or spin... And I did not mean you were to silly to do it!!! I still have the greatest respect for you.
I got serious beating by a few guys here and; accused of having no idea what I am talking about and the mods taking my posts offline... It is correct, I had no idea that the advanced a/c can basically not be recovered; and that surprised and worried me a lot. Now, a few days later, the same professionals debate about how this should be possible, should be trained, etc. You even add another perspective, that an airbus can't be hand flown in an instant (when the pilot decides to), but requires quite extensive elaboration and switches..., something impossible in an emergency situation.
Previous "professional" statements like "better don't get into such situation" or "I know how to avoid a stall" or "the aircraft can't stall" are all very obviously incorrect. Both AF and Air Asia crashes involved a stall...
"Happy" at least that my "PPL view" of last week is now shared by a lot of professional drivers as well...

LiveryMan

26th Jan 2015, 07:18

A friend of mine suggested this is might be a partial repeat of the XL888T crash at Perpignan. Not the flight testing part obviously, but that part where water gets into the angle of attack sensors, freezes them in place, so computers think the plane is descending and command an ever increasing pitch up untill it stalled out and plummeted toward the water.

Not being technically minded, I thought I'd throw it out there to you wolves to tear apart.

onetrack

26th Jan 2015, 08:10

I see no reasonable way the fuselage could have been raisedIt could have been easily done with a large crawler crane mounted on a large barge, and using a "spreader bar" with multiple slings.
I feel the Indonesians are seriously lacking in technical skills and knowledge in this area, or they have simply not canvassed the available options - by way of people with appropriate knowledge, and by way of local companies who possess such equipment - for effective recovery of the likes of the fuselage and wings.

The attempts to drag aircraft components bodily from the water using winches, a la fishing nets, is doing nothing to promote their desired image of competence in recovery techniques.

BlankBox

26th Jan 2015, 08:53

http://img4.hostingpics.net/pics/345488363574620.jpg

Gigi Cocok 100 Persen, Dua Korban AirAsia Dikenali | -nasional- | Tempo.co (http://www.tempo.co/read/news/2015/01/25/058637534/Gigi-Cocok-100-Persen-Dua-Korban-AirAsia-Dikenali)

...sorry not in English...but you can decipher quite a bit via the numbers...

...and regarding the rescue effort...the Indonesian SAR...are in no way shape or form trained professionals...nowhere near western standards are used...and likely for most involved here...this is their first kick at a major accident and they're learning on the go - it is what it is...welcome to Asia and the way things get done.

Blake777

26th Jan 2015, 09:25

I take it the artistry is exaggerated as from the condition of the radome the aircraft either did not hit the water nose first - or for some reason it separated before impact.

HarryMann

26th Jan 2015, 09:39

Leightman957 said "
The largely intact wings really present a puzzle. When we think of wings being pulled off by aerodynamic loads, and how relatively easy that is... "

What makes you say that... how many instances are there of that from
aerodynamic loads alone these days?

Mahatma Kote

26th Jan 2015, 09:55

The largely intact wings really present a puzzle. When we think of wings being pulled off by aerodynamic loads, and how relatively easy that is... "

The wing to wing spar /body joint is the strongest part of the aircraft (often called the wing box). It has to lift the entire weight of the aircraft plus maneuvering forces of up to 6 times that weight.

It will be the last thing to break. You will see the fuselage fall off and the wing ends fall off before any break in that part.

bud leon

26th Jan 2015, 10:52

It could have been easily done with a large crawler crane mounted on a large barge, and using a "spreader bar" with multiple slings.
I feel the Indonesians are seriously lacking in technical skills and knowledge in this area, or they have simply not canvassed the available options - by way of people with appropriate knowledge, and by way of local companies who possess such equipment - for effective recovery of the likes of the fuselage and wings.

The attempts to drag aircraft components bodily from the water using winches, a la fishing nets, is doing nothing to promote their desired image of competence in recovery techniques.

Indonesia has plenty of capacity in this area. It has a sizeable offshore oil and gas industry and salvage operators would not be ignorant to salvage techniques. It's much more likely to be an issue of access to resources. Do people think the kind of specialised resources suitable to this type of offshore salvage are just waiting around in close proximity on call to respond to this kind of event? it can take weeks to mobilise specialist resources,which more than likely will already be tied up in current and planned commercial operations.

There has to be a weighing up of costs and benefits and availability of resources. And maintaining current morphology is not actually essential for incident investigation.

Most of you are living in some kind of fantasy land. Sad, traumatic and important as this is, it's not something the world stops for.

skippybangkok

26th Jan 2015, 11:17

Kaki = feet
Menit = minute

Basically started at 32k kaki. Started climbing 400, the 1400, until hit 11000 fpm. Max height 37k something feet.

Then fell

All based on radar observation I believe

henry_crun

26th Jan 2015, 12:38

...spreader bar...

yes, exactly, that was omitted in the example i mentioned and presumably missed here

anyone who doesn't recognise the term should not be involved

Global_Global

26th Jan 2015, 12:43

Indonesia has plenty of capacity in this area. Mmm we are talking Indonesia here.. All other countries in the region I would agree but this is the country with the biggest `we know best` chip on the shoulders... In other words politics and ego´s over best practice is not uncommon. I have not been on the site and know NOTHING about salvage so I refrain from commenting on this case in particular but having worked in the region a lot I was not surprised with any screw up like this :uhoh:

RF4

26th Jan 2015, 12:45

I am a little confused by Blankbox's post ( the graphic therein). As far as I knew 8501 had requested a weather diversion to port which was initiated some time before experiencing the zoom climb. As far as I knew, the attitude of the A/C was still in left bank ( or some facsimile) until reaching apogee. Am I wrong ? The graphic shows a turn to starboard, before the zoom and that is news to me. Anyone ? .

Heathrow Harry

26th Jan 2015, 13:18

Global

The Indonesians have a large offshore oil & gas sector that works subsea all the time - they probably know as much about it as anyone else and the kit they've deployed looks better than most countries could come up with

ramble on

26th Jan 2015, 13:52

The pictures of the radome show some marks & evidence that lightening strike & radome separation shouldn't be discounted.....if it had indeed been blown off by such a strike it would have had an interesting disruptive affect too on the sensors that measure AOA & pitot static pressures ...

Lightening in this region is some of the strongest I have ever seen.

training wheels

26th Jan 2015, 14:19

Here's a video of divers getting right up close to the fuselage. You can get an appreciation of how difficult it is for the recovery team to enter the fuselage and search for bodies given the poor visibility and limited sunlight reaching down to that depth. On top of that, you have a tangled web of wreckage, metal fragments, cables and wires to deal with. The footage also shows what appears to be the top of the left wing (with spoiler still deflected up) where the over wing emergency exits are.

9nG9qnIGXbw

phiggsbroadband

26th Jan 2015, 14:26

The Radome also has a split at the very front, which could have let air pressure build up inside the dome... just a few psi over pressure could cause it to pop its rivets.

SAMPUBLIUS

26th Jan 2015, 14:45

LEIGHTMAN TODAY @ 15:50 SAID " When we think of wings being pulled off by aerodynamic loads, and how relatively easy that is,'

Relatively easy ?? Suggest you review your comment- its not easy with a maintained airplane. As to surviving water impact, the load direction with forward motion is mostly fore to aft- sort of like a water ski. The engines come off probably at the pylons which leaves the inboard mostly intact as shown.

LiveryMan

26th Jan 2015, 14:51

RE video in #2575 (http://www.pprune.org/8840697-post2575.html)

The wing spoilers appear to be deployed. Would deployed spoilers remain up after the aircraft coming apart like this? Or are they just floating freely, as their is no pressure in the hydraulic lines?

sandiego89

26th Jan 2015, 14:58

I would also image that water impact, regardless of entry angle or velocity would cause damage to the radome, including cracking, paint being delaminated due to "denting", separation of the radome etc.

Are the marks really indicative of a lightenting strike? Could they just be the paint being craked where there was an impact? Looks much like a car bumper that has been impacted and the paint becomes cracked/crazed due to deformation of the underlaying structure...

Niner Lima Charlie

26th Jan 2015, 15:12

The radome is held on with screws, not rivets. This allows maintenance access to the weather radar equipment (dual R/T, antenna drive, flat plate, etc.).

Volume

26th Jan 2015, 15:16

The wing spoilers appear to be deployed. Would deployed spoilers remain up after the aircraft coming apart like this?If there is a strong current from the rear end, possibly. Due to gravity it should be closed by now. Even with hydraulics fully intact, after some weeks external forces push it where they want to. Anyway remarkable that the spoilers are intact.

Stabilo31

26th Jan 2015, 15:44

FYI radome has a hinge on his upper side and is maintained closed by two latches (approx. 4 and 8 o'clock).

aguadalte

26th Jan 2015, 17:02

despegue:By the way...if in doubt: all automatics OFF, go to DIRECT LAW, fly the aircraft raw data like any other aircraft in the world .

There is no need to put the aircraft in DIRECT LAW. One just have to take TWO ADRs OUT and put it in ALTERNATE LAW.

In ALTERNATE LAW the aircraft will obey your inputs and the pilot will be able to steer from any unusual attitude, keeping the required information to fly the aircraft away from overspeed protection, as well as low speed. The protections in ALT LAW may be overridden while maintaining structural integrity. (Load Factor).

Stabilo31

26th Jan 2015, 18:08

Previously limited by physically hardening the pitch axis on the yoke. With FBW it's just implemented in computers (ELAC's, PRIM's,...)

SAMPUBLIUS

26th Jan 2015, 19:46

" The protections in ALT LAW may be overridden while maintaining structural integrity. (Load Factor)"

So in other words one cannot ' bend ' the airplane no matter what ( unless somehow HAL allows direct control "

IMO that is a logic flaw. I posted earlier about a China 747 in 1985 that had an engine failure, went into a major spiral dive, and was recovered- and still flew - but it was severly bent with estimated 5G loads during pullout. Wings were bent up permanently sever inches at tips, who knows how much during flight. Part of horiz stabilizer was torn away.. plane landed safely, and was repaired and put back into service. Structural limit load test the wings on a 747 go several feet above top of fuselage at 2.5 to 3 G equivalent ..

Alternate law limits would have prevented that- and plane and passengers would have been swimming with the fish . .

see China Airlines Flight 006 - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/China_Airlines_Flight_006)

and the official report can be found

" here were two serious injuries on board: a fracture and laceration of a foot, and an acute back strain requiring two days of hospitalization. The aircraft was significantly damaged by the aerodynamic forces. The wings were permanently bent upwards by 2 inches (50 mm), the inboard main landing gear lost two actuator doors, and the two inboard main gear struts were left dangling.[1] Most affected was the tail, where large outer parts of both horizontal stabilizers had been ripped off. The entire left outboard elevator had been lost along with its actuator, which had been powered by the hydraulic system that ruptured and drained."

Now about load protection ??

http://www.airdisaster.com/reports/ntsb/AAR86-03.pdf

http://www.airdisaster.com/reports/ntsb/AAR86-03.pdf

BTW early in the 707 commercial introduction a 707 went into a spiral dive cuz autopilot kicked out at night, nobody minding the store. Luckily they recovered, but wings were permanentely bent..

Load protection would have done What ??

Lonewolf_50

26th Jan 2015, 19:57

... nobody minding the store ... It appears that this can't be fixed with technology. :cool:
Back to our thread, it's a good bet that in this accident, Air Asia Indonesia, the pilots were indeed minding the store and trying to resolve their difficulty with the weather they were encountering.

747SP5

26th Jan 2015, 21:27

IndonesianTempo AirAsia Plane Often Experiences Trouble, Former Pilot Says | Economy & Business | Tempo.Co :: Indonesian News Portal (http://en.tempo.co/read/news/2015/01/23/056637174/AirAsia-Plane-Often-Experiences-Trouble-Former-Pilot-Says) - 'Former Garuda Indonesia pilot Capt. Shadrach M. Nababan, said -based on its logbook data - that the Airbus A320-200 serving AirAsia flight QZ8501 had experienced problems as much as nine times on its auto rudder trim limiter flight control in 2014.
Three days before crashing on December 25, 2014, flight QZ8501 experienced a 'return to apron' twice, according to Shadrach.
Flight QZ8501's last transmission were revealed by the Transportation Minister Ignasius Jonan in a meeting with the Transport Commission Legislature on Tuesday, January 20, 2015, saying that the aircraft tried failed gain altitude and stalled before vanishing from Soekarno-Hatta's ATC (Air Traffic Control) radar. The plane attempted to climb from 32,000 feet to 37,000 feet between 06:17 to 06:17:54 am (Western Indonesian Time).
At 6:17:54, the plane stalled from 37,000 feet to 36,000 feet in six seconds, and to 29,000 feet in 31-seconds. At 6:18:44 am, the plane vanishes from ATC radar.
Shadrach explained that commercial planes are not allowed to do any extreme climbs in the air because it does not have the same power as fighter jets.'

OzSync

26th Jan 2015, 21:30

SAMPUBLIUS:

Your post raises some thoughts.

Why are Airbus machines designed with such load protection?

Are they built like Boeings?

Would an A330 survive in the above circumstances if full control were allowed?

auraflyer

26th Jan 2015, 22:13

Load protection would have done What ??

Is there a related problem: I'm getting the impression that when things go seriously amiss, will the computers try to hang on too long, and in the process make such extreme inputs, that when they finally drop and hand control back, things are already much further gone than they should be?

ventus45

26th Jan 2015, 22:52

auraflyer

Quote: Load protection would have done What ??
Is there a related problem: I'm getting the impression that when things go seriously amiss, will the computers trying to hang on too long, and in the process make such extreme inputs, that when they finally drop and hand control back, things are already much further gone than they should be?

Auraflyer - In a word - Yes

What we have is the reverse of the Sourcerer and the Apprentice.

In normal life, when the Apprentice screws up, the Master Tradesman has to perform the rectification, and the apprentice gets a kick up the arse and "learns".

What we now have, in the bus philosophy, is the system is the annointed Master Tradesman, that is so much smarter, quicker, better, and all knowing, than the apprentice pilot.
When the Master Tradesman screws up, makes a mess, then compounds it into a bigger mess, and then gives up, it throws it all back onto the apprentice pilot with the instruction:-
"OK son, let's see how much you have learned so far. Show me how to fix this problem, and save our arses, before we loose the contract."

Put bluntly, the real responsibility for these "make it worse - then give up and kick off" automation facilitated needless disasters - should be laid fairly and squarely where it belongs - at the front door of the mentoring and sponsoring certification authouity - La Direction de l'Aviation Civile.

despegue

26th Jan 2015, 23:50

Thanks, Aguadalte,

That is what I meant, haven't flown Airbus since 2003.

peekay4

27th Jan 2015, 00:58

Alternate law limits would have prevented that- and plane and passengers would have been swimming with the fish . .

Except that on the 'Bus, with one engine out you are still in Normal Law with full protections in place, and the plane would not have rolled into the spiral dive to begin with...

Simply losing one engine on a 747 should not result in total loss of control by the crew and a 5g 30,000ft plunge. Once he took manual control, the Captain made several basic mistakes and nearly killed everyone on board. That is completely unacceptable.

If anything, China Airlines 006 further supports the Airbus design philosophy.

RifRaf3

27th Jan 2015, 01:17

GPS gives you ground speed (the speed of the aircraft relative to the ground), which is irrelevant for flying. You need Airspeed (the speed of the Aircraft relative to the air).

Since icing can affect all IAS probes at the same time, thus nullifying cross-checking, it would make sense to have GPS crosschecking for errors implicit in rapid changes in IAS. If IAS changes unexpectedly and GPS does not then a warning could be issued re suspected icing. It's just another bit of filtering software that perhaps needs to be written.

glendalegoon

27th Jan 2015, 01:33

maybe we just need hotter pitot tubes and vanes

oh, and even without GPS, if you are level at cruise power settings, there is a good chance you are making cruise indicated airspeed.

I want all those who are NOT pilots to know that we train, when we are very early in our pilot training how to deal with loss of indicated airspeed, for what ever reason...

In the larger planes, we actually have a piece of paper in our POH which gives power setting, pitch attitudes and other things which could be used IN LIEU of indicated airspeed...even offering things like this if we LOSE THE RADOME for some reason or another.

I love all the advice from non pilots or engineers, but the best thing I could ever offer to an aircraft designer is the term: KISS

KEEP IT SIMPLE STUPID

but it seems that modern planes have done just the opposite.

Sailvi767

27th Jan 2015, 01:36

That's what the Airbus A330 company pilots thought when they pulled a engine back to simulate a failure with one hyd system shut down. It did not go so well for them.

Gretchenfrage

27th Jan 2015, 02:16

auraflyer

Quote:
Quote: Load protection would have done What ??
Is there a related problem: I'm getting the impression that when things go seriously amiss, will the computers trying to hang on too long, and in the process make such extreme inputs, that when they finally drop and hand control back, things are already much further gone than they should be?

Auraflyer - In a word - Yes

What we have is the reverse of the Sourcerer and the Apprentice.

In normal life, when the Apprentice screws up, the Master Tradesman has to perform the rectification, and the apprentice gets a kick up the arse and "learns".

What we now have, in the bus philosophy, is the system is the annointed Master Tradesman, that is so much smarter, quicker, better, and all knowing, than the apprentice pilot.
When the Master Tradesman screws up, makes a mess, then compounds it into a bigger mess, and then gives up, it throws it all back onto the apprentice pilot with the instruction:-
"OK son, let's see how much you have learned so far. Show me how to fix this problem, and save our arses, before we loose the contract."

Put bluntly, the real responsibility for these "make it worse - then give up and kick off" automation facilitated needless disasters - should be laid fairly and squarely where it belongs - at the front door of the mentoring and sponsoring certification authouity - La Direction de l'Aviation Civile.

Amen to that. It has been my argument since 20 years.
Being called a Airbus enemy for so long, i only opposed the stated above. It would take a minor programming change to rectify the system. But it would admit something ...... and therefore we will cynically witness another Airbus c-u.

Did you notice the absence of the usual Airbus lobby-suspects' trivial contributions on here? It's getting ever more difficult to find excuses for the system with the similarity of these accidents.

But you have pointed out the real responsible body in your contribution.

FlyerBabe

27th Jan 2015, 03:56

So, it looks like the latest plan is to skip the floatation bags and just winch it up.

FlyerBabe

27th Jan 2015, 05:31

So last word is they may have 6 meters secured.

Ollie Onion

27th Jan 2015, 06:30

Quite simply, the investigators KNOW what happened by now. The fact that there has been exactly ZERO airworthiness directives sent our to Airbus Operators would suggest that we can almost rule out Structural / Equipment failures and are more likely looking at a mishandled event. Investigators probably just trying to figure out what to pin on airbus to shift some of the blame much like the ASIANA investigation which concluded that the pilots were duped into crashing by a 'hard to understand' autothrottle.

Clandestino

27th Jan 2015, 06:47

SAMPUBLIUS, auraflyer and Gretchenfrage.

Your concerns about load protections on Airbus have been repeatedly raised ever since features of Airbus FBW have become known, about a year or two before first A320 entered service. They have being pointedly ignored by aeronautical powers that be ever since. There are at least two possible explanations for this:

1) there is a worldwide conspiracy started by Airbus and involving all of the aviation authorities that accept the Airbi to be of airworthy design (that means all of them!). This conspiracy is very stealthy, no one has ever managed to prove its existence yet it is so powerful it suppresses data on every incident where overstressing the aeroplane saved the day so perception is there was none and whistelblowers are sidelined to anonymous fora.

2) whoever proposes that 2.5G with 50% reserve is too low for transport category aeroplanes combined with FCS actively maintaining the limit has no clue about certification criteria, structural design of the transport category aeroplane or operations of them - provided they really believe what they write - and everybody who is somebody in aviation is right to ignore that contributions as their notions represent acts of spectacular ignorance. There really is not any accident where transport aeroplane could be realistically saved by being built stronger instead of not getting it her into mess in the first place.

Your pick.

IMO that is a logic flaw. I posted earlier about a China 747 in 1985 that had an engine failure, went into a major spiral dive, and was recovered- and still flew - but it was severly bent with estimated 5G loads during pullout. Wings were bent up permanently sever inches at tips, who knows how much during flight. Part of horiz stabilizer was torn away.. plane landed safely, and was repaired and put back into service. Structural limit load test the wings on a 747 go several feet above top of fuselage at 2.5 to 3 G equivalent ..Transport category aeroplanes have to withstand 3.75 G at MTOW, clean, without damage or plastic deformation. For the time being, I'll leave it at this to solicit some entertaining responses.

Dynasty 6 accident happened near the end of cruise. They were lucky to be light yet with enough fuel in wings.

Loss of engine on FBW Airbi does not degrade control laws. Alpha protection would prevent the stall and bank protection would prevent the spiral dive.

Did you notice the absence of the usual Airbus lobby-suspects' trivial contributions on here? Speaking of devil.

PastTense

27th Jan 2015, 07:01

A preliminary report into last month’s crash of an AirAsia passenger jet that killed 162 people will not include an analysis of the black box flight recorders, an Indonesian investigator said today.
The preliminary report, which the ICAO requires within 30 days of the date of the accident, will include “information on the plane, the number of passengers and other information like that”, NTSC investigator Suryanto told Reuters.
It will not include analysis from the cockpit voice recorder and flight data recorder, both of which were recovered by divers from the bottom of the Java Sea.
Data from radar and the aircraft’s two “black box” flight recorders is providing investigators with a clearer picture of what occurred during the final minutes of Flight QZ8501.
Flight QZ8501: Investigator says crash report won?t include black box data | Malaysia | Malay Mail Online (http://www.themalaymailonline.com/malaysia/article/flight-qz8501-investigator-says-crash-report-wont-include-black-box-data)

So it doesn't look like there will be an official explanation anytime soon as to what happened.

FlyerBabe

27th Jan 2015, 07:08

Latest I hear is they may not raise the fuselage and the ships involved may be recalled. No additional bodies have been recovered.

edit: ships were recalled.

DType

27th Jan 2015, 08:13

The GPS speed question that keeps on getting asked ad nauseam usually only gets half the answer it deserves.
With many apologies to those who know this well, IAS is INDICATED Air Speed. If a plane climbs at constant IAS then, due to the decreasing air density, its actual air speed will increase progressively. This is good, because if the speed didn't increase then the lift would decrease.
IAS is an exceptionally useful parameter, GPS speed in itself is doubly useless because it is:- 1) not airspeed, and, 2) requires desnity information and calculation to be of any use.
But GPS height data could be useful, in this engineer's opinion.

GunpowderPlod

27th Jan 2015, 08:21

27Jan15 1700 hk
QZ8501 recovery operation called off: Indonesian military - Channel NewsAsia (http://www.channelnewsasia.com/news/asiapacific/qz8501-recovery-operation/1618784.html)

JAKARTA: The Indonesian military on Tuesday (Jan 27) called off efforts to recover the wreckage of an AirAsia plane that crashed into the Java Sea last month after failing for several days to lift the fuselage.

"All of our forces are being pulled out," said Rear Admiral Widodo, a navy official overseeing the search and rescue operation. "The operation has been ongoing for 30 days so the joint team has pulled out," he said. "We apologise to the families of the victims. We tried our best to look for the missing victims."

8che

27th Jan 2015, 08:21

The 787 supplies a back up IAS derived from AOA and inertial system should you lose all pitot's. The GPS takes over altitude. A very nice feature.

p.s the GPS/FMC ground speed indicator is always a vital tool for handling unreliable airspeed especially if you still have wind data.

The vital difference with Boeing is that there are no control FBW inputs to overspeed/underspeed. (Only trim inhibit) This means any failure is passive and Boeing want the pilot to have control over the aircraft not the other way around.

I am losing count of the number of airbus incidents of altitude climb/bust due to transitory overspeeds.

Throttles that don't move/ sidesticks cancelling each other out with inputs not visible to the other pilot/ FBW system that commands pitch manoeuvres without the autopilot..........The airbus design philosophy counters not helps piloting ability and awareness especially when its needed most with out of the ordinary occurrences.

Andrewgr2

27th Jan 2015, 08:43

I'm surprised that no-one has commented on the report above by 757SP5 at http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-130.html#post8841170 about rudder problems with this aircraft.

Apparently Former Garuda Indonesia pilot Capt. Shadrach M. Nababan, said -based on its logbook data - that the Airbus A320-200 serving AirAsia flight QZ8501 had experienced problems as much as nine times on its auto rudder trim limiter flight control in 2014.
Three days before crashing on December 25, 2014, flight QZ8501 experienced a 'return to apron' twice, according to Shadrach.

Could 'auto rudder trim limiter flight control' problems be the key to this crash?

Sorry to see the wreckage is now to be abandoned, but I guess there is a trade off between the risk to recovery team members and any benefit that might result from their efforts.

de facto

27th Jan 2015, 08:46

Agree,how many times did Airbus aircraft safety systems avoided a disaster due to crew.....

deanm

27th Jan 2015, 08:56

Failures in airspeed measurements seem to be increasingly implicated in aircraft upsets.

Rather than relying on pitot-based airspeed measurement, a number of contributors have suggested GPS-based approaches (which are unsuitable, as discussed).

As a back-up, could a fuselage-embedded mini-RAT type system work - an air-driven impeller (probably rather smaller than the cooling fan in your PC)?

Or perhaps even an acoustic system: a simple hull-mounted microphone which detects airspeed-derived noise? Frequency tuning or filtering could avoid interference from engine-derived noise.

Momoe

27th Jan 2015, 09:39

RifRaf3

27th Jan 2015, 09:45

Each of those is still subject to icing. You need sensing from something that does not involve air that's moisture laden. A combination of inertial reference and GPS can compute a probable wind until you get out of icing conditions. There is no simple solution.

Ian W

27th Jan 2015, 11:15

Deanm - good observation.

Building on that, why not incorporate GPS data as a fall back, in the event of a pitot head icing up, the FMS advises that there is a mismatch between GPS/Pitot speeds and does nothing pending pilot acknowledgement.

Descent to a lower level at this stage gives more speed margin whilst reducing a stall scenario, the secondary effect of pitot head icing is loss of awareness and subsequent loss of control (in some cases).

Just thinking about being in heavy turbulence/wind shear at night with no speed awareness and close to speed margins gives me the sweats.

Prevention is better than cure.

It is already being done - only a matter of time before the 'bus has it too

The 787 supplies a back up IAS derived from AOA and inertial system should you lose all pitot's. The GPS takes over altitude. A very nice feature.

p.s the GPS/FMC ground speed indicator is always a vital tool for handling unreliable airspeed especially if you still have wind data.

Lonewolf_50

27th Jan 2015, 12:07

Didn't Airbus family aircraft have a recent Airworthiness Directive release having to do with pitch control in certain circumstances? (The answer is yes). If you look at the longer view of the A320 and its successors, in terms of the concept of the control architecture, this is a mature system. That they are still "discovering" certain issues with it (albeit having to do with rarely encountered variables lining up) speaks to the complexity of the system and its software suite.
With the complexity of the system in mind, it speaks for a need for more training and education on the system, not less, for those who operate it. What is curious is that this seems to be counter-intuitive to the suit wearing sorts who design and who buy these systems to be put into place to deliver people from here to there for a fee.

Maybe it's a paradox: if one is to build an airplane that "a concierge can fly" you have to build a complicated plane, which means that this concierge actually needs to have more invested in him on an ongoing basis in training and education in order to operate this complex machine.
No Free Lunch?

Analogue: you need deeper training and education of flying a swept wing transport aircraft than you do a Piper Cub.

Blake777

27th Jan 2015, 12:41

Air Asia Indonesia's update tonight states that today's operation was called off due to strong currents. The decision to abandon recovery was made in part because observations of the fuselage have led investigators to believe there are no further bodies to be recovered from this part of the wreckage. (I suppose the cynical part of me observes that two failed attempts to haul the wreckage aboard a recovery vessel, the second of which resulted in the fuselage breaking into two pieces, is not likely to have helped that situation.)

I'm disappointed there has not been confirmation of the location of the cockpit or recovery of occupants. Whilst it may not be of material benefit after recovery of the FDR and CVR, it would have been fitting to lay them to rest, having gone down fighting it seems.

CX-HOR

27th Jan 2015, 12:42

Disclaimer, I fly Airbus so don't call me a Boeing troll.

How does a whole family of aircraft (Airbus FBW) get certified when through Air Asia, Air France, Lufthansa and the many other reports of, at the least UAS show a major design flaw still not corrected.

Ie where a simple disagreement between similar sensors (AOA, Pitots/ADRs), followed by computer rejection of perhaps the one accurate sensor, may then cause the computers to initiate drastic action on the flight controls requiring IMMEDIATE and correct response by the pilot whilst assessing what is valid and what isn't, with spurious ECAMS, Aural Alerts and displays, be allowed to be certified safe.

All, without letting the pilot know in the first instance of a disagreement between sensors and then the decision by the computer to reject a specific sensor, giving the pilot no indication anything is abnormal until all hell breaks loose?

To me this borders on the criminal, at the very least an immediate ECAM advising a specific sensor has been rejected, or even better is approaching the limits of rejections similar to "IRS excessive drift etc", would at least advise the pilot that something is amiss and the computers maybe about to take drastic action..

barit1

27th Jan 2015, 13:03

The recovery process of lifting a fuselage-full of water (wings too) is the subject of a tech memo from Boeing. And it was proof-tested in March 2002 when the precious, sole survivor B-307 was ditched in Puget Sound.

The essential limitation is the rate at which the aircraft is lifted as it surfaces; the water must be allowed to drain slowly, and the hoisting reduced to a rate of inches per hour, to insure the aircraft structure and lift cables/straps are not overloaded.

Granted, the 307 was in sheltered and shallow water of Eliott Bay (in fact, still awash, not submerged). This made the job easier, and the place was probably swarming with engineers from Boeing's nearby offices, and the hoist was successful. The 307 was restored to airworthy, and was ferried one last time to the NASM facility at IAD. History, Travel, Arts, Science, People, Places | Air & Space Magazine (http://www.airspacemag.com/ist/?next=/flight-today/restoration-soggy-stratoliner-32605581/)

The precautions from the Boeing 307 experience were probably impossible to execute in the choppy Java sea, leading to the inevitable breakup of the AirAsia ship.

Less Hair

27th Jan 2015, 13:55

What a/c parts could be secured? Now what do we do if we end up in need of -for example- certain electronics boxes and maybe rudder actuators to find out what really happened? Go back into the mud with another campaign again? If we end it here we might lose all traces.

Or is this case considered solved already behind the scenes?

CONF iture

27th Jan 2015, 14:05

Loss of engine on FBW Airbi does not degrade control laws. Alpha protection would prevent the stall and bank protection would prevent the spiral dive.
I like that glassy brochure too ...

BARKINGMAD

27th Jan 2015, 14:54

Doubtless Airbus and AirAsia know by now EXACTLY what happened and the lawyers are gearing up for the fight.

If it was a rudder hardover, as allegedly previously reported, then the maintenance is in the frame.

If it is an Airbus peculiarity about which few if any of the crew were/are aware, then manufacturer is in the frame.

If it was mishandling of a recoverable situation, then the airline and xAAs are in the frame for inadequate training.

Meanwhile the amateur accident investigators in this forum will continue to speculate about spins, stalls, aerobatics, impact angles, aircraft flight paths, configurations and so on ad nauseam.

Perhaps if those in the industry might take some time out, particularly the designers of the kit which is confusing the operators so regularly (no particular type specific) to try reading "The Organized Mind" by Neuroscientist Daniel Levitin (as soon as it is published 2 days from now,if you want it Kindled).

If that task can be achieved, either before their next flight or before their next design, then we might be part-way down the road to reducing the rates of such tragic mishaps.

Listening to the author on BBC Radio 4 Monday 26th January at 0900Z (try pod-casting it?) it appeared that someone with knowledge of the brains of the species of advanced-level Bonobo Chimps, who design and operate the current airliner fleets, has cast light on the supreme arrogance of said Chimps and their (possibly) false confidence in their ability to multi-task and handle multiple information threads.

If I may quote: "The brain worked well enough for our Stone Age ancestors and has barely changed since then, during which it has been forced to absorb vastly more knowledge than ever before in human history. "

I await the incoming fire, yours truly, a retired Bonobo! :)

airsound

27th Jan 2015, 15:02

BarkingmadBBC Radio 4 Monday 26th January at 0900Z try
BBC Radio 4 - Start the Week, Organising the Mind (http://www.bbc.co.uk/programmes/b0505zw1)

An excellent listen

Stabilo31

27th Jan 2015, 15:09

Could 'auto rudder trim limiter flight control' problems be the key to this crash?

Never heard of such device on A320. Is it RTLU (Rudder Travel Limiting Unit)? Or rudder autotrim with AP?

Ian W

27th Jan 2015, 16:03

Perhaps if those in the industry might take some time out, particularly the designers of the kit which is confusing the operators so regularly (no particular type specific) to try reading "The Organized Mind" by Neuroscientist Daniel Levitin (as soon as it is published 2 days from now,if you want it Kindled).

If that task can be achieved, either before their next flight or before their next design, then we might be part-way down the road to reducing the rates of such tragic mishaps.

Listening to the author on BBC Radio 4 Monday 26th January at 0900Z (try pod-casting it?) it appeared that someone with knowledge of the brains of the species of advanced-level Bonobo Chimps, who design and operate the current airliner fleets, has cast light on the supreme arrogance of said Chimps and their (possibly) false confidence in their ability to multi-task and handle multiple information threads.

If I may quote: "The brain worked well enough for our Stone Age ancestors and has barely changed since then, during which it has been forced to absorb vastly more knowledge than ever before in human history. "

I await the incoming fire, yours truly, a retired Bonobo! :)

It is also an idea to look at the research on cognitive tunneling/cognitive narrowing/attentional tunneling and multiple cognitive resource theory. All of which, unfortunately, appear to have been disregarded in recent cockpit system integration.

Try reading the paragraph above and reciting a nursery rhyme - you can't do it - your brain only has one verbal analysis channel if someone speaks to you at the same time you may not even hear them - unless a trigger word like your name is part of what they say. If you are listening to an audible alert message with words in it - you will be unable to comprehend an ECAM alert saying something different or what the other crew member is saying. These are cognitive limitations to multitasking that all humans have.

This is the reason using cell phones while driving can be dangerous as if the caller starts describing the route to get somewhere and what to look out for... you start overloading the 'spatial analysis channel' in your brain and your brain start to imagine what is being described and stops seeing things in reality and following the road accurately and you may miss turnings and wander on the road.

In the cockpit if the system sensing a problem delivers a cacophony of alerts and messages plus the 'automation surprise' that something has stopped working properly 'that never fails', your brain will stop hearing things and/or stop being able to read and comprehend ECAM messages or what the other pilot is saying and will focus (cognitive tunnel) on one aspect of the environment and try to 'fix' that often to the exclusion of everything else.

bubbers44

27th Jan 2015, 18:08

"It is also an idea to look at the research on cognitive tunneling/cognitive narrowing/attentional tunneling and multiple cognitive resource theory. All of which, unfortunately, appear to have been disregarded in recent cockpit system integration."

At one time not too long ago inconveniences like AP malfunctions, UAS, stuck Loc or GS needles and instrument failures were not cause for alarm because we just flew the plane. A stall warning was an embarassment because you weren't paying attention and easy to remedy.

No accident should ever be blamed on any of these inconvenient failures because you as a pilot shoulld be able to fix it easily with a bit of airmanship.

Lonewolf_50

27th Jan 2015, 18:14

bubbers, the industry seems to think that stick shakers are a handy way to provide tactile clues to an impending stall. They've been in use for some time.
Do you disagree with them in principle?

I think I understand that adapting that to sidesticks as Airbus use them in the control scheme, stick shaker is counterproductive to the desired control inputs being made by the pilot flying when controls to prevent a stall are the desired output (from him/her).

(If I misunderstand the reasoning, anyone with an understanding please correct me).

Leightman 957

27th Jan 2015, 18:27

As more than one poster has pointed out, the continued emphatic Yes-they-do No-they-don't FMC arguments between people trained on and flying A320's IS having an effect on those not privy to front seat ops, to whit: increasing certainty that neither side has all the answers. However one side of that argument is saying that not all answers are known (and the search for additional details is at minimum prudent), and the other side seems to be saying that enough is known and the systems are presently predictable and sufficient for all flight regimes that a competent pilot would ensure.

An explanation delivered with great emphasis and and certainty does not guarantee accuracy. All the people lower in piloting stature and experience than current A320 pilots have a valid interest about what happened and why. Virtually all those people either maintain or very much want to maintain great confidence in the front seat pilots. I am finding that the most emphatic, contentious, and posts filled with certitude are the ones in which I have the least confidence, and I do not like having that reaction.

This current argument may appear unjustified following a clean accident rate five years from now, but will be viewed as prescient and proper if additional accidents continue.

Blake777 FWIW there were earlier but subsequently unconfirmed reports that the captain's body had been recovered. Given the local currents and sectioned fuselage, the expectation that all bodies will be recovered is unreasonable. An unseen blossoming of neutral or lower buoyancy fuselage contents that must have occurred just from additional damage during the raising attempts must be presumed.

RE Bonobo's: One can simultaneously and successfully task multiple inputs if those specific interactions have been repeated until they become familiar lower brain categories. We have all kinds of such examples. But ask someone to juggle multiple simultaneous questions which are all unfamiliar, or which were repeated during “education” a few times far in the past, or for which subsequent training or subsequently discovered details demand different actions (or even opposite actions) from that previously learned, all of which in the present demand high level brain function, and the thought process then goes sequential and the required time expands.

IanW is right: cognitive functions of different categories can't co-exist and, unfortunately, that varies to some degree between people. Demonstrations of quick “answers” to complex questions fall into either the lengthy rote category, or in a few individuals, intuition which is outside the scope of direct teaching and still involves luck. I disagree with Ian to the small extent that multiple aural alarms made familiar by virtue of “rote” repetitions do not need to continually demand attention, but rather a signal filed in short term memory requiring only that its cessation be noted. But an aural alarm is an on-off decision, not one (by itself) of conflicting information simultaneous with multi-relational possibilities that, perhaps just this once, is outside of all rote memory.

Sop_Monkey

27th Jan 2015, 19:32

It ain't an idea, it is a necessity for recovery from nose high unusual attitudes to "unload the wings", if all else fails. It is practiced every 6 months where I and every other non computer pilot, carried out recurrent training at any reputable ATO. If you are in a "deep stall" and unable to get the nose down, that is your only hope, apart from a recovery parachute attached to the tail. I ask the question again, has A/B carried out a fully developed/deep stall at altitude in their "un-crashable aircraft"? If not, then why not and why was it certified in the first place? Tell me please.

Yes pilots shouldn't be getting themselves into a stalled situation in the first instance but it is happening, mainly due to overall inexperience.

Computers are very good. I am the first to admit, are better at flying normal procedures than any human will ever be and in addition frees up the handling pilot for extra monitoring etc.. However if a human is allowed to get the a/c into a scenario due to human error, the computers are not so good at recovery action. Remember AF.

I have never flown an Airbus and never will. So I am both.

captplaystation

27th Jan 2015, 19:41

Contact Approach,

I refer you to my post above (2624 ) which quotes a current Airbus pilot. . . . I am happy you are happy, but less happy that you don't see beyond the superficial "joy", which may cloud some less savoury "truths" in your machines "mindset" .

Contact Approach

27th Jan 2015, 19:44

I refer you to an article on the Flight Global website regarding the real problem on the matter, and note its not all airbus!

AirAsia QZ8501 accident may reflect a trend - 1/22/2015 - Flight Global (http://www.flightglobal.com/news/articles/airasia-qz8501-accident-may-reflect-a-trend-408168/)

Accidents involving aircraft cruising through tropical or sub-tropical zones – like AirAsia flight QZ8501 in December – are happening often enough now for the industry to have cause to refresh pilots on the risks. Therehave been three accidents in similar circumstances over the last 10 years.

When the Air France flight 447 accident report was published in 2012, French accident investigator BEA warned of the need to prepare pilots for high altitude aircraft handling – and comparisons with the AirAsia Airbus A320 accident in December have highlighted this apparent area of operational weakness.

But now that Indonesian accident investigators have formally stated they will not publish a preliminary factual report about flight QZ8501, it could be up to two years before operators know whether BEA's concerns are reflected in this case.

There have been three other accidents in the last 10 years that merit comparison with the flight QZ8501 incident over the Java Sea. These four accidents all took place during cruise in tropical/subtropical zones, and in all cases the crews were manoeuvring to avoid bad weather. In all four cases control was lost, and none of the crews made an emergency call.

Before the AirAsia loss in December, there was the Swiftair/Air Algerie McDonnell Douglas MD-80 loss over Mali last July, Air France Airbus A330 flight AF447 over the South Atlantic in June 2009, and West Caribbean Airways’s loss of an MD-80 series aircraft over the Caribbean Sea in August 2005.

Looking back further in the archives, jet accidents with this combination of circumstances did not happen – although flights in those days were just as likely to have to fly through the tropics as they are now. Tropical weather seems to be a “modern” accident trigger, in the same way that loss of control in flight (LOC-I) is now accepted by ICAO as the greatest aviation risk to life, when before the 1990s it was not.

There is an uncanny similarity between the onset of problems in the AirAsia and Air France cases. Both pilots suddenly selected a nose-up attitude that gave them a rapid rate of climb – estimated to be 6,000ft/min in the AirAsia case, and known to be 7,000ft/min for AF447 – at an altitude where demanding such performance from the aircraft is unrealistic. In the Air France case the aircraft was in a full stall within 46s, and the crew did not apply stall recovery procedures.

When it published pilot training recommendations in the AF447 official accident report,BEA called for a list of actions to be implemented. These including training in stall recovery at high altitudes as well as low, practicing flying in alternate and direct control law as well as normal law close to the edges of the flight envelope, carrying out flying exercises to test whether pilots understand aerodynamics in practice, as well as in theory, and introducing more surprise events into recurrent training scenarios to prepare pilots – individually and as a crew – to react calmly to the unexpected.

Pilots are instructed not to disconnect the autopilot at high altitudes for good reason: with reduced vertical separation minima applying above 29,000ft in most parts of the world now – and handling being a delicate matter at high altitude – manual flying is banned there. BEA points out, however, that if the automatics trip out – as they did in the case of AF447 – the pilots have no choice, so they should be reminded in the simulator of what handling is like up there.

Pilots talk glibly on Internet forums about “Coffin Corner” – as applied to the aircraft’s narrowing flight envelope when it approaches the high altitude edges of its performance capabilities – but according to BEA they need to be reminded of what to do if the edges are actively breached, because recovery is almost always possible if the correct procedures are applied.

Yes I agree, Airbus needs to take the above into consideration... nothing a bit of training and/or a big red disconnect button couldn't solve.

fgrieu

27th Jan 2015, 20:03

Failures in airspeed measurements are increasingly implicated in aircraft upsets.

Am I right that it would be feasible to obtain a reliable (if not extremely accurate) airspeed measurement from a gas turbine engine, derived from some of its existing pressure and rotational speed sensors, under the assumption that the blades did not suddenly break?

[edit: since there are at least two such engines on commercial airliners, we have some built-in redundancy; the idea is is of course to feed this to HAL as a substitute for iced pitots]

Sop_Monkey

27th Jan 2015, 20:17

fgrieu

Correct and the power parameters apply to all a/c. If a/s is lost refer to a known power setting and attitude.

This should be second nature to an overall experienced pilot who has come up through the ranks. What the sausage factories are churning out now one can only guess.

bugg smasher

27th Jan 2015, 22:35

This is a very difficult argument, all of the Airbus protections built into the system architecture, how many lives has that saved over the years? Not quantifiable, none of those 'saves' ever get reported. Who wants to do paperwork? GroundSpeed Mini for example, these airplanes are well protected from wind shear events.

As an Airbus driver though, I am increasingly concerned at the growing number of OEB's in our QRH; if this happens, ignore what the airplane is telling you, do this instead. Counterintuitive in many ways, some of us have developed nervous ticks in one body part or another.

I think most will agree though, a combination of thorough system knowledge and simulator training make us at least as statistically safe as the Boeing folks, although I'm sure the Seattle boys have their dark places too.

Needle, ball, and airspeed, if I could voice one criticism of the AB design, in abnormal and emergency modes, it tends to work against the natural instincts of a well rounded and experienced pilot.

IcePack

27th Jan 2015, 22:44

How many years since AF 447 & simulators STILL do not represent correct handling at high altitude. It's criminal if you ask me.

FullWings

27th Jan 2015, 22:47

Correct and the power parameters apply to all a/c. If a/s is lost refer to a known power setting and attitude.
Yes, but this may not work immediately in an aircraft with hard envelope protection which is suffering from erroneous inputs. It may be trying to follow a trajectory that is completely wrong for the situation and making it harder to recover by doing things like trimming the stabiliser in the opposite direction needed.

By the time this has been diagnosed by the crew, probably during a time of very high workload in the middle of an upset, it leaves little time for remedial action. “Known power setting and attitude” will not recover the aircraft from a stalled condition. Much more aggressive inputs are required.

GarageYears

27th Jan 2015, 22:49

How many years since AF 447 & simulators STILL do not represent correct handling at high altitude. It's criminal if you ask me.

Take that issue up with the airframe manufacturers. Simulators are only as good as the data available. If the aircraft manufacturer can't provide stall data (for example), because stalls aren't included in the data package, how can the sim manufacturer model the behavior? Certain extrapolations and modeling data may be valid, but remember simulators aren't magic - the underlying model can only be as good as the data provided.

BenThere

27th Jan 2015, 22:49

The A-320 family is a great aircraft to fly. The automation does so much for you.

But it can surprise you with anomalies, especially in the first year you fly it when you forget to activate the approach, fail to reach planned cruise altitude and don't re-cruise, don't know the implications of alt star and vertical speed, et al.

If you're like me, you learn by being surprised, then learn to avoid the surprises. For example, I know that when the magenta diamond is below the horizon line, it's bad.

But my main point is this: WHENEVER YOU ARE SURPRISED THE MOST IMPORTANT MEMORY ITEM IS:

AUTOPILOT OFF
AUTO THRUST OFF
FLIGHT DIRECTORS OFF
IF AIRSPEED IS UNKNOWN, SET 90% N1

Then figure out what is going on and fix it.

Sop_Monkey

27th Jan 2015, 23:04

Benthere

Well put and totally agree on the surprise memory items.

ACMS

27th Jan 2015, 23:23

1/ if you forget to activate approach it will do it anyway

2/ in cruise where most of these unreliable airspeed incidents have occurred you don't set 90% N1. You adopt the normal attitude and thrust your Aircraft uses for cruise.

The memory items in the QRH are for initial actions after takeoff, they then say ABOVE MSA CIRCUIT ALTITUDE LEVEL OFF FOR TROUBLESHOOTING.....

If you are in Cruise you would compete the memory items ( A/T F/D A/P off ) then skip to the last step in the table giving you the pitch attitude and thrust setting for level flight.....something you should already know anyway because you've just be looking at it for a long time....

For the A330 it's 2.5 deg and 78% N1 for most conditions.

BenThere

27th Jan 2015, 23:38

If you haven't activated the approach, it won't always, or even most of the time, activate itself. You might find yourself accelerating, fully configured, at TOGA thrust.

90% N1, on the A320, will assure a safe airspeed (assuming you're not approaching a stall). While the appropriate power setting, based on configuration, weight, etc., will likely be a bit lower, 90% will keep you flying and avoid a stall until you stabilize the aircraft and adjust power to the appropriate setting. Figure that out later, when you have the time. If you don't know your airspeed, and your protections are suspect, erring on the high side, I think, is wise.

FoxHunter

28th Jan 2015, 00:23

Contact Approach
I'm sure there have been more accidents due to your '90 degree bank' idea.
I'm a 320 driver and can only say that the aircraft is a joy to fly, those slamming it on here are either misinformed or have never flown her.

Funny you have a problem with the 90 degree bank. I was a test subject in the creation of an jet upset training program being put together. At the time I was a current MD11 Captain with a 40 year airline career with an all civil background. My fellow test subjects had varied backgrounds. One was a 757 Captain, ck Airman former USAF/ANG A7 pilot. One a 777 F/O Major US airline. One flew the A320 for a low cost airline, unknown background. I believe there were two others that I don't recall their background. I recall one of the first things we did was to go to the trainer which was set up like a B757, but single pilot. A big link trainer on a small centrifuge. The pilot sat in the seat and the computer flew the device to the desired position and then the pilot received the command Recover. I recall one of the events was around 80 degrees nose up wings level. How would you recover? Another was about the same, but inverted! How would you recover? The most challenging was simulated wake turbulance at I recall 6000 feet. After all the bumps the simulator had a bank of 120 degrees to the right, aircraft nose down of 20 degrees, airspeed had increased from 250K passing 340k increasing when the machine says "Recover". Only two of us managed not to hit the ground the first try. The trainer had a visual and conditions were CAVOK. I stood at the operator panel an watched the A320 guy hit the ground 6 times in a row. He never had a successful recovery.

maggotdriver

28th Jan 2015, 01:17

FullWings completely agree.

Yes, but this may not work immediately in an aircraft with hard envelope protection which is suffering from erroneous inputs. It may be trying to follow a trajectory that is completely wrong for the situation and making it harder to recover by doing things like trimming the stabiliser in the opposite direction needed.

By the time this has been diagnosed by the crew, probably during a time of very high workload in the middle of an upset, it leaves little time for remedial action. “Known power setting and attitude” will not recover the aircraft from a stalled condition. Much more aggressive inputs are required.

These quotes are from (mine and others') posts about 2337. Whilst we're speculating on possibilities, I think we should also look at this in the broader discussion of modern Flt Cntrl laws.

Ice crystal icing, blocked pitots, the aerolane thinks it oversped, then pulled up, overriding the pilots, stalled, got confused and gave up.

How many Airbus incidents have had confusion over what the laws were meant to be doing in the control environment? And, yes I have nearly 5000 hours flying the euro version.

60% of known ice crystal icing events occur in the tropics. If the rate is beyond the ability of the pitot system then I would infer that the blockage would occur at almost identical times and magnitude because of the homogenous nature over the small frontal area of the aircraft. Thus, resulting in an artificial overspeed due to it climbing. It is very difficult for any pilot to do something when the systems designed to protect you are now putting you in harm's way, against your intentions. I've never had it in the aeroplane but have done it in the sim and its the most uncomfortable feeling having FULL forward stick whilst the aeroplane pitches up opposite to your inputs.

Unfortunately, if your climbing, its night time, your looking at the radar, there's associated flashing from lightening, you're thinking about your clearance (or lack thereof), turbulence and the aircraft now "detects" an overspeed... Goodluck! You need to turn off multiple parts of the FCCs or the inputs i.e. ADR an IR (from OEBs) to regain the ability for you to simply level off and set the correct attitudes and power settings. This is from the Airbus manuals - combine voting logic from the section in the FCTM with Overspeed Protection from the Normal Law section in the Ops Manual, throw in a little knowledge about what happens in a climb with blocked pilot tubes and the rest looks like a replay of AF447.

Now, I'm not saying this is what happened but it is feasible and IMHO the most probable scenario. When the pilots most need manual control of the aeroplane, it can be taken away from them. Only, to be handed back when the AoA senses the CA has been exceeded. Helping or hindering?

Something is flawed in the design. Why can't you select the "big,red button" sure, its connected to the ACARS so the company, NTSB and Airbus will all be wanting to know why you did it but you should be able to!

electric-chris

28th Jan 2015, 02:41

Exclusive: AirAsia probe vets possible computer glitch, crew response | Reuters (http://www.reuters.com/article/2015/01/27/us-indonesia-airplane-computer-pilots-idUSKBN0L00X020150127)

(Reuters) - Investigators probing the crash of an AirAsia jetliner are examining maintenance records of a key part of its automated control systems, and how the pilots may have handled the plane if it failed, two people familiar with the matter said.

An outage of the twin Flight Augmentation Computers (FAC) could not have directly caused the Dec. 28 crash, experts say, but without them the pilots would have had to rely on manual flying skills that are often stretched during a sudden airborne emergency.

"There appears to be some issue with the FAC," a person familiar with the investigation said, adding that more information was being sought from the manufacturer and airline.

The pair of computers comprising the A320's FAC system is mainly responsible for controlling rudder movements and helping to keep the airplane stable, as well detecting windshear, or sudden changes in wind speed or direction.

Indonesian magazine Tempo reported a series of maintenance problems with the computerized rudder system of that particular aircraft in the days and months before the loss of Flight QZ8501.

Machinbird

28th Jan 2015, 03:08

Flight Augmentation Computers

http://www.efbdesktop.com/flight-controls/images/fcc-1s-1.jpg (http://www.efbdesktop.com/flight-controls/sys-7.5.1.html)http://www.efbdesktop.com/flight-controls/images/fcc-1s-2.jpg (http://www.efbdesktop.com/flight-controls/sys-7.5.2.html)http://www.efbdesktop.com/flight-controls/images/fcc-1s-3-red.jpg (http://www.efbdesktop.com/flight-controls/sys-7.5.3.html) FAC - The Flight Augmentation Computers have three main functions:

Rudder trim
Rudder travel limits
Yaw damping inputs
Alternate yaw
Flight envelope and speed computations
Wind shear detection

The most significant function wrt to QZ8501 is probably the Flight envelope and speed computations.
SourceFlight Control Computers | Flight Augmentation Computers (http://www.efbdesktop.com/flight-controls/sys-7.5.3.html)

MountainBear

28th Jan 2015, 03:21

Several posters have danced around the issue but it's better to be direct. The "Airbus philosophy" of envelope protection needs to be revisited. Either one of two things is going to happen over the long haul.

Scenario One: Flight envelope protection is lost and the pilot guesses. Solution: If all the pilot is going to do is guess the computer can do that just as well and probably better.

AUTOPILOT OFF
AUTO THRUST OFF
FLIGHT DIRECTORS OFF
IF AIRSPEED IS UNKNOWN, SET 90% N1

That isn't difficult logic for software engineers to program...

Scenario Two. Flight envelope protection is lost and the pilot uses his superior flying skills to correct the problem.

This for many is the preferred solution. However, it seems that maybe it isn't working as well as one would hope. :}

SAMPUBLIUS

28th Jan 2015, 03:32

Seems to me either

1) " HAL" although degraded has the final word or

2) the Pilot has the final control.

All else is 'yea but " arguments.

BA enables 2, Airbus uses 1.

Time for a review or change ??

Volume

28th Jan 2015, 07:13

All of which, unfortunately, appear to have been disregarded in recent cockpit system integration.This is because standardisation and saving money by doing things "the way we always did it" is found more important these days, than doing the right thing.
If latest cockpit design would be significantly different from the last generation, pilots would have problems to change between the two, which would either limit the types they are allowed to fly (with all the organisational and financial impact) or it would create new risks. It would also impact all training organisations being forced to teach two different ways of flying. However, somehow we already do have that situation with different aircraft having different levels and type of automation and protections.
Sometimes it is safer if everybody uses a poor design but is well aware how to handle it, than some using an improved design and losing awareness...

Sailplaneflier

28th Jan 2015, 07:19

Beginning on 1 June 2009, when 229 souls perished in the Atlantic crossing the ITCZ, in a thunderstorm, punching a cell, crossing extreme shear layers and entering 100+ mph updrafts combined with warm air (https://www.youtube.com/watch?v=CDNMp55wHIk&list=LLo_JQHvvibcRR7cccIBUh_w&index=24), followed by 100+ mph downdrafts of cold air and icing conditions -- the anvil portion of the cloud being even more extreme -- weather effects upon AB software have been discussed and re-discussed and recommendations made.

In December of 2014, 168 souls perished in the Java Sea crossing the ITCZ, in a thunderstorm, punching a cell, crossing extreme shear layers and entering 100+ mph updrafts combined with warm air (http://avherald.com/h?article=44e48c2a/0000&opt=0), followed by 100+ mph downdrafts of cold air and icing conditions -- the anvil portion of the cloud being even more extreme....

Both occurring while a FBW system reacts to such insults by trimming the THS to full nose up, and commanding its pilots to follow a flight director that indicates further full nose up, leading to these zoom climbs (The FAC's panicked response to perceived overspeed due to lack of speeds from the pitots and over-reliance on merely the pitots to determine speed.)

I remember being transfixed by the loss of AF447 and following most every thread since then with great interest.

WHENEVER YOU ARE SURPRISED THE MOST IMPORTANT MEMORY ITEM IS:

AUTOPILOT OFF
AUTO THRUST OFF
FLIGHT DIRECTORS OFF
IF AIRSPEED IS UNKNOWN, SET 90% N1
I remember posting back in 2009 on another forum that, at minimum, AB's should have a way of reverting back to manual, 'pilot mode' by providing its pilots with as much information (AoA display being one of them) as possible and providing the pilot with full, cable-and-lever, type control.

After another 168 lives lost, it is time to demand revision to the software. Revisions that gives the pilot back full control of a fully re-trimmed, completely flyable aircraft. I'd suggested back in 2009 that instead of dumping the the aircraft upon the pilots in an out of shape, out of control, "I give up" fashion, that:

The AIRCRAFT in the seconds during transition between AP to manual control, should instead:

1. Re-trim the THS to neutral
2. Re-trim the elevators and ailerons to neutral
3. Rapidly transfer fuel forward to achieve a more forward CG to improve flyability
4. Depending upon the altitude of the craft, and similar to the "exponential rates" and "dual rates" familiar to most radio control flight enthusiasts, give the pilot a dumbed down, not as twitchy/touchy and difficult to fly aircraft when at cruise altitudes, yet still able to provide plane-bending 100% deflections if necessary when control is completely handed to the pilot
and then

5. AUTOPILOT OFF
6. AUTO THRUST OFF
7. FLIGHT DIRECTORS OFF
8. IF AIRSPEED IS UNKNOWN, SET 90% N1
The big red button ought to do all of these things and HAL should press the button himself when unreliable airspeed/AoA sensors or otherwise promote AP auto-disconnect at cruise altitude.

(It would also seem that Airbus now has at least two sets of pretty good data on entry to stall, stall behavior, transition to deep stall, flat spin, 10,000+ per minute altitude loss.... to add to their simulator's knowledge base.)

Gretchenfrage

28th Jan 2015, 07:29

Not an Airbus pilot...

Are you saying that the autopilot can never be turned "off" on an Airbus?

You can never "disconnect" the A/P totally and fly it manually at any time?

I have to answer with a "yes, you still can, but ....".

If you "disconnect"the Autopilot, you still guide the Airbus via a Flight Control Computer, which in turn directs actuators and finally the aircraft (same goes with the Boeing FBW system).

The FCC has the famous "protections" built in who restrict the inputs to stay within the programmed envelope. To restrict and direct, the computer needs sensors and those sensors can fail or ice up or etc. etc. The "feeling" of the computer is therefore wrong, just as the resulting restrictions of the inputs will be wrong.

So in short: HAL is still fiddling with flight commands even with the AP OFF.

In an Airbus you would need to force the computer into the famous "Direct Law" to shut out the protections and have almost unrestricted control.
The whole problem is "if", or "how" and "how user friendly" can this be done.

Or in other words: When the :mad: hits the fan, does the average pilot have enough possibilities / skill / hard- or software to achieve this in survivable time.

Jet Jockey A4

28th Jan 2015, 08:45

The FCC has the famous "protections" built in who restrict the inputs to stay within the programmed envelope. To restrict and direct, the computer needs sensors and those sensors can fail or ice up or etc. etc. The "feeling" of the computer is therefore wrong, just as the resulting restrictions of the inputs will be wrong.

So in short: HAL is still fiddling with flight commands even with the AP OFF.

I understand the fact that Airbus designed a "protection" software into their flight computer's program which is fine by me, however are you telling me that if a pitot tube of some other probe (a la AF447) freeze up or just fails to function properly the FCC will still take its inputs and get confused instead of disregarding them and work in a "less than ideal" situation?

If this is a fact then Airbus is pretty stupid in letting a "compromised" sensor render their "protection envelop design" and/or aircraft controllability vulnerable. The compromised sensor should automatically be taken out of the loop and a warning to the crew annunciated immediately.

Furthermore, if the whole FCC is somehow compromised, there should be an easy and quick way to disable it so that the pilot can regain full control of the aircraft (like in a conventional aircraft) albeit without the "protection envelop".

Finally, what I would like to know from an Airbus driver is assuming this crash was related to another stall like the AF447 flight, can a pilot regardless of the FCC's confusion just say "I have control", switch the AP off, the AT to off and FD if need be, fly straight and level and set a given N1 or EPR value to keep the aircraft from falling out of the sky?

ironbutt57

28th Jan 2015, 09:16

Finally, what I would like to know from an Airbus driver is assuming this crash was related to another stall like the AF447 flight, can a pilot regardless of the FCC's confusion just say "I have control", switch the AP off, the AT to off and FD if need be, fly straight and level and set a given N1 or EPR value to keep the aircraft from falling out of the sky?

Yes you can....fly Pitch-attitude-thrust.. and that is the procedure for "airspeed unreliable"...do all pilots who grew up in these planes have the skills? NO...it's a skill that has over time disappeared...having facilitated many LOFT scenarios in the simulator, it never ceased to amaze me when conflicting information from ADR sensors caused problems, to see how many pilots seemed UNAWARE of their present thrust and attitude before things went haywire..the other fairly recent "surprise" on the 320 family anyway, was water penetrating the seals on the AOA vane shaft and freezing the vane in some undesirable position, forcing the crew to select 2 of the 3 ADR off to place the system in alternate law, removing stall/overspeed protections..

Jet Jockey A4

28th Jan 2015, 09:43

Thank you for the answer.

So in the end in an Airbus you can still fly it like a non FBW conventional aircraft if you required to do so regardless of which "Law" it's in or faults/malfunctions in "Normal Law" or am I and others missing something?

It's all down to basic pilot skills, which it seems a lot of new pilots don't have these days.

This should not be an Airbus versus Boeing debate then.

LiveryMan

28th Jan 2015, 09:44

...the other fairly recent "surprise" on the 320 family anyway, was water penetrating the seals on the AOA vane shaft and freezing the vane in some undesirable position, forcing the crew to select 2 of the 3 ADR off to place the system in alternate law, removing stall/overspeed protections..

I alluded to this earlier, the XL Air /Air New Zealand A320 crash off the coast of Perpignan. It pitched up into a nose high attitude, stalled and fell into the water below. This AirAsia flight seems to have made a carbon copy manoeuvre.

My suggestion that this crash might well be similar to that one was lost in all the Airbus v Boeing control philosophy discussion.

Volume

28th Jan 2015, 09:57

however are you telling me that if a pitot tube of some other probe (a la AF447) freeze up or just fails to function properly the FCC will still take its inputs and get confused instead of disregarding them and work in a "less than ideal" situation?As usual : it depends.
Normally a system should detect that the available data is corrupted/invalid/senseless/outside the norm, and hence should not get confused, but take the action it is designed for. Which might be switching off and leave it to the pilot to solve the issue. All input to systems should come from independent and redundant sources (e.g. 3 pitots feeding three air data computers), If one delivers data which is different from the other two, the system should understand that this one is at fault, and use the data from the others. If all three differ, the system should understand that no valid data exists, and act accordingly.
However, being designed by humans there might always be situations nobody ever thought of, and hence did not design the system for. This is valid for any complex product of any manufacturer.
See the Spanair accident in Madrid, where the very old fashioned pre-computer-systems got confused, and did not warn the pilot when having full authority to do something wrong.
Sometimes protections kill you, sometimes the lack of protections does. Sometimes systems safe the day. Some days they fail. Most important is that the pilot fully understands the systems in his aircraft and works with them, not against. A pilot which is not trusting his systems should not fly that aircraft.

NigelOnDraft

28th Jan 2015, 10:47

So in the end in an Airbus you can still fly it like a non FBW conventional aircraft if you required to do so regardless of which "Law" it's in or faults/malfunctions in "Normal Law" or am I and others missing something?
Not strictly true ;)

If you turn off AP/ATHR and FDs (in fact FDs academic) then you should be flying "attitude and power". However, if you are in Normal Law, then auto-trim and protections are still present, and you are not commanding control surface movements, but performance changes that the computers action.

In Direct Law the computers are still between the stick and controls, but in a fairly dumb way (stick relates to control surface movement).

Alternate Law somewhere between.

The "problem" some seem very concerned with is when a "protection" incorrectly kicks in, or limits you inappropriately, such it prevents you from flying normal attitudes. The recent OEB I believe, after many years of Airbus FBW Ops, is the first acknowledgement this can happen. Others here claim it has happened on numerous occasions.

I do not see that issue relevant to AF447. Without re-reading the whole thing again, I do not believe a "protection" incorrectly kicked in? Sensor failures led to FBW downgrades and allowed the pilots to fly the aircraft conventionally into an unrecoverable situation.

Perpignan ditto - again faults then downgrades allowed the pilots to crash the aircraft.

China 747 - yes he pulled 5g or whatever, bent the wings, and people say "couldn't do that in an Airbus - they'd all have died". With FBW working, the Airbus would have stopped the situation arising. Even if it got into that, I see no evidence the 747 recovery required 5g? A correctly flown 2g recovery would likely have been fine.

A Qantas A330, then the run up to the OEB, did see the aircraft behave inappropriately due FBW protections (as did a SQ777?). I suspect these are now designed out, or being so.

There are 000s of FBW Airbuses out there, and I doubt more accidents than say 737s? It is not perfect, but then nor is the 737. I very much doubt we will see a significant change (linked sidesticks, moving throttles, Law basics) requiring extensive hardware modification and re-training. I suspect we will see tweaks to the software and training to try and close off the loopholes as they arise.

Won't stop the Airbus bashers on here, who are now so desperate they blame the Airbus for the 777 SFO crash :eek:

I fly it, fine piece of kit, would prefer something "fun/real" to fly like the 757, (but get that outside the airline world). It's pretty reliable, albeit has it's French quirks. As is being repeatedly shown however, like all aircraft in the past, it is not crash (or idiot) proof.

RAT 5

28th Jan 2015, 11:01

If latest cockpit design would be significantly different from the last generation, pilots would have problems to change between the two, which would either limit the types they are allowed to fly (with all the organisational and financial impact) or it would create new risks.

You're, perhaps, forgetting history. In 1985 I was flying a very, really basic B732. The BY boys will remember. Then we became the lead airline for B767. Wow; this was a quantum leap. Fortunately we had a very strong pilot orientated Flt Ops management right up to MD level. They expected the pilots to be able to fly these new toys with the same competence as B732. We took it to some similar places; i.e. very basic Greek island airports = night circle to land at Corfu, or Kos or Heraklion etc. No problem. The culture was correct. My point is that such a mega change happened without a/c falling out of the sky and without the training department collapsing.

Most important is that the pilot fully understands the systems in his aircraft and works with them, not against. A pilot which is not trusting his systems should not fly that aircraft.

So how did we achieve this safe transition? Firstly the ground school was very thorough. The FCOM's had been simplified & diluted and CBT introduced. But, this was backed up at the end of every day with 'chalk & talk' FAQ's with a FE. During the CBT we used the sim as an FTD to explore and experiment with what we had learned that day. We came out with a healthy knowledge. We then went on the line and plugged in the automatics at 400' and watched what the a/c did; how it wanted to fly. We tried all the different methods via AFDS and learnt in depth how to fly & manage this new beast. The LTC/TRI/TRE's had a real indepth understanding and passed it on. When possible we hand flew it as that was the culture of the company. We'd learnt from the automatics what was necessary and then we applied it manually. Job done.

I then went to an airline who transitioned from B727 to B757. I took my BY philosophy with me, but had a culture clash. They insisted in a complete change and it became a "follow the FD" type operation with full automatics. The older guys flew it manually, but the newbies didn't know how and switching off the VNAV for descent planning was scary. Switching off the FD for departures from small empty airfields or approaches was even more so. Switching off the ILS on a CAVOK day was the final straw for some. All this was 25 hears ago.
I now fly B738 and HAL has sporned and migrated onto that fleet as well. I still fly my way and try to encourage others, but the ice is too thin and they don't want to venture out. OFDM is watching and they are afraid. When some do 'have a go' it might be a mess. They have a confidence blip and complain they don't get enough practice. True. I also teach TQ and they never had the skills in the first place via the syllabus. Hand flying on the line is discouraged and strict SOP's cause a trained monkey mentality. They know only a fraction of the capabilities of both the a/c and the systems.
You can only "trust your systems" if you understand them. Guys fly the a/c using the A/P looking only at the FD. It will always be centred, even when you stall or hurtle into the ground. It tells you nothing and can mask the truth of what is happening, i.e. attitude and V/S. In todays automated dependant operation the basic instrument scan, to confirm the FD is giving healthy guidance, is no more. It wasn't taught, save the raw data ILS, and it not used on line due to fully automatic operations. What chance have they got when the missing skills are needed? The way EASA is going I do not hold out any hope. Keep everything as cheap and as simple as possible just to be inside the safety envelope. Enough. Perhaps the days of single crew 99% automatic flight is not that far away. I'm sure the companies would love it. Any faults and the computer tells you how to fix it. That's not always a success if the computer is at fault. Bit like asking the police to conduct the enquiry into the police. The independent pilot needs to sort out the mess and decide if the computer solutions are correct: if not take over and save the day.
Ah, but that takes us back to the beginning and the circle is complete, or rather the spiral.

Jet Jockey A4

28th Jan 2015, 11:09

Not strictly true

If you turn off AP/ATHR and FDs (in fact FDs academic) then you should be flying "attitude and power". However, if you are in Normal Law, then auto-trim and protections are still present, and you are not commanding control surface movements, but performance changes that the computers action.

In Direct Law the computers are still between the stick and controls, but in a fairly dumb way (stick relates to control surface movement).

Alternate Law somewhere between.

I basically understand the 3 "Laws" of the Airbus.

Again I am asking, in "Normal Law" with all protections working, if you get caught in a 3000 feet/min updraft (for argument's sake) and you feel the aircraft is not reacting properly for what ever reason it (the aircraft's FCCs) think, can you just turn "OFF" the AP, AT and FDs and fly the darn thing to a pitch/attitude with a manual power setting on the engines to prevent or reduce the rate of climb or will the FCCs say tough luck pilot I'm still deciding what's best?

NigelOnDraft

28th Jan 2015, 11:18

RAT 5 - an excellent summary of where we are today :D People scared to manually fly, either through lack of familiarity, fear of QAR monitoring, get rusty..

Again I am asking, in "Normal Law" with all protections working, if you get caught in a 3000 feet/min updraft (for argument's sake) and you feel the aircraft is not reacting properly for what ever reason it (the aircraft's FCCs) think, can you just turn "OFF" the AP, AT and FDs and fly the darn thing to a pitch/attitude with a manual power setting on the engines to prevent or reduce the rate of climb or will the FCCs say tough luck pilot I'm still deciding what's best? If the Aircraft is serviceable, and you do not demand an "outrageous" attitude or speed, then yes - and indeed it is "Airbus Golden Rule 1" I think: If you don't like or understand what it's doing, take control.

I likely confused you with the earlier reply because the simple answer would be picked up by the Airbus bashers that in exceptional circumstances, the one word answer would not be correct ;)

DrPhillipa

28th Jan 2015, 11:23

I alluded to this earlier, the XL Air /Air New Zealand A320 crash off the coast of Perpignan. It pitched up into a nose high attitude, stalled and fell into the water below. This AirAsia flight seems to have made a carbon copy manoeuvre.

My suggestion that this crash might well be similar to that one was lost in all the Airbus v Boeing control philosophy discussion.

Is there any evidence of this aircraft also having been heavily pressure washed immediately prior to takeoff such that both AOA sensors were filled with water below the seals causing them to freeze at exactly the same time (at FL370+ ) and AOA and thus confuse the AC into thinking they were both still working?

LiveryMan

28th Jan 2015, 11:55

No, but replace pressure washers with rains aloft hitting a fuselage doing 300 something MPH and the effect should be similar

Jet Jockey A4

28th Jan 2015, 12:02

So that I understand this "autotrim" feature that keeps coming back in many conversations...

Do I understand this feature correctly as a pitch trim that is activated automatically whenever an extended pitch input (up or down) is commanded through the sidestick?

In other words if I command a pitch up from the sidestick, it trims in a nose up to relieve back pressure on the elevator not unlike when you fly a conventional aircraft in manual when using the electric trim or basically the same way an aircraft on AP would trim itself?

If I'm correct in my assumption as stated above? If so what is the big deal behind this feature on the Airbus?

Assuming the THS is trimming a nose up attitude because you are pulling on the sidestick, wouldn't the reverse also be true… that with a nose down input the THS would trim nose down?

Assuming you are in cruise at FL370 in an Airbus 320 and all is working properly, you then encounter some mountain wave turbulence or something similar to it, what usually happens at first in an increase in speed with a tendency in an increase in altitude usually compensated by the AP.

In the above scenario in a very conventional aircraft with a basic AP without AT, if left unchecked and if the power is not reduced, the aircraft may overspeed but it should hold altitude. In an aircraft with AT, the FADECs should automatically reduce to a power setting as to avoid overspeed.

However it seems from reading on here that the Airbus might want to pitch up to avoid overspeeding, is this correct? At what point will the Airbus reduce power to keep it within the normal parameters?

Back to AF447 and the THS being full nose up. If the pilot would have just inputed a full nose down on the sidestick, wouldn’t the THS trimmed nose down too?

SRS

28th Jan 2015, 12:23

surely attitude must be set in addition to the 90% N1. On a Boeing it is safe to use 2-3 degrees nose up in cruise. Must be the same for Airbus

jimjim1

28th Jan 2015, 12:51

it trims ... to relieve back pressure on the elevator Yes.

If I'm correct in my assumption as stated above? If so what is the big deal behind this feature on the Airbus?There is no deal at all.

Assuming the THS is trimming a nose up attitude because you are pulling on the sidestick, wouldn't the reverse also be true
that with a nose down input the THS would trim nose down?Yes.

Back to AF447 and the THS being full nose up. If the pilot would have just inputed a full nose down on the sidestick, wouldn’t the THS trimmed nose down too? Yes.

NigelOnDraft

28th Jan 2015, 12:53

Do I understand this feature correctly as a pitch trim that is activated automatically whenever an extended pitch input (up or down) is commanded through the sidestick?I believe it essentially alters the "THS" (Horiz Stab) to automatically align with the elevators - much as can be done on a 1950s jet fighter ;) So as you say:In other words if I command a pitch up from the sidestick, it trims in a nose up to relieve back pressure on the elevator not unlike when you fly a conventional aircraft in manual when using the electric trim or basically the same way an aircraft on AP would trim itself?
If so what is the big deal behind this feature on the Airbus?None whatsoever to you & me - but to the Airbus bashers, it means it was the Airbus' fault that when a pilot held the stick back and flew to a crazily low speed (AF447 & Perpignon), and the aircraft duly trimmed back, it meant when the pilot let go and/or belatedly tried to recover, it was trimmed back in a way that did not assist recovery.

There is the aspect, which is true, that if the aircraft then drops into Direct Law (as happened in Perpignon), the auto-trim stops, so your:If the pilot would have just inputed a full nose down on the sidestick, wouldn’t the THS trimmed nose down too?would not happen in this case. In AF447 I don't think it dropped to Direct law, so yes, stick forward and trim forward would have happened...

Assuming you are in cruise at FL370 in an Airbus 320 and all is working properly, you then encounter some mountain wave turbulence or something similar to it, what usually happens at first in an increase in speed with a tendency in an increase in altitude usually compensated by the AP.

In the above scenario in a very conventional aircraft with a basic AP without AT, if left unchecked and if the power is not reduced, the aircraft may overspeed but it should hold altitude. In an aircraft with AT, the FADECs should automatically reduce to a power setting as to avoid overspeed.

However it seems from reading on here that the Airbus might want to pitch up to avoid overspeeding, is this correct? At what point will the Airbus reduce power to keep it within the normal parameters?To be precise, it does not pitch up to prevent overspeed, but minimise the extent when it has oversped. First the AP disconnects, then further into overspeed it pitches up a small amount (which you can override) until eventually even full forward stick will not counteract it. It does not reduce power - albeit if ATHR is on, that should have selected idle.

The ATHR is not great, and in Mountain Waves it can be almost hilarious as speed oscillates between Min and Max, and Power between full & idle, but out of phase. Disconnect ATHR, set sensible power and leave, and the speed oscillations are far lower :ok:

I suspect it would be fair to say that in the well known overspeed events, poor monitoring or inappropriately high cruise speeds may have been factors. Certainly my experience on the line is poor choice of speeds with regard to conditions - any excuse to go faster seems the aim :sad:

Volume

28th Jan 2015, 13:02

Do I understand this feature correctly as a pitch trim that is activated automatically whenever an extended pitch input (up or down) is commanded through the sidestick?Not only for extended input. Any sidestick pitch input commands g´s, or a change in the flight path vector (it needs a normal acceleration to change the flight path vector). Releasing the stick does not mean you return to your previous trimmed pitch, but you maintain the new one. This is achieved by use of the trim.
To oversimplify a little bit, the Airbus Sidestick is the same as the trim switch on some conventional aircraft. Which is exactly what you need for a large transport aircraft. You need the stick to rotate, to level off at cruise altitude, to initiate the descend and to level of for touchdown. In the optimum situation, two of these actions are done via the A/P selector panel, so you basically need the elevator control twice per flight (both times in the nose up direction). So for some pilots that is all they have done with the stick in the recent years...

Back to AF447 and the THS being full nose up. If the pilot would have just inputed a full nose down on the sidestick, wouldn’t the THS trimmed nose down too? Unless that would have resulted in an overspeed situation, and hence the system would not have allowed it, yes. So maybe it would have required to throttle down or use the speed brake until the system would have accepted this input.

LiveryMan

28th Jan 2015, 13:14

None whatsoever to you & me - but to the Airbus bashers, it means it was the Airbus' fault that when a pilot held the stick back and flew to a crazily low speed (AF447 & Perpignon)

The pilots @ Perpignan were pushing their sticks forward once the upset because apparent, not back. They apparently had not seen the "USE MAN TRIM" warning on centre display. The AP had commanded full nose up trim just prior to that, causing the elevator deflection the control sticks could commands to be too small to recover the aircraft before it stalled.

I'm not blaming or bashing anyone or anything, just stating it as it was.

The extreme nose up attitude and then sudden drop toward the ocean AirAsia is tracked to have done is remarkably similar to the Perpignan crash, just higher in altitude.

Jet Jockey A4

28th Jan 2015, 13:19

Thanks to all for the answers.

Personally I don't see anything wrong with this autotrim function...

Whenever the Airbus pilot inputs a command via the sidestick when flying the aircraft manually, it helps relieve the pressure on the elevator in the same way on a conventional aircraft one would have to use the electric trim (or manual trim) to relieve the pressure and this is not unlike the autotrim feature on aircrafts I flew when the AP is engaged it's just the Airbus takes it a step further and does it when being flown manually.

RAT 5

28th Jan 2015, 13:22

This is not Airbus bashing; it's not even FBW bashing, but....

There have been so many threads on here, before and after incidents/accidents, that I am curious about one point. The discussions about what might have happened, what did happen, what the pilot should have done, why they didn't do it, why they did something else or nothing, why they were confused etc. etc. mostly there has been no clear concise answer to these queries; and many AB training captains have been involved in the debate. There have been answers such as " it did exactly what it was supposed to do, but they didn't realise it."
The discussion going on now is how a pilot can take over and fly the a/c in the most basic manner and what its capabilities will be when you do so. And to a non-AB driver it is not clear. I really wonder if you put 100 experienced AB plots in a room, and through these type of questions at them, how many of them would come up with the correct answers. If less then 90 that is very worrying, and why do I expect that to be the case? Am I wrong? And how many would not understand the question?

Capn Bloggs

28th Jan 2015, 13:22

To be precise, it does not pitch up to prevent overspeed, but minimise the extent when it has oversped. First the AP disconnects, then further into overspeed it pitches up a small amount (which you can override) until eventually even full forward stick will not counteract it. It does not reduce power - albeit if ATHR is on, that should have selected idle.

Say what? :confused:

Back to AF447 and the THS being full nose up. If the pilot would have just inputed a full nose down on the sidestick, wouldn’t the THS trimmed nose down too?
Unless that would have resulted in an overspeed situation, and hence the system would not have allowed it, yes. So maybe it would have required to throttle down or use the speed brake until the system would have accepted this input.
Overspeed? Correct me if I'm wrong, but when the sidestick was pushed forward the speed increased to above 60KIAS and then the Stall Warning came back on. What sort of feedback message does that send? :cool:

Autotrim when handflying is bad news; nice and geeky when all is well but precisely what you don't want when things go pear-shaped. Either you're fully flying it ie trimming as well or the AP is in.

jientho

28th Jan 2015, 13:35

Just to be absolutely clear, with FBW you can never fly it completely "like" a non-FBW aircraft. There is simply no physical linkage there to fall back to. There is only a wire (wiring) electronically communicating intended control movements to control surface actuators.

Of course in Direct Law the intention is that pilot inputs are directly and immediately communicated as-is to the actuators. (And maybe that is what was meant by "like" above.) But technically it is still a simulation of a direct physical link (something like controlling avatars in video games).

Not that there is anything wrong with that, by any means. Just that it is different. With differing failure modes, for instance.

LiveryMan

28th Jan 2015, 13:41

Jet Jockey

The AOA Sensors were found to have frozen in place at a relatively flat angle. (due to the plane being washed with a pressure hose, allowing water to get into the workings of the AOA sensors which then froze at altitude)

The crew initiated a near stall moment as part of their flight tests to make sure the AP did it's flight envelope protection job.

Aircraft pitched up as it's supposed to, but due to the stuck AOA sensors, it thought it was still on the level and still slowing, so commanded the nose further up. By the time the crew realised something wrong, the AP had given up, given them control and threw up the "USE MAN TRIM" warning.

As I understand it, with the stabiliser in full nose up trim, and with the low airspeed, the elevators did not have enough authority in the time given to recover the plane before it stalled.

The investigation noted that the pilots were most likely relying on the plane fixing the situation. But when it called quits, it left them with a plane about to fall out of the sky.

ManaAdaSystem

28th Jan 2015, 13:50

I don't fly Airbus, but I find it a bit strange that an aircraft with so many protections built in, will allow the autotrim to trim the stab into a position that you simply can't fly out of.

Capn Bloggs

28th Jan 2015, 14:00

Mods are quick...

Is there not enough elevator/stab authority in the Airbus to unstall it assuming it was fully stalled by pushing the nose down even if fully trimmed nose up?
In my "normal" aeroplane, I suspect that if stalled with full nose-up trim, I could imagine that I may not be able to get out of it by simply pushing full forward; I suspect that there may not be enough elevator authority, hence the comment about using the Stab trim to get the nose down in the stall recovery procedure. Pushing and trimming, of course, is second-nature to me because that's the way you hand-fly my aeroplane. I would expect that in a full stall, trimming forward (via a convenient pair of buttons on the control column) would also be second-nature.

KatSLF

28th Jan 2015, 14:15

I am SLF but I have written numerous computer programs containing sanity checks on inputs.

It seems to me there two main ways to make a plane go faster -- nose down or extra thrust. In both cases the speed builds up, doesn't jump 50 or 100 instantly, right? and there are duplicate/triplicate inputs in the knowledge that they may develop faults?

Looks to me they've written it for ideal conditions, assuming the inputs are all true, when the triplications show they know full well they may not be.

So if the nose is more or less level and thrust has not been increased, why on earth should it think it has started to overspeed in the space of 2 or 3 seconds?

Why can't the computer be told to STAY CALM in any situation when an impossibly fast change of speed (or altitude I guess) seems to have occurred? ideally also tell the pilots there are input discrepancies, to get them alert and ready, just in case.

paull

28th Jan 2015, 14:22

After AF447 did Airbus modify the software for loss of speed sensors?

If the aircraft goes from 3 consistent indications to 2, to all different, then why cannot it not provide a warning along the lines of:

"Loss of one speed indicator, if another is lost then the AP will drop , set pitch and power and see you later"

Of course, it may be that you go immediately from 3 to none in which case there is no time for a long speech but even saying "AP dropout due to unreliable speed indications", gives you less surprises.

We must surely have the data to see how long it takes for loss of one speed indicator to all three inconsistent, who knows, it might even provide enough warning to start unloading the trim as well. What does the data say?

captains_log

28th Jan 2015, 14:27

Which Is what a few including myself have been saying:

http://www.pprune.org/8824575-post2010.html

Quoted below:

It certainly didn't help 2 very respectable pilots on D-AXLA A320-232 over Perpignan when THS stayed nose up, with the combination of the AOA sensors freezing and normal law dropping out they were unable to save it with the above aforementioned method (stick hard forward). After reading the report, i was worried by the lack of notification to the pilots, in that moment id expect 'USE MAN PITCH TRIM' to be slapping me round the face. This warning in that instance was possibly the most understated warning.

Why not use a clear indication:

Warning - Normal law FAIL, Law STATUS NOW Alternate/Direct etc
Cause - AOA/ADIR mismatch(or the like)
Result - USE MAN PITCH TRIM(?(perhaps added flashing audible warning if stick forward/backward exceeds Xseconds)) or AUTO PITCH TRIM DISABLED!

(In the tragic Perpignan instance above i believe the PITCH TRIM warning dropped off the display shortly after changing flight law.) i stand to be corrected. Which then left the guessing game of 'what did the AP change we haven't worked out yet'.

Smilin_Ed

28th Jan 2015, 14:29

One of the first things I was taught in primary training is, "Don't trim into a stall." Apparently, the AB system will do just that in some circumstances.:(

Jet Jockey A4

28th Jan 2015, 14:32

Overspeed? Correct me if I'm wrong, but when the sidestick was pushed forward the speed increased to above 60KIAS and then the Stall Warning came back on. What sort of feedback message does that send?

Autotrim when handflying is bad news; nice and geeky when all is well but precisely what you don't want when things go pear-shaped. Either you're fully flying it ie trimming as well or the AP is in.

Well I think I understand why the Stall Warning came back on when the airspeed increased past 60KIAS and if so it's makes perfect logical sense.

However I think Airbus should have programmed their software better so that the aircraft's system knew they were in the air in a very unusual
circumstance (low speed) but at altitude. Basically I think the system is programmed by logic that if it sees a speed below 60KIAS it is silenced.

I still can't fathom why the crew could not understand (at least initially when the captain wasn't in the cockpit) in which state of flight their aircraft was in.

I don't care if the stall warning was going on and off several times which is a major clue in itself when your airspeed is at or near zero, your attitude is wrong, your bloody VSI is showing a rate of descent near 10,000 feet/minute (or whatever is the max rate is on the Airbus) and with the altimeters showing you that you are losing altitude at a very high rate and speeding towards the surface of the Earth or ocean in this case... All of those clues are a tell, tell that you are in a stall!

Autotrim when hand flying is a bad thing, not necessarily imo!

I see it as an option, a built-in option and I think I know why Airbus included it. I think they decided this in the initial phase of development... Why not have Autotrim when the pilot is flying it manually since we have it (as in most other aircrafts) when the AP is controlling the aircraft. It would be just a nice feature and relieve another action (manual/electrical trimming) from the piloting. I'm sure some of the very old timer said the same thing about autopilots, flight directors, FMS systems etc when they first appeared.

Now not knowing the Airbus, it would perhaps be better if not only a visual warning appeared but also an audio warning like "Autotrim Fail" or "Use/Revert Manual Trim".

Same thing could/should be done with both visual and audio warnings when the aircraft reverts from "Normal Law" to "Alternate or Direct Law".

This would perhaps help the situation in a critical phase of flight.

SAMPUBLIUS

28th Jan 2015, 14:44

So if the nose is more or less level and thrust has not been increased, why on earth should it think it has started to overspeed in the space of 2 or 3 seconds?

Seems to me some sort of ongoing comparison between flight computer responding to various sensors (AOA, Airspeed, etc ) and an Inertial reference system would immediately reveal such a difference.

And while one cannot fly for long using an inertial reference system due to lack of true airspeed data, the few seconds with a large delta between the systems could trigger a simple warning. YO dude- Garbage in- Garbage out !!

A body in space cannot change in any direction without some sort of acceleration/deceleration vector as measured by precision accelerometer /gyroscopes. So why isn't such info used ?? In days of yore- it was a simple gyroscope to determine attitude , etc and a pendulum device/equivalent to determine certain vector accelerations.. . . :confused:

NigelOnDraft

28th Jan 2015, 14:52

Now not knowing the Airbus It would perhaps be better if not only a visual warning appeared but also an audio warning like "Autotrim Fail" or "Use/Revert Manual Trim".When you go into Direct Law:

USE MAN PITCH TRIM on PFDs
ECAM message "F/CTL DIRECT LAW"
Above message includes "MAN PITCH TRIM" message
ECAM message will have associated audio alert
However, as with any ECAM message, it is prioritised according to what other messages / alerts are being triggered.

One of the first things I was taught in primary training is, "Don't trim into a stall." Apparently, the AB system will do just that in some circumstances.An Airbus is slightly different to your primary trainer. Firstly, it is not "trimming" as such, but aligning THS and Elevator to retain full elevator control. There are circumstances you need to fly, and retain control at low speed: landing / GPWS / Windshear, and retain control to manoeuvre and counter rapid trim changes (e.g. full power).

So if the nose is more or less level and thrust has not been increased, why on earth should it think it has started to overspeed in the space of 2 or 3 seconds?Because in the cases where it has "pitched up", AFAIK they were genuine overspeeds. The aircraft does not need to accelerate to go from cruise speed, to significant overspeed.

A body in space cannot change in any direction without some sort of acceleration/deceleration vector as measured by precision accelerometer /gyroscopesAs above, you are misunderstanding basic aerodynamics. Overspeed refers to IAS/M number - nothing to do with KE / Ground Speed.

Jet Jockey A4

28th Jan 2015, 15:13

One of the first things I was taught in primary training is, "Don't trim into a stall." Apparently, the AB system will do just that in some circumstances.

That is why on all our recurrent training sessions we go through 4 types of stall exercises and they are done all with the AP on so that you do get a nose high trim situation because the AP will trim until the shaker is activated.

Stall exercise #1: 10,000 feet, clean, then slats out only, then slats + flaps at the two takeoff settings, then slats + full flap. One of these will also be done with a continuous 20 degree bank turn.

Stall exercise #2: Low altitude stall in a landing configuration (full flaps), again with the AP on. While levelling off on an approach at the MDA (400' AGL), the trust levers are brought back to idle where they stay until stick shaker/AP disconnect.

Stall exercise #3: High altitude (high 30s) with a heavy aircraft.

Stall exercise #4: Takeoff with an engine failure with a climbing turn. The AP is selected on at the proper altitude and a 15 degree bank continuous turn is initiated using the VS mode and set a high rate of climb until stick shaker/AP disconnect and recover.

silverstrata

28th Jan 2015, 16:20

Ben There:

Setting 90% N1, on the A320, will assure a safe airspeed (assuming you're not approaching a stall)

An even better strategy, is to select 80% and a degree or two of attitude below normal cruise.

What you want to do is get out of coffin corner, especially if things are getting turbulent, and by setting a lower thrust and attitude you will do just that.

Manaada:

I don't fly Airbus, but I find it a bit strange that an aircraft with so many protections built in, will allow the autotrim to trim the stab into a position that you simply can't fly out of.

So will the Boeing, as was amply demonstrated on an approach into a European base a few years back, when the pitch got out of control.

Unfortunately, there are still some corners of the envelope that will always remain outside controllability. You want a trimmer, but that trimmer can also reduce elevator authority (as it can on the Boeing). You want lots of thrust, but that thrust can reduce elevator authority (as it can on the Boeing). You want rudders, but that yaw can diminish roll control (as it can on the Boeing).

With an aircraft, both in design and in flight, you are juggling any number of interconnected and competing controls and parameters - any or all of which can be present at the same time. And from that melee of actions and reactions you need to create stable flight, which is not always possible at the far edges of the envelope. So the idea is (and always has been) to remain within the established safe envelope, where all of those many competing aeronautical parameters can coexist and operate in safe harmony.

The problem lately, is too many aircraft exploring the edges of the envelope where they should not even dream of going. So rather than fiddling with the edges - changing this law or that law - we need to concentrate more on not letting aircraft get to the edges and corners of the envelope where they have absolutely no business being.

One minor mitigation would be to take fuel restrictions off pilots, so they are not flying in the higher corners (and adjust the pax figures to compensate). But no company will do that....

mtwittm

28th Jan 2015, 16:46

Maybe we need Lawyers instead of Pilots to fly these airbus aircraft in order to interpret - Direct Law, Alternate Law...etc.

Leightman 957

28th Jan 2015, 17:11

Nigel@2677: "An Airbus is slightly different.............it is not "trimming" as such, but aligning THS and Elevator to retain full elevator control."

But "full Elevator control" at an AOA (and airspeed, temperature, pressure etc) presumed by the computer to be correct, along with previous pilot inputs. I thought that was exactly what happened in the case where excessive (and unnoticed?) FCC-directed the THS trim to alter to the point where elevator authority was insufficient.

If the above is incorrect, are you saying that the FCC will correctly notice and interpret unusual variations/combinations in AOA, OAT, pressure, and airspeed over a time period that varies from nearly instantaneous to several seconds, and during that time correctly interpret that a sensor set is beginning to disagree? If you are, the system is pretty smart. But if the system can't interpret all those inputs and correctly categorize them as this or that type of situation where action A is the best, the pilot has lost 5 or 10 or 20 seconds while the computer thought, then annunciated, and then the pilot began to redirect his attention from wherever it was.

I'm getting back to the time involved in the FCC process here. Explanations by posters about the FCC logic chain haven't mentioned time. The system must ignore (or filter) instantaneous changes as either transitory or anomolous, so that means that a certain number of samples over time must be taken for all sensory inputs. Can someone knowledgeable speak to the amount(s) of time the system consumes before it acts in various ways?

RE AOA: Because AOA failure seems to have recurred, do not (or could not) the sensor be mechanically exercised to confirm its movement is free? This need not necessarily involve actual movement but could be forced applied to a pressure transducer. Or is this already built in?

Rananim

28th Jan 2015, 17:16

One of the first things I was taught in primary training is, "Don't trim into a stall." Apparently, the AB system will do just that in some circumstances

Autotrim when handflying is bad news; nice and geeky when all is well but precisely what you don't want when things go pear-shaped. Either you're fully flying it ie trimming as well or the AP is in.

100% correct.Strange counter-intuitive design.I dont like it.And I dont need to fly it to know that.In normal flight,pilot doesnt trim so becomes accustomed to not trimming.When something happens,the computer has trimmed into the stall and then the computer leaves it to the pilot to manually trim.No,its not right.No pilot would ask for that design.Keep it simple.Human brain doesnt work well with startle factor and lots of ECAM warnings blaring.

Coagie

28th Jan 2015, 17:18

I think how QZ8501 and AF447 got into stalls and attempted to recover will be different. The pilot flying in the AF447 crash, Bonin, makes it difficult to compare it to other air accidents. Reading the CVR transcript, makes it clear to me, even through translation from French to English, that Bonin was very uncomfortable, even frightened, well before the UAS incident even took place. He was spooked (frightened) by his first exposure to St Elmo's Fire and the the "electrical smell" AF447 encountered. Most, especially when the phenomena were explained to them, would think "That's cool" or "neat", but Bonin reacted as if they were supernatural, as if he'd seen a ghost. He may have even known something about what to do in a stall situation (wasn't one of his hobbies flying gliders?), but he was already in "fight or flight" mode (wrong kind of "flight" mode. No pun intended), and using his reptilian brain instead of his reasoning brain. He didn't think things through or go back to his training. Each person is different, but Bonin "cracked up". Some people would use the term, "freaked out". The threshold to "crack up" is different for everyone, but can change with training and experience. This is why some training/testing involves putting the student in pressure (stressful) situations, to see if they can still perform. Some people perform and pass testing fine, when not under pressure, only to falter, when pressure is applied. Military training is often known for this, but so are some civilian training regimens. I think the amount of "Pressure Testing" and "Trickery Testing" in airline pilot training and testing, needs reviewing and revising to better insure, that a pilot's "cracking up" threshold isn't too low. Some schools and airlines may already have it "right" and some may have it "wrong" or some degree in between.
Bonin had a low "cracking up" threshold in the situation he found himself in, and since every person can be vastly different, it may be apples and oranges to compare his incident and the Air Asia QZ8501 incident, as far as how they got in a stall, and how they attempted to get out of it. I think his "cracking up" threshold just happened to be lower than an average person, and, I imagine, much lower than an average pilot. I can't bring myself to blame the computer in the AF447 crash, as it didn't throw an unmanageable situation in Bonin's lap. He created the unmanageable situation. It could prove different in QZ8501
Anyway, hopefully, we'll see, when an Air Asia QZ8501 CVR transcript is published. I don't think the chances are that another pilot, with a low cracking up threshold, slipped through the cracks and caused a crash in the same way. Hopefully, I'm not wrong, or things are worse than I thought.

Coagie

28th Jan 2015, 19:12

"Coagie

You aren't suggesting actual, all round air experience might even count towards the selection criterion now?? Goodness gracious me, what are you on?"

Sop_monkey,
It's the madness flaring up. Forgive me.

glendalegoon

28th Jan 2015, 19:19

cnn just reporting and will comment in minutes about finding 2 bodies 600 mies from crash site.

NigelOnDraft

28th Jan 2015, 19:54

One of the first things I was taught in primary training is, "Don't trim into a stall." Apparently, the AB system will do just that in some circumstances Having reviewed the FCOM in fact the auto-trim stops as AoA protection kicks in i.e. well prior the stall (providing AoA protection is working of course...)

Neptunus Rex

28th Jan 2015, 20:03

Without flotation assistance, such as an inflated life vest, a human body immersed in water will sink. It then takes at least 48 hours for the decomposing body to produce enough internal gases for the body to rise to the surface. During this process, as well as afterwards, the body is subject to any local currents.

As the accident occurred on 28 DEC 14 and these two bodies were found on 27 JAN 15, 29 days have passed, which amounts to an average movement of under one knot for the whole period.

"Do the Maths."

Perfectly explicable.

enola-gay

28th Jan 2015, 21:29

Regular as clockwork, someone pops up to say" Wait for the official report. Stop speculating"

Aye just like the Chilcott inquiry!

If you read the blog title it is called "Rumours and News", so I presume people are allowed to post published news, speculate about its veracity and consider all the rumours that abound?

If not change the blog title to "Facts Only":=

mickjoebill

28th Jan 2015, 22:02

Numerous posters have pointed out that from their first hand experience there is a significant difference between severe turbulence in real life compared to the comfy environment of the sim.

Should task specific "turbulance and stall sims" be used?
They need only have essential cockpit instruments to reduce building and running costs. One version for bus and a second for boeing.

This bare bones, one size fits all generic cockpit concept has proven successful for HUET training.

Is there enough data from recent stall spins or severe turbulance to replicate the physical environment?

Rather than a full size cockpit on multiple actuators a smaller capsule on the end of a robotic arm is able to generate the G forces in any plane, like these F1 and Diamond aircraft sims.
German researchers claim first: Robotic flight simulator - AOPA (http://www.aopa.org/News-and-Video/All-News/2013/April/3/German-researchers-claim-first-Robotic-flight-simulator)

Giant robot arm used as F1 simulator - CNET (http://www.cnet.com/uk/news/giant-robot-arm-used-as-f1-simulator/)

AirScotia

28th Jan 2015, 23:58

Numerous posters have pointed out that from their first hand experience there is a significant difference between severe turbulence in real life compared to the comfy environment of the sim.

Many pilots here have reported how difficult it is to see the instruments in severe turbulence conditions. Do simulators adequately replicate this effect?

Machinbird

29th Jan 2015, 00:09

As mentioned yesterday, the FAC computers are being considered as a possible causative element in the QZ8501 accident. Leaked information from those who do not wish to be identified has indicated this area of interest by the investigators. QZ8501: Probe team vets possible computer glitch, crew response | New Straits Times (http://www.nst.com.my/node/71095)

"The aircraft remains fully controllable if you lose the two FACs,” an Airbus spokesman said by email. This may be dissembling however. Suppose the FACs remained engaged in Normal Law and failed to detect a failure of that system. Could the FAC system then activate overspeed protection without an actual overspeed?

This need not be a design issue, it could also be a maintenance issue. In any case, if the DFDR data continues to point to the FACs, they should probably salvage the cockpit and recover those boxes for examination.

A320FOX

29th Jan 2015, 00:50

The FAC "boxes" are not in the cockpit, they are under the forward galley in the E & E compartment.

scard08

29th Jan 2015, 01:47

(It would also seem that Airbus now has at least two sets of pretty good data on entry to stall, stall behavior, transition to deep stall, flat spin, 10,000+ per minute altitude loss.... to add to their simulator's knowledge base.)I was thinking about this too. Does the FDR from a crash that involves a stall help improve the simulated responses to a stall?

The answer is obviously yes, but how many sets of data about performance outside the envelope (i.e. flights) do you need before you can program a simulator to correctly simulate performance there? What does it take to make simulations significantly better? My gut feeling is "a lot".

Any folks who build simulators out there? Any test pilots who can speak to this?

HeavyMetallist

29th Jan 2015, 02:07

This may be dissembling however. Suppose the FACs remained engaged in Normal Law and failed to detect a failure of that system. Could the FAC system then activate overspeed protection without an actual overspeed?

This need not be a design issue, it could also be a maintenance issue. In any case, if the DFDR data continues to point to the FACs, they should probably salvage the cockpit and recover those boxes for examination.
Apart from the fact that the FACs aren't where you think they are, I think you'll also find that the ELACs and SECs are the flight control computers that apply the control laws, and not the FACs.

Propduffer

29th Jan 2015, 03:07

News from the data and voice recorders is apparently leaking out.

The WSJ has a story including this:
They have said icing of speed sensors, a phenomenon that contributed to numerous fatal airliner crashes and serious incidents over the years, doesn’t appear to have been a factor this time.
AirAsia Co-Pilot Was Likely at Helm Before Crash - WSJ (http://www.wsj.com/articles/airasia-co-pilot-was-likely-at-helm-before-crash-1422493886)

RifRaf3

29th Jan 2015, 03:09

Notice in the video that it's rotating at about 1 rev/sec to simulate G. If you are pulling straight out of a dive you are not rolling at 360 degrees/sec, so you gain one simulation aspect (sustained G) at the expense of another, namely rotation on another axis. Only a very large, 3D printer style, framework could do a reasonable job of isolating each axis of rotation and the cost would be prohibitive.

Machinbird

29th Jan 2015, 04:23

Quote:
Originally Posted by Machinbird http://www.pprune.org/images/buttons/viewpost.gif (http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-135.html#post8844238)
Suppose the FACs remained engaged in Normal Law and failed to detect a failure of that system. Could the FAC system then activate overspeed protection without an actual overspeed?

This need not be a design issue, it could also be a maintenance issue. In any case, if the DFDR data continues to point to the FACs, they should probably salvage the cockpit and recover those boxes for examination.

Apart from the fact that the FACs aren't where you think they are, I think you'll also find that the ELACs and SECs are the flight control computers that apply the control laws, and not the FACs.HeavyMetallist, I'll have to defer to your experience on the actual physical location of the FAC computers.:O
Guess they will have to grab the piece from ahead of the wing (which may also include the cockpit).

As to the exact manner the FACs interface with the ELACs and SECs, my reading of FCOM indicates that it is the FACs that have the function of monitoring the flight envelope and invoking the protections (among other functions). The ELACs and SECs would then implement the response. Take a look at FCOM 1.22.40 Auto Flight-Flight Augmentation-and see if you read it that way.

B4MJ

29th Jan 2015, 05:43

Has anyone ever explained why AF447 PF obsessively pulled back?

Survey some old experienced simulator instructors. Ask how many times they've observed pilots pull back (even obsessively) on the yoke/stick when they're suddenly startled (frightened) by an upset.

Human nature isn't going to change and until upset training becomes really realistic (and everyone is forced to receive it), this type of event will always be lurking.

foxcharliep2

29th Jan 2015, 06:19

When AirAsia Flight QZ8501 crashed, the co-pilot was flying the plane. The pilot was monitoring the flight. And things may have gone wrong in a span of just three minutes and 20 seconds.That's the assessment of Indonesia's National Transportation Safety Committee.

Investigator: Co-pilot flew AirAsia Flight QZ8501 - CNN.com (http://edition.cnn.com/2015/01/29/asia/airasia-disaster/index.html)

toffeez

29th Jan 2015, 06:25

So the capt just sat there for 3 minutes while they were out of control and didn't press the takeover button.

formationdriver

29th Jan 2015, 07:02

Volume

29th Jan 2015, 07:25

but I find it a bit strange that an aircraft with so many protections built in, will allow the autotrim to trim the stab into a position that you simply can't fly out of. Since aircraft have grown in size beyond what the pilot can control by his own power, trim has always been a feature that can kill you if wrongly used. This was one of the reasons why the L1011 did not have a trimable horizontal stabilizer with an elevator, but an all flying tail. It was eliminating a possible risk. Due to the complexity of that type of pitch control, the concept was abandoned in favour of cheaper, lighter and simpler systems. Stabilizer trim requires appropriate trim setting, pilots and computers can mess that up (and did in the past). Autotrim is always designed for a certain purpose, and works perfectly for that. If you operate out of the design assumptions, the system can bite you. Even with autotrim installed, it is still the pilot who is responsible for the correct trim setting. Autotrim should assist him, not dominate him. It is the pilots fault if he lets it happen. It is the role of the pilot to check what his systems are doing. All the indications and warnings are there right in front of his eyes. The pilot should just not stop monitoring, which has never failed in all of his life...

Has anyone ever explained why AF447 PF obsessively pulled back? I think this is pretty self explanatory for anyone who is able to fly an aircraft and read FDR plots in the report. The pilot did not obsessively pulled back unless the aircraft nose dropped by some 25 degrees (between 2:11:45 and 2:12:00, the normal behaviour of a stalled aircraft). However the normal behaviour of a trained pilot should have been different... But for a pilot in panic, pulling when your pitch is -10° is understandable, but should not happen. Especially if a synthetic voice is trying to point out an important detail for about a minute already...

Sop_Monkey

29th Jan 2015, 08:30

When you have an A/C manufacture boasting that one of their latest models are "uncrashable" I think it will happen again but hope I'm wrong

This attitude and arrogance has astounded me ever since.

HeavyMetallist

29th Jan 2015, 09:12

Quote:
Originally Posted by Machinbird http://www.pprune.org/images/buttons/viewpost.gif (http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-135.html#post8844238)
Suppose the FACs remained engaged in Normal Law and failed to detect a failure of that system. Could the FAC system then activate overspeed protection without an actual overspeed?

I think I misunderstood what you were suggesting had failed; by "failure of that system" I thought you were referring to the system(s) that was applying Normal Law,whereas I now think you're referring to the FAC not detecting a fault within itself? So your scenario would be:

ELACs/SECs working normally, in Normal Law

Controlling FAC suffers a failure, but doesn't detect that failure
As a consequence of the failure, the controlling FAC signals an overspeed condition
ELACS/SECs carry out pull-up manoeuvre

Quote:
Originally Posted by Machinbird http://www.pprune.org/images/buttons/viewpost.gif (http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-135.html#post8844238)
As to the exact manner the FACs interface with the ELACs and SECs, my reading of FCOM indicates that it is the FACs that have the function of monitoring the flight envelope and invoking the protections (among other functions). The ELACs and SECs would then implement the response. Take a look at FCOM 1.22.40 Auto Flight-Flight Augmentation-and see if you read it that way.
I agree with your interpretation.

RifRaf3

29th Jan 2015, 10:06

5.5g, yes, but without accompanying wings

Ian W

29th Jan 2015, 10:29

Interesting timing: FO's family sues the airline (for dangerous & illegal operations) two days before Airline says CVR proves he was the Pilot Flying at the time of the accident. The airline also says 320's FDR shows no mechanical,electrical or weather-related-related malfunctions before steep climb, subsequent loss of control, and crash. The family's suit is based on "illegal operation" ( flying without a route permit on that day), "overcrowded skies" "excessive ATC workload." Lawyers... !!

France-La famille du copilote d'Air Asia porte plainte | Reuters (http://fr.reuters.com/article/frEuroRpt/idFRL6N0V633920150127?pageNumber=2&virtualBrandChannel=0&sp=true)
AirAsia : le copilote français pilotait l'avion au moment du crash (http://www.lefigaro.fr/international/2015/01/29/01003-20150129ARTFIG00065-airasia-le-copilote-francais-pilotait-l-avion-au-moment-du-crash.php)

The family's suit does not appear to have any merit. The area was not overcrowded, it was standard delay in clearing an aircraft for higher and in any case higher would make an aircraft more susceptible to severe turbulence. This was just luck-of-the-draw finding an active possibly 'dry' cell at the wrong moment.
However, not being an engineer, you probably did not note the items missing in that carefully enumerated list which I have bolded: Software, algorithm, system design are not mentioned. It is possible that all the protections worked as precisely as designed but this is a particular rare occurrence of a 'bus zoom climbing, as the WSJ reports, because the systems worked as designed.

jientho

29th Jan 2015, 11:34

NAROBS, you are confusing speed with acceleration. 55 m/s is speed. You seem to be assuming 55 m/s/s (=~5.5g). That would only be the case if the change in vertical speed all occurred in one second.

klintE

29th Jan 2015, 11:38

AirAsia Flight 8501: Co-Pilot Was Flying Plane at Time of Crash, Officials Say - ABC News (http://abcnews.go.com/International/airasia-flight-8501-pilot-flying-plane-time-crash/story?id=28570604&cid=fb_abcn)
Siswosuwarno said the plane was struggling to recover as stall warnings sounded until the end of the recording.

Audible stall warning in Airbus (or stick shaker in Boeing) is designed to inform pilots that stall condition is approaching, but tracking the fate of an Air France flight 447 we can observe that stall warning appeared many times even when aircraft have already been stalled.

So imagine, you are simply falling with terminal velocity without any lift force on the wings and airplane is saying "watch out, you will stall in a moment"
Isn't that confusing?

Sop_Monkey

29th Jan 2015, 11:38

General

You have every cause and right to be concerned. You are not alone believe me. A/B never learnt from the findings of one or if not the first "computers" to run amuck at Bangalore. It was found to be "too complex an a/c to operate" for the training the crew were given.

Shouldn't operations be keep simple? If it's relatively simple then its relatively safe. When things are complicated humans don't normally cope so well. I am aware the Europeans are true masters at complicating things.

NigelOnDraft

29th Jan 2015, 11:39

GF

but scary at the same timeI would not suggest you need to be ;)

If pilots on such a professional forum start interpreting thingsI would not assume many of the posters have done much flying outside MS Sim :{ I can claim to be a 14000hr ex-RAF FJ A320 Captain, but so can they?

The takeaway for me from all these discussions is that nobody really knows how an Airbus is programmed In software terms, no we do not. But in flying terms, we do not need to in practical terms?

Why is that an Airbus always stalls during turbulent weatherAny aircraft will stall if you pull the stick back and hold it there and the aircraft does not stop you :{

...and does not allow pilots to get back the control?It does, but you need to correctly interpret the situation (known as "SA"), and apply the correct procedures - but if you had that SA / knowledge, it is unlikely you would be in the situation on the first place. Catch 22 :ugh:

I, for one, will never fly an Airbus againI would say an irrational decision. Boeings also crash for "Human Factors". If you want something to be concerned about, it is the "dumbing down" of pilots, their training and role (which the Airbus I will agree is a step towards), "Pay to Fly" etc.

Vilters

29th Jan 2015, 12:34

NO !

First :
The autopilot was flying, the co-pilot in charge of turning the autopilots buttons, but he certainly was NOT flying.

Flying is : "I" have the stick and throttle in hand, and "I" control where the aircraft is going.

For reasons yet unknown to us, the autopilot could not cope with the situation, and left the pilots in an unrecoverable position.

Be it autopillot or sensor failure feeding wrong information to the autopilot is another discussion.

On a second side note:
Concerning general flying.
What ever happened to basic attitude indicator flying?????

Looking back at the AF447 video, the attitude indicator is clearly showing a WAY nose high attitude from start till impact.

Whatever, whenever fails, go back to basics.

Attitude, attitude, attitude. Wings level, nose ON the horizon, ball centered.

On a third note:
Basic aerobatics should be MANDATORY for ALL pilots.

IFR aerobatics on instruments should ALSO be made MANDATORY simm training.

Simple upset recovey is clearly NOT sufficient.

Well, the results are here. Pffft, unfortunately, again.

Lonewolf_50

29th Jan 2015, 13:13

What ever happened to basic attitude indicator flying?????
Possible answer: it has been replaced by Flight Director flying? This would imply that the scan has fewer items in it as a matter of habit, and thus scans used in attitude flying atrophy from disuse.
Am prepared to be wrong in this guess.

NigelOnDraft

29th Jan 2015, 13:21

Looking back at the AF447 video, the attitude indicator is clearly showing a WAY nose high attitude from start till impact.But your "modern" pilot does not fly attitudes - they consider "Manual Flying" to be moving the stick to centre the FD bars :{ Ask them to fly an approach with the FDs off, and they struggle.

Elsewhere asked "why did AF447 climb / stick back?". Noteworthy that in the mode changes the FDs went off and then reappeared as systems restored themselves. Of course, in these circs they come back in "basic modes" - vertically current V/S. So:
FDs disappear (due ASI issue)
For reasons various (see above) and in absence of AP/FDs PF pulls back
FDs reappear in V/S climb
However, due decreasing speed, and inappropriate pitch attitude selected by PF, that V/S is unattainable
PF continues to "pull" to match FD and is still trying as aircraft stalls and gives up - but pull is still maintained
No doubt this will be Airbus' FD "at fault".

As you say, if we wish to prevent this then regular manual flying (without FDs) required on the line. Actual visual UP, then Instrument UP recoveries required in appropriate aircraft - and not just trained, but currency maintained.

The issues are not the individual pilots' fault, but the current state of the industry. I doubt they will have the inclination / $$$ to solve it, just accept the odd accident and blame the pilots :ugh:

Rananim

29th Jan 2015, 13:22

Sorry if this was discussed before..but I read somewhere that when making a cruise climb in an Airbus and speed has been rolled back prior to climb THR set,that ATHR will actually command idle thrust.Can this be true?
If so,this might explain how they lost their speed margin,so critical at altitude in bad weather.In an effort to climb quickly,I can see how someone might roll back the speed selector prior to climb THR set and not notice(thrust levers dont move on an Airbus,no tactile feedback)what ATHR is actually commanding.
Could some Airbus pilot please clarify?I am used to FLCH/LVL CHG where full climb thrust is automatically commanded even if speed selector rolled back.

DrPhillipa

29th Jan 2015, 13:26

For reasons yet unknown to us, the autopilot could not cope with the situation, and left the pilots in an unrecoverable position.
Be it autopillot or sensor failure feeding wrong information to the autopilot is another discussion.

This is rank speculation. What caused the AP to disengage and how Normal Law was left has not yet been revealed at all. Gysbreght's idea of Alpha-prot inspired panic followed by wrong decisions sounds a good possibility to me but I only fly couches nowadays.

NigelOnDraft

29th Jan 2015, 13:28

Rananim

You describe what I earlier referred to as a "French quirk" in the Airbus.

What you say is partially true, however it should not result in an issue since the outcome is really just a slow start to get the climb going - and of course, the pilots should be monitoring it. After all, in a non-Airbus, do you assess thrust from where the TLs are? Or what the gauges say? Is an engine failure indicated by the gauges? Or the TL retarding itself ;)

NigelOnDraft

29th Jan 2015, 13:29

What caused the AP to disengage and how Normal Law was left has not yet been revealed at allDo we know the AP disconnected?

Do we know Normal Law was lost?

DrPhillipa

29th Jan 2015, 13:36

Nigel: probably we do not even know that

though, again from my couch, it seems unlikely to me that an A320 can fly that self destruct path under AP in Normal Law with full protections.

Rananim

29th Jan 2015, 13:36

Thanks for your reply NoD.
the pilots should be monitoring it
Yeah,of course..but what if they werent?What if their attention was elsewhere?Like the weather?

C_Star

29th Jan 2015, 14:48

Out of curiosity - what is the 'French Quirk' described by Raanim and NoD?

Been flying the Bus for a couple of year now and never heard of it...

Coagie

29th Jan 2015, 15:04

I read that the FO was flying at the time of the crash, but I also read that the Captain took over, when they started to climb. This mixes things up. Do they mean that the Captain held the stick back until they stalled, or the FO? Maybe it wasn't the FO's fault after all? Maybe the leaked information is no good (who'da thought it?). I'd like to know if they tried using the apparently oft forgotten trim wheel to get their AoA under control.

fireflybob

29th Jan 2015, 15:05

It's a few years since I flew the A320 but does the autopilot still have the "Expedite Climb" mode?

Is it possible this was selected (possibly in error)?

NigelOnDraft

29th Jan 2015, 15:25

Out of curiosity - what is the 'French Quirk' described by Raanim and NoD?

Been flying the Bus for a couple of year now and never heard of it...I earlier, somewhat humoursly, just referred to "French Quirks".

The one later referred to it, when you go OP CLB, instead of smartly going to CLB power, and pitching for Climb Speed, it sometimes does nothing with the power, or even takes it off. It seems to be unsure of whether to "enter" the climb by leading with power or pitch? Some P2s fool it by winding the speed up for a bit, I just make some offensive remark about it's abilities and await it sorting it's life out (which it eventually does).

Rick777

29th Jan 2015, 16:13

N O D. I have never heard of the Frenck Quirk, but I have seen it pitch for climb speed before adding power if you are going faster than climb speed. That is how I was taught to fly a light plane although it never made much sense.

CONF iture

29th Jan 2015, 16:26

The one later referred to it, when you go OP CLB, instead of smartly going to CLB power, and pitching for Climb Speed, it sometimes does nothing with the power, or even takes it off.
What are you talking about ... One function of OP CLB is to maintain maximum climb thrust when A/THR is active.

NigelOnDraft

29th Jan 2015, 16:31

One function of OP CLB is to maintain maximum climb thrust when A/THR is activeExactly :ok: So why sometimes does it take power off, or just do nothing, for 5-10s ;)

It wasn't be who brought it up BTW!

WhyByFlier

29th Jan 2015, 17:01

NOD I think what you're referring to is when flying a higher selected speed than cost index climb speed in the FMGC whilst level, then getting a climb clearance and pulling open climb then managing the speed. It's your naughty FOs flying higher than company CI, increasing engine internal drag and wasting fuel. In this case it'll come back to idle. It gets me once every 6 months. Pushing EXPED then straight away pulling OPEN CLB or pushing CLB locks the thrust at CLB thrust and prevents this.

mm43

29th Jan 2015, 17:29

I have been looking in vain for the maximum bank angle the AP uses. I thought it would be either 25 or 33 degrees. Do you know if there is such a limit?

The following I have associated with A330/A340, and most likely is same for A320:-

If the AoB exceeds 45°, the AP will disconnect and the FD bars will be removed.

NigelOnDraft

29th Jan 2015, 17:33

I have been looking in vain for the maximum bank angle the AP uses. I thought it would be either 25 or 33 degrees. Do you know if there is such a limit?25deg I think it says?

However it seems quite happy to use up to 35deg to capture a Localizer :confused:

WillFlyForCheese

29th Jan 2015, 18:00

Quote:

The one later referred to it, when you go OP CLB, instead of smartly going to CLB power, and pitching for Climb Speed, it sometimes does nothing with the power, or even takes it off.

What are you talking about ... One function of OP CLB is to maintain maximum climb thrust when A/THR is active.
For A320 reference:

A320 Panels / Displays: A320 Displays and Panels | Panels (http://www.efbdesktop.com/help/panels.html)

FCL: Flight Control Laws | Direct Law - Description (http://www.efbdesktop.com/flight-controls/sys-7.3.0.html)

Abnormal Attitude Law: Flight Control Laws | Abnormal Attitude Law - Description (http://www.efbdesktop.com/flight-controls/sys-7.3.20.html)

Reconfig FCL: Flight Control Computers | Reconfiguration of Flight Control Laws Diagram (http://www.efbdesktop.com/flight-controls/sys-7.5.4.html)

AP/FD Vertical Mode FMAs: AP/FD Vertical Mode FMAs | SRS (http://www.efbdesktop.com/nav-autoflight/sys-13.3.16.html)

cwatters

29th Jan 2015, 19:06

"Veering left and wobbling" before the climb..

Investigator: Co-pilot was flying AirAsia Flight 8501 (http://www.usatoday.com/story/news/world/2015/01/29/airasia-crash-investigation/22511899/)

Ertata Lanang Galih, a senior pilot and investigator, said after requesting permission to ascend, the plane was veering left and wobbling, CNN reported. Siswosuwarno said it then ascended to 37,400 feet in about 30 seconds, according to the broadcaster.

BenThere

29th Jan 2015, 19:11

[QUOTE] does the autopilot still have the "Expedite Climb" mode? [/QUOTE}

It does (actually a flight director mode) and it seeks to match climb thrust to green dot speed (roughly l/d max in the clean configuration). The engaged autopilot will grab on to the flight director and ride it up, but I don't think you'd ever see it command more than 4,000 fpm or so, even on the lightest 319.

I think I could rule out 'expedite climb' as a factor in this accident.

I agree that waiting till the investigation is complete before saying you know what happened is a good idea. But I feel pretty comfortable saying that a 6,000 fmp climb was not manually flown and involved some serious failure in the autoflight/reality interface.

juice

29th Jan 2015, 19:41

More stuff of uncertain provenance:

AirLive.net: BREAKING Pilots of AirAsia #QZ8501 Disabled Critical Computers Moments Before AirAsia Crash (http://www.airlive.net/2015/01/breaking-pilots-of-airasia-qz8501.html)

CONF iture

29th Jan 2015, 20:05

So why sometimes does it take power off, or just do nothing, for 5-10s
Power off I doubt it, doing nothing could be just a question of perception as selecting OP CLB won't produce instantaneously climb thrust as such altitude especially if thrust has been significantly reduced just before selecting OP CLB.

But if you want to link a video to what you refer as a "French quirk" in the Airbus I'll be interested to watch it.

I read somewhere that when making a cruise climb in an Airbus and speed has been rolled back prior to climb THR set,that ATHR will actually command idle thrust.
It is normal for the A/THR in level flight to reduce the thrust to match a speed reduction request, but once OP CLB is selected, the request is for climb thrust. Selecting a different vertical mode such as V/S could be different ...

I am used to FLCH/LVL CHG where full climb thrust is automatically commanded even if speed selector rolled back.
And this is not different from OP CLB.

captjns

29th Jan 2015, 20:07

Too bad the Indonesian Investigation Board refuses to indicate whether or not a full initial report on the crash will be released to the public.

Why am I not shocked.:ugh:

nickp

29th Jan 2015, 20:13

When I first learnt to fly it was drummed into me that you used the controls to put the a/c where you wanted it and then trimmed out the stick forces - you did not use the trims to fly the a/c.
With hydraulic controls the manufacturers went to a lot of trouble to devise feel systems that told the pilot what he was doing. If you took the autopilot out and pulled the stick back then you would have to pull harder and harder to keep the nose up as the speed fell - there was a clue there!
Then along came FBW and we have a/c that no longer have feel but, worse still, trim the nose up as the speed falls with the a/p disconnected (sort of). Presumably someone decided that the feedback that feel provided was a waste of money or just impossible with a sidestick.
I flew t-tail jet airliners and we regularly hand-flew them because we enjoyed doing it and we were expected to be competent at it in the sim.
Incidentally I thought that pickle switches were used for dropping bombs.:O

Coagie

29th Jan 2015, 20:22

Too bad the Indonesian Investigation Board refuses to indicate whether or not a full initial report on the crash will be released to the public.

Maybe these leaks, that not only peak curiosity, but fuel wild speculation, will motivate the Indonesians to release the FDR data and the CVR transcript sooner than they planned.

It's possible, that the investigators just need more time to sort out why the computers were disabled prior to the incident, if that is indeed true. Why they were turned off, might not be evident at a glance.

ventus45

29th Jan 2015, 21:01

I said very early on, when we first saw the ruptured pressure bulkhead, that the aircraft probably broke up, in the last few seconds, in the air, from overload, in a desperate pull-up.

I still think so, and I think the latest "information" could support that view.

The latest "information" - is that the "critical" computers were allegedly" switched off in the last few seconds.
That, if true, seems to me, to be consistent with a desperate last ditch attempt to force direct law.

How many seconds ? 5, 10, 15, 20 seconds ? The time is crucial.

Consider this hypothesis.
Out of control descending at around 12,000 feet per minute (to keep the numbers realisticly simple), which is 200 feet per second.
As best I can fathom, from weather reports at the time, local cloud base was probably about (verry rubbery) 1600 feet.

So, pop out of cloud - now visual, (it was just after local dawn) - 8 seconds to impact.

You, me, and every other pilot who has ever flown anything, are of course just going to sit there, fat dumb and happy, and just kiss our arses good-night.

Sure we will.

The other "information" that is out there, is interesting, not from what it said, but most pointedly, from what it did not say.

It is said that the fdr and cvr both stopped.

Two questions.

1. Did they stop simultaneously ?
(they were both mounted on the frames next to the missing ruptured sector of the pressure bulkhead).
(refer to the graphic showing the rupture outline on an installed bulkhead at post 1702 http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-86.html#post8819798)

2. What was the last recorded altitude ?
It will of couse be recorded as qnn, but we can convert it to local qnh from the met records.
Was it FL016, FL015, ..14, ..13, ..12, ..11, ..10, ..09, ..08, ....................... 01, ..00 ?

I may still be wrong of course.

jfkjohan

29th Jan 2015, 21:13

AirLive.net: BREAKING Pilots of AirAsia #QZ8501 Disabled Critical Computers Moments Before AirAsia Crash (http://www.airlive.net/2015/01/breaking-pilots-of-airasia-qz8501.html)

Indication that they turned 2 ADRs off and the plane was in ALT LAW?

On_The_Top_Bunk

29th Jan 2015, 21:18

I may still be wrong of course.

More than likely.

WhyByFlier

29th Jan 2015, 21:56

It says they switched off FACs not ADRs.

glendalegoon

29th Jan 2015, 22:17

Wow.

And of course when you only partially release information, you leave a number of interpretations of it:

1. They were screwing around with something to troubleshoot something that might not be consistent with published procedures. And this lead to the upset.

2. They shut down something after something had gone wrong in an effort to
recover from a problem.

I would sure like to more fully understand what is being said only in whispers and not from the authorities (as of yet) .

training wheels

29th Jan 2015, 22:40

There was a report in the Singapore Straits Times a few days ago, that the aircraft had 9 writes ups in the tech log for FAC issues in 2014.

Machinbird

29th Jan 2015, 23:38

For non-Airbus readers, the following may be useful regarding just what functions the FAC computers perform. Bolded sentences are by my action.

AUTO FLIGHT
FLIGHT AUGMENTATION
1.22 40 P 1
SEQ 001 REV 25

GENERAL

The aircraft has two flight augmentation computers (FACs) that perform four main
functions:
• Yaw function
— Yaw damping and turn coordination
— Rudder trim
— Rudder travel limitation
• Flight envelope function
— PFD speed scale management
* Minimum/maximum speed computation
Maneuvering speed computation
— Alpha-floor protection
• Low-Energy Warning function <I
• Windshear detection function <

In performing these functions the FAC uses independent channels :
Yaw damper
Rudder trim
Rudder travel limit
Flight envelope
Each FAC interfaces with the elevator aileron computers (ELACs) when the APs are
disengaged, or with the FMGS when at least one AP is engaged.
Both FACs engage automatically at power-up.
The pilot can disengage or reset each FAC (in case of failure) by means of a
pushbutton on the flight control overhead panel.
When a FAC is disengaged (FAC pushbutton set off) but still valid, the flight envelope
function of the FAC remains active.
If both FACs are valid, FAC1 controls the yaw damper, turn coordination, rudder trim,
and rudder travel limit, and FAC2 is in standby.
FAC1 keeps the aircraft within the flight envelope through F171 ; FAC2 performs this
function through FD2.
If a failure is detected on any channel of FAC1 ,
FAC2 takes over the corresponding
channel.
Note that the FACs also control the rudder operation in certain important ways. For that reason, it may be more than coincidental that the following comment by 747SP5 reported repeated problems with that aircraft's rudder system, the same computer system that the crew was attempting to shut down during their departure from controlled flight.
http://www.pprune.org/rumours-news/553569-air-asia-indonesia-lost-contact-surabaya-singapore-130.html#post8841170

Did they use the switches on the overhead panels? That would not work too well if the flight envelope protection part of the FAC system was malfunctioning without failing its self test.

training wheels

30th Jan 2015, 00:24

This preliminary report that was not meant to be made public has more leaks than a sieve.

JAKARTA (BLOOMBERG) - The pilots of AirAsia flight QZ8501 cut power to a critical computer system that normally prevents planes from going out of control shortly before it plunged into the Java Sea, two people with knowledge of the investigation said.

The action appears to have helped trigger the events of Dec 28, when the Airbus A320 climbed so abruptly that it lost lift and began falling with warnings blaring in the cockpit, the people said. All 162 aboard were killed.

The pilots had been attempting to deal with alerts about the flight augmentation computers, which control the A320's rudder and also automatically prevent it from going too slow. After initial attempts to address the alerts, the flight crew cut power to the entire system, which is comprised of two separate computers that back up each other, the people said.

- See more at: AirAsia flight QZ8501: Pilots disabled critical computers just before crash, say sources - South-east Asia News & Top Stories - The Straits Times (http://www.straitstimes.com/news/asia/south-east-asia/story/airasia-flight-qz8501-pilots-disabled-critical-computers-just-crash-#sthash.nif6KwKR.HiSUCRxG.dpuf)

ventus45

30th Jan 2015, 01:48

Machinbird, re your post 2748.

<quote>
When a FAC is disengaged (FAC pushbutton set off) but still valid, the flight envelope function of the FAC remains active.
</quote>
<quote>
Did they use the switches on the overhead panels?
</quote>
<quote>
That would not work too well if the flight envelope protection part of the FAC system was malfunctioning without failing its self test.
</quote>

If the FAC is "disengaged" by pushing the FAC set off button on the overhead panel, why does the FAC remain active in any way at all ?
Off is Off in my book.
This reads like "Off is only half off".
What is the point in having an off button for one part of the FAC's funtionality, but not the other part (the flight envelope functions).
Perhaps there should be two buttons, the second one for "flight envelope functions off".
Since there is none, the only option left is the circuit breaker ?

How many times would a FAC be turned off in normal ops, do you think ?
How many times, in what circumstances, and for what reasons, would turning off a FAC by done in the sim ?
Is it implicit in those sim exercises that the the flight envelope functions remain engaged, always, or not ?
If not, is it implicit in ANY sim exercise that total FAC shutdown, by pulling circuit breaker is required, and what the specific circumstances are, and what the consequent implications then are for continued safe flight ?

Do any of those sim'd circumstances and exercises "gell" in any way with any credible circumstance that may have occured on 8501 ?

In other words, could they have identifed (or thought they had identified) a circumstance that they had trained for, and were just working the list, or is it more likely they were in a whole "new world" situation, not trained for, no list printed, and trying to work it out themselves from square one ?

The whole idea of an off button that is not actually an off button at all bugs me.

Talk about "man - machine interface" - jesus - what are the engineers thinking when they design these things ?

It sounds more like a case of - "I can't let you do that Dave" - as in - a "prime directive" - sort of thing.

Machinbird, I do not understand the rationalle of the logic in play here.:confused:

"Please Explain".:)

(It is lunchtime. Perhaps I should go and have some fish and chips).

HEALY

30th Jan 2015, 02:08

While not on the 320, I do fly airbus and maybe its worth putting some points out regarding the last bit of leaked info. My interpretation may not be perfect and may even be incorrect for the 320 however...

The airbus does not have Circuit Breakers (except for a number in the Electrical Control Area beneath the cockpit). The airbus has Reset buttons. Wether you want to argue that its the same thing but that's what they are called. It allows the many multitude of computers that run the different systems to be disconnected, then after resetting allow a boot up of the system. My interpretation is that a CB wont necessarily reboot a system but rather just cut the power to it.

Airbus designed these reset buttons too not just cut the power but also enable the system once reset to "start again". Many issues found on the ground are fixed by just resetting the system. Not always the case but a "get out of jail card" in some respects.

The important thing to remember with reset buttons is that they may be restrictions on when they can be used and a reset performed. Some systems can be rebooted on the ground or also in the air, some may even need the engines off. Airbus provides a Quick Reference Handbook which has a section on computer resets, it will tell you of any restrictions imposed on resetting a certain system.

Wether or not some frantic button pushing of the Flight Control Computers were going to save the day it would be interesting to know wether correct interpretation of these resets and its consequences may off set off a chain reaction.

etudiant

30th Jan 2015, 02:10

The A320 is presumably always under computer control, just the extent of it modulated in case of glitches to alternate or direct. That the FAC would likewise have residual action seems unsurprising. How those specific demarcation lines are laid out or understood by the crew is less clear.
It does seem that a very experienced Captain and a mature and dedicated FO felt they were at the controls of an unresponsive or defective aircraft.
They may have been mistaken, but how that mistake came about needs to be made clear, if only to prevent recurrences.

Flo121142

30th Jan 2015, 03:07

How many times, in what circumstances, and for what reasons, would turning off a FAC by done in the sim ?
Is it implicit in those sim exercises that the the flight envelope functions remain engaged, always, or not ?
Actually, this (FAC1+2 Fault) was one of the main topics in my last sim session - the sim was setup with one 1 FAC Inop (according to MEL) and then the instructor failed us the second one after departure - it's a pretty straightforward procedure, just follow the ECAM, the only 'challenge' is that you loose both Autopilots, both Autothrust Channels and normal law, so you have to fly manually all the time. And while its true that you don't switch off the flight envelope function of the FAC by switching of the pushbutton, the aircraft should drop you to alternate law, if you turn both FACs off, so you loose protections (except loadfactor protection) - even if their computation is faulty it should not be affecting the flight too much...

Propduffer

30th Jan 2015, 03:09

It seems likely that this crew were doing the same thing that's been asked about in this thread - trying to get it in direct law.

Machinbird

30th Jan 2015, 03:13

Machinbird, I do not understand the rationalle of the logic in play here.:confused:

"Please Explain".:);):DI'm not really qualified to answer other than to offer philosophy. I've never flown an Airbus, just read up on the systems. I'm just one of those real old steam gage jet pilots who has a bit of a knack for systems.
Take HEALY's points on Airbus computer re-set philosophy and consider that Airbus engineers gave the computers dominance over the pilots in their operational concept. That you and I think this is foolish has no bearing on what has transpired already. We may only be able to influence what direction future aircraft designs take.

I suspected early on that QZ8501 was going to be a man/machine authority horror story and so far, I am not dissuaded from that viewpoint.

I'd like to point out that without the FDR data, we readers do not know when the FAC shutdown attempt occurred relative to the flight control problems. The sooner they can publish the FDR readouts, the sooner we can begin to properly wrap our minds around what caused this accident.

RatherBeFlying

30th Jan 2015, 05:48

If the pilots felt that the active law was blocking their recovery effort, they might decide to turn off some boxes in a last ditch effort.

When software goes bad, it takes considerable time just to identify exactly what the misbehavior constitutes. And it might not be the result of a miscalculation, but an unanticipated set of conditions that has the software running off in the wrong direction.

I like the philosophy in the Embraer Legacy 500 as described in the current issue of Flying.

There's just two laws: Normal and Direct.

Much less potential for confusion when things start going wrong:ok:

SKS777FLYER

30th Jan 2015, 06:12

Not bashing Airbus here........ It bashes its ownself quite well enough.
Reading the accident reports and explanations from "experienced" pilots, sounds like the old "Who's on first, what's on second" comedy routine of Abbott & Costello. This newer version is, unfortunately such a tragic one and seeming to repeat itself.
Cool sidestick controller whose position is hidden from other pilots, and if being manually manipulated by both pilots for whatever reason simultaneously, the A320 flight computers ADD the 2 control inputs together. Combined with thrust levers that don't move when FADEC makes thrust changes, major or minor.

Also it is alleged the 320 family will sometimes fully move the THS to full or nearly so, nose up, causing big difficulty,in getting the nose down - particularly when high thrust is occurring in low speed flight.

They said Titanic was unsinkable.

The 320, cant be stalled.

NigelOnDraft

30th Jan 2015, 07:50

Hi HealyWhile not on the 320, I do fly airbus and maybe its worth putting some points out regarding the last bit of leaked info. My interpretation may not be perfect and may even be incorrect for the 320 however...

The airbus does not have Circuit Breakers (except for a number in the Electrical Control Area beneath the cockpit). The airbus has Reset buttons. Wether you want to argue that its the same thing but that's what they are called. It allows the many multitude of computers that run the different systems to be disconnected, then after resetting allow a boot up of the system.I presume you are 330 or 340?

340 IIRC just had the small panel above/behind the overhead panel, and were mostly/only "resets". A320 series has this, but also has the whole rear wall of the cockpit as CB panels - and most of these I think are normal CBs? Often have to reset "CBs", iaw FCOM, and FCOM is clear some are CBs (wiring protection) and some are reset buttons. FCOM also clear as to what / where you can reset.

A quick look does not reveal FACs as items to be reset, especially airborne. ELACs and SECs can be reset, albeit only one at a time. In my Airbus time, the emphasis has shifted from "keep resetting" things to rest only when specifically required by that table. I am very clear with my colleague as to what is being done, why and by whom - and if in doubt do not rest / play.

NigelOnDraft

30th Jan 2015, 08:36

Link (http://www.scribd.com/doc/45265529/Reset-Breakers#scribd)

It may be a small point in the end, but the FACs are controlled by switches, not by CBs.Usually yes. But there are FAC CBs (proper ones!) - see link above B3 & B4.

The press article earlier might be implying these were pulled, rather than the pbs?

Roo

30th Jan 2015, 08:44

An engineer explained to me that green simply means the circuit is monitored and will display on the ECAM CB page when tripped. Nothing to do with it being reset vs cb. Black CBs are un monitored by ECAM and important ones would logically need to be where pilots could check them. Hence 330 EEC being full of green CBs while 320 has wall of black ones in the cockpit. In the 330 in our outfit the reset switches are black btw.

JY9024

30th Jan 2015, 08:47

Abnormal Attitude Law..

The A320 will switch to this mode whenever the normal envelope is exceded in pitch, roll or speed.
The aircraft will revert to alt/direct law and stay there even after recovery.

My guess is that this has occurred and the crew have tried a FAC reset to try and recover normal law..

Sikpilot

30th Jan 2015, 09:16

In time all the info will be released due to the lawsuits. It may take a while.

So far Airbus has not issued any AD's regarding the FAC's or anything else regarding this crash. Is this telling in any way or just too early????

Ian W

30th Jan 2015, 09:25

Take HEALY's points on Airbus computer re-set philosophy and consider that Airbus engineers gave the computers dominance over the pilots in their operational concept. That you and I think this is foolish has no bearing on what has transpired already. We may only be able to influence what direction future aircraft designs take.

I suspected early on that QZ8501 was going to be a man/machine authority horror story and so far, I am not dissuaded from that viewpoint.

I'd like to point out that without the FDR data, we readers do not know when the FAC shutdown attempt occurred relative to the flight control problems. The sooner they can publish the FDR readouts, the sooner we can begin to properly wrap our minds around what caused this accident.

We can already surmise a lot from the lack of reaction from Airbus who already know all there is to know of the DFDR/CVR recordings. There have been no urgent ADs issued by Airbus as one would expect if something had 'failed'. So we know that in Airbus' view the aircraft worked as expected.

So I tend to agree with Machinbird that although Airbus thinks that the aircraft worked as expected, the crew may not have had that opinion at the time. Were they trying to fix a zoom climb by pulling the FACs - is that the action for frozen AOA vanes? If they dropped into Alternate Law - from the AFR447 thread - the roll is far more sensitive than pitch - is that the case in the A-320? In trying to fix a problem did they put themselves into a test of their hand-flying the A-320 IMC, at height, in severe turbulence and in alternate law?

vilas

30th Jan 2015, 09:25

If AOA sensors had frozen at lower speed(higher AOA) then they would trigger Valpha protection(not alpha floor) at higher speed/M as the AOA threshold for the protection is lowered and the speed tape will show valpha prot/alpha max masking VLS and since it is a latching condition normally would cause a descent at Valpha prot. But if there is an updraft and speed tends to increase the aircraft in trying to maintain Valphaprot will climb. If AOA vanes were normal you can get out of alpha prot by simply pushing the stick forward. In QZ8501 case the aircraft was cruising at .78 for a while without a problem so AOA vanes should be at the correct threshold so I am not convinced that is the cause. Also to override protections you need to switch off two ADRs and not FACs unless they did it by mistake. After evoking abnormal attitude law, after recovery the aircraft goes in alternate law and stays in alternate even with gear down. If you want direct law for landing you need to reset ELACs

matkat

30th Jan 2015, 10:10

Sikpilot, just for clarity purposes Airbus have no authority to issue any AD only a national authority has that power.

cwatters

30th Jan 2015, 10:31

Many issues found on the ground are fixed by just resetting the system...

Makes me cringe to hear that. The problem might go away but it's not "fixed" unless you find out why it needed resetting. Repeatedly resetting a system is a very bad habit to get into. It's a bit like repeatedly replacing a fuse that keeps blowing.

Derfred

30th Jan 2015, 11:28

If it latches on alpha-prot and starts an uncommanded climb, wouldn't the pilot just push forward on the stick?

The leaks thus far have indicated that the pilot in the RHS pulled back on the stick.

vilas

30th Jan 2015, 11:34

Gysbreght
Alpha prot and alpha max are AOAs but shown on PFD as speeds hence the V. Aircraft when latches on to that it maintains by doing whatever it takes. Since reduced bank angle protection to 45 degrees is not relevant to the topic I didn't mention it. Also from alpha to alpha max side stick commands alpha and not load factor and it maintains the alpha prot(AOA) and not G.

rideforever

30th Jan 2015, 11:37

We can already surmise a lot from the lack of reaction from Airbus who already know all there is to know of the DFDR/CVR recordings. There have been no urgent ADs issued by Airbus as one would expect if something had 'failed'. So we know that in Airbus' view the aircraft worked as expected.
They must conclude that an AD would not prevent recurrence of the event. Which would be the case if for instance the human / computer interface needs "refinement".

I do wonder what is occurring at Airbus right now. They must be quite frantic.

"You can never make the same mistake twice; because the second time you make it, it's not a mistake, it's a choice."

lapp

30th Jan 2015, 11:41

I do wonder what is occurring at Airbus right now. They must be quite frantic.

Why they should be. Millions of their airplanes fly everyday perfectly fine. We don't even know what has happened beside unofficial leaks that do not even positively indicate a defect.

lomapaseo

30th Jan 2015, 11:47

They must conclude that an AD would not prevent recurrence of the event. Which would be the case if for instance the human / computer interface needs "refinement".

I do wonder what is occurring at Airbus right now. They must be quite frantic.

"You can never make the same mistake twice; because the second time you make it, it's not a mistake, it's a choice."

Yes

But who made the mistake under what regulation ?

Design/manufacture, Maintenance or Operations?

vilas

30th Jan 2015, 12:18

Derfred
In the level bust of A330 I think with over speed the pilot disconnected AP and when the aircraft climbed initially due to high speed protection and then due to alpha protection he didn't do any thing. It was strange because pushing the stick is a normal reaction to prevent climb and would have that cancelled the protection. In QZ case we do not know the whole story. The problem with the protections is that they were designed to prevent pilot indiscretions and not severe environmental changes. They work fine and may saved hundreds of incidents which we do not come to know. But the protections get triggered even when the environment is the cause like Bilbao, where aircraft was sinking in a severe down draft and pilot was struggling to pull but alpha prot was triggered by the high AOA and pitched the aircraft down as it was programmed to. In the level bust it maintained the alpha prot but due the prevailing AOA due to wind changes caused the aircraft to climb instead pitch down as it would have normally.