As others pointed out, are pilots at many airlines told to not use the autopilot during a go-around, and what Boeing calls VNAV, FLCH or VS, along with Heading Select for steering? I used the autopilot in Las Vegas after the Approach Controller suddenly "inserted" a B-737 ahead of us, while telling us to go-around (between the invisible mountains at night-I told the Captain that we would remain over the valley lights in case the controller gave us a questionable vector to downwind or base). Were the Gulf Air pilots already on an ILS or LOC etc hand-flown approach when the go-around began, and if so, either the plane flies the published go-around procedures or the pilots could have pushed an autopilot button and used simple vertical and lateral control modes, while following the flight directors??

I know almost nothing about the sad tragedy, but are Gulf Air pilots (and also pilots at YOUR airline) encouraged to engage whichever autopilot modes allow the easiest control of the airplane when the workload is very high? Did those Gulf Air pilots by chance have very little instrument curency or total experience? I'm not blaming the unfortunate crew a bit, but wonder just what we all can learn.

My impression of the Airbus A-320/319/321...(which I've never trained on-yet) design philosophy was that both the aircraft's unique fly-by-wire technology and the most "advanced" level of automation, could compensate for limited pilot experience, if necessary. If advanced technology=safer aircraft, and can be proven to be so throughout global airline operations, then I will avoid using quotation marks in a skeptical manner in the future. Of course several tragedies have taken place due to pilot complacency or misunderstanding this technology, if I understand the articles about the accidents in France and India.

I have not forgotten the accidents involving the B-757, MD-11 etc. What is the common denominator among accidents involving newer aircraft types, limited simulator+line training or cockpit culture and CRM?

Lodestar if you can't operate the aircraft through all stages of flight with the automatics then you have no business being there. It was always the problem during the early days of the 75 that guy's were getting confused and the most frquently asked question was 'whats is doing now?' The answer is not to reduce the aircraft to the level of a C150. Just about all of the losses of 757's have been crew induced, the same with the Airbus although here the leap in technological interface in the early days caused problems. It was an easy cop out by the instructers to say revert to manual, it is very rarely necessary or desirable to do so. What runs through all these discusions is the underlying belief that when the going gets tough then pilots must be pilots, you don't take the automatics out, you use the increased capacity to do what the automatics can't, think ahead, programme and operate the aircraft not reduce yourself to the level of mere driver and degrade the effectiveness of your other pilot who now has to monitor your flying in addition to his other tasks.

Okay Seriph, I tought that was exactly what I said as well.

Pilots should be able to perform all manoevres 100% perfect by flying Raw Data as well as using the automatics!!!!!

But what about if you where flying an approach all raw data (No AP/AT/FD) and you are instructed to do a go around....In many cases there will not be enough time to reengage automatics (except the FD which reengages automatically when setting ToGA)....When you where flying a hand flown approach you should be able to do the same during the event of a Go Around!!

Also it is a misconception that the automatics in an airbus will compensate for the pilots skills! I have flown both Boeing and Airbus and I have to say that the automatics in an airbus are much tougher to handle and sometimes really are totally against Human Nature! Do not forget as well that the whole fly by wire system has several stages of degradation which means that an airbus (Normal stage is normal law..)becomes difficult to handle when it degrades from normal law into Alternate law-Direct law and mechanical back up!!!!
From Direct law on these aircraft requires a lot of stick and rudder so stating a pilot can afford to be less skilled .....NO WAY!!!!

A bus driver has to be equally skilled as in any other advanced airplane and has to really know all his systems....despite what Airbus is trying to say in its commercials!!!They as well need to study up and invest time to keep their knowledge up to date!!!Otherwise an Airbus becomes an unpredictable and consequently dangerous aircraft in the hands of a badly trained pilot!!!!

Lodestar
 

I agree to all what you say ... except one thing: the GF072 did not have the automatic reinsertion of FD bars at TOGA ... Airbus considered this not useful when the plane was built (1994), and afterwards they have only recommended this implementation, but did not make it mandatory ... so the crew intiated the go around in dark night, without any kind of reference .. in fact, the SOP impended them to re-engage the FD bars in SSR mode until reaaching the acceleration altitude (in BH, 1500 feet as I believe).

Scaring ...

And I insist that, upon the lecture of the DFDR readouts, a very suspicious lack of response is seen in the graphics: the PIC commands almost AFT sidestick for two seconds, releases and repeats the command, but the plane keeps descending and descending ...

"the PIC commands almost AFT sidestick for two seconds, releases and repeats the command, but the plane keeps descending and descending ..."

Can this be correct?Have we been maligning the crew when the aircraft was at fault all along?
Please somebody confirm that if you pull back on the stick in one of these deathtraps,the nose comes up.Thats basic.Trim-related lag?Sounds bs to me.U can get out of trim in a Boeing and still get the a/c to do what u want.Now they want to build a 800 seater....

Rananim

My god! You mean an 80 ton aircraft travelling downwards at over 100 ft/s doesn't immediately start going up again as soon as you pull the stick back? :D

No it is NOT true that the aircraft didn't respond to the pilot's input. At least not according to the official report I'm looking at. I don't know what report Chuck Hog has been reading, but I guess there's always someone who wants to flog that tired old Airbus/Scarebus theme again. It really is getting a bit boring and a bit tiresome.

The report clearly states that nose down pitch trim is automatically applied to counteract the tendency to pitch up when TOGA thrust setting is selected. Now, I'm not an Airbus jockey (actually, I'm not any kind of a jockey), but could an Airbus operator confirm that this is normal operation please?

Even if the trim was acting counter to the elevators, I rather suspect that full aft stick deflection would have considerably more authority than trim. Again, I'm no expert. Maybe an Airbus pilot could comment?

Here are some salient excerpts from the official report that should lay your fears to rest. Unless, that is, you subscribe to some conspiracy theory that Airbus are somehow involved in a big cover-up and have doctored or influenced the official report. :rolleyes:

If you were fighting against a control system that was ignoring your commanded input and flying you into the ground, wouldn't you be yanking back on that sidestick for all you're worth and not letting go until the GPWS shut the hell up? I think even the most rabidly anti Airbus among you will have to admit that this was a CFIT incident.

The report continues:

Now admittedly that was the simulator, but the FDR did not show anything like this kind of response from either pilot. The PNF didn't even touch his sidestick during the entire accident sequence, so he clearly wasn't aware of the situation.

[edited to correct VS estimate at first nose-up command input]

Well Lodestar perhaps the aircraft should be operated in conditions such a these with the automaics engaged i.e. no hand flown raw data nonsense then it's simply 'go-around, flap---, gear up, autopilot re engage if the go around was late and the autopilot had been disengaged tfo landing. Hand flying is for nice clear conditions in good weather and only then in a quiet environment. The time cannot be far off when the autopilot will be engaged with the before take off checks and disengaged at the end of the landing roll, 'pilots' will have to know how to operate the system and their should be far fewer cfit's.

rananim + covenant
 

See the objective data yourself and create your own opinion, do not lean on indirect subjective comments.

DFDR readouts, vertical parameters, 5 last minutes, 1 last minute.

http://www.bahrainairport.com/civil/...1/planche1.gif

http://www.bahrainairport.com/civil/.../planche1b.gif

green line, sidestick deflection; red&grey line, elevators; light blue line, TRIM (I recall automated TRIM controlled by the fly-by-wire system); purple line, pitch attitude. Analyse the graphics together and you may see something strange.

See also another interesting graphic in

http://www.bahrainairport.com/civil/...t/pdf/fig3.pdf

and check the effects of the commands of the pilot in the pitch attitude and the vertical speed.

Regarding the nose pitch down on a go-around.

Virgin had an "incident" about two months ago in Hong Kong. A340 entered wind-shear and the crew selected toga at low level (300ft or so). Trouble was that despite full aft on the sidestick the aircraft pitched nose down (sufficiently for the chap in the Dragon Air office (next to r/w?) to see the entire upper surface of the wing!

The a/c reached 70ft before the nose came back up and the a/c climbed away.

Mr Airbus saw this as full vindication of his systems, the operators are not so sure...

As told to me by a BA training Capt.

757 Gti

So he pulled out at 70 ft in a microburst. Many non FBW aircraft of different manufacture were not so lucky!

757 Gti
 

Same thing happened more than a year ago in Bilbao, with an A320 from Iberia. Hard windshear, go around decision, full aft sidestick from both pilots, and the TRIM automatically noses down to avoid a computed potential stall condition (alpha floor protection). Result: the aircraft impacts the runway at a high vertical speed, nobody resulted seriously hurt but relevant structural damages in the plane.

Airbus recognized the problem of the software and has implemented the change of the ELAC computers. Maybe the Virgin aircraft you mention had not made the neccessary improvements ...

If the Iberia incident had been a 737 and both pilots had pulled full back, the result would not have been a hard landing. It would have been a conventional stall, spin, splat!

The GF 072 aircraft was functioning perfectly and did exactly as its pilot commanded it to do. The fact that the pilot commanded a totally inappropriate flight path is a different matter entirely. A 2-second aft-stick input does not miraculously 'recover' the aircraft; it merely lessens the angle of dive by a corresponding amount. Again, the aircraft did what it was told to do.

Seriph, with your desire to fully automate every aspect of flight, please enlighten me as to how the aircraft is going to perform your 'auto-takeoff, auto-approach and autoland' at a field where there is no landing aid, or when the glideslope is knocked out during a storm, or when you need to do a visual circling off an ILS to a different (non-equipped) runway..... And fyi, yes, I do fly A320 and A340 into fields exactly like these.

Much has been written since that sad night nearly 2 years ago. I believe the report is accurate, and damning. It is, however, indicative of the Company culture and mindset at the time. GF believed for too long that 'it would never happen to us'. Then it did. Much has changed since then, but the cost has been extremely high.

Tulips, I beg to differ only on your remark about a B737 in a similar situation to the Iberia aircraft.

I have flown B737, 747, 757 and 767 in Europe, Middle East and Far East and experienced some really bad weather scenarios, why do you say that a B737 would go,'coventional stall, spin and splat!' whereas the A320 only did a heavy landing? Presumably you are only referring to the action of pulling fully back on the control column by both pilots and having stall protection?

In a B737 and upwards, from my experience, in conditions other than a microburst the Boeings will, with full power and two pilots who know they cannot apply full rearward control without the strong possibility of a stall, climb away.
What makes the Airbus so special?

Chuck Hog

So you are saying that the official investigation overlooked, either deliberately or inadvertently, something in the data that you have managed to discover from looking at a couple of FDR graphs. :rolleyes:

For what my opinion is worth, I read the graphs differently from you anyway:

1) The captain commanded significant nose down input between 16:29:43 and 16:29:53 - more than ten seconds of it anyway. During this time, the aircraft went from about 6 deg nose up attitude to about 16 deg nose down. The aircraft responded to his aft sidestick commands thereafter by reducing the nose down attitude to about 5 deg nose down at the end of the FDR recording. This strikes me as ain aircraft doing exactly what it's been told to do. The elevator response (brown and grey lines) confirms this.
2) Soon after TOGA power was selected, trim moved to around about 1.5 deg nose down and stabilised there during which time the aircraft was established in a stable climb. When the captain started to command the nose down on his sidestick, the trim blipped down a fraction, maxing out at maybe 2 deg nose down and finally ending at about 1 deg nose down.

I don't believe you are telling me that this trim graph indicates that the aircaft flew itself into terrain against the commands of its pilot! All the evidence points to CFIT. There is no evidence from the CVR that the pilots were surprised by the response of the aircraft to their command input. The PF didn't pull back as if he was fighting to bring the nose up. The PNF didn't even activate his sidestick. How do you answer those rather more pertinent facts?

The other incidents that have been mentioned regarding the A320/A340 response in windshear do not have any real bearing on this incident, and particularly not on the question you have raised about the automatic trim. A40-EK was responding only to TOGA selection and had established a steady climb with nose-down trim. In the case of windshear, it is likely that the FBW system was limiting command authority to the elevators because of approaching stall conditions.

I'm sorry Chuck Hog, but I have to say that I think what we are seeing from you and the other Airbus scaremongers is technophobia (I give the benefit of the doubt that it is not just parochialism or, worse, protectionsim). No matter how advanced we become technically, and how educated or professional we are, we still do not really trust machines deep down in our psyche. In essence (and I apologise for crossing threads here), this is why we had the tragic loss in the recent mid-air collision over Germany. In the face of a conflict between advice being supplied by a computer, and the reassuring human voice of the ATC, the Russian pilot chose to trust the human.

Unfortunately, as has been amply demonstrated so many times, most tragedies are caused in some way by human error. If we could bring ourselves to trust machines more, maybe the world would be a slightly safer place.

Come of it Tulips, you know what I mean. If you can't fly ILS then you still fly non precision or even visual approaches with the autopilot, certainly flying manual circling approaches in difficult conditions has proved time and time again to be dangerous! 3D sat nav will take care of the lot, won't be long, in the meantime when I'm down the back I keep my fingers crossed that we haven't got a macho pilot up front.

Hi Blue Eagle

Yes, that is exactly what I am referring to. I mean that in the Iberia case, both pilots were able to apply full 'back stick', resulting in probably the best possible resulting landing (taking into account sink rate, full power and pitch input). In a 737, or an A300B4, it is not possible to pitch to exactly the critical alpha; one can only get close, which is not quite as efficient.

No slur intended on Mr Boeings product, merely an observation that you cannot have both pilots applying 'full back stick' in a low energy situation in a 737 aircraft without ending up in an unusual attitude, to say the least! The same comparison would be valid in the case of the very first A320 crash. The aircraft ended up in the forest, but my point is that it was the right way up, which has to help limit injuries or fatalities.

The Airbus is not special, just different!:cool:

Tulips

are you sure you are flying BOTH A320 and A340?:rolleyes:

Hi SID; yes, I reckon a few thousand hours in each type qualifies.

Seriph, if you look at FCOM3 you will see that it is against Airbus SOP to fly a visual circuit with the autopilot. ('At the beginning of downwind, deselect FD'). Further, there are times when a manual circling approach is required, whether the pilot likes it or not. That is the time when he/ she had better be able to fly the airplane. Depending on pilot ability, there can be a big difference between a 'demanding' approach and a'dangerous' one, to use your phrase.

Don't forget, the conditions in BAH that night were neither demanding nor dangerous. Clear skies, serviceable aircraft. It was only dark.....

The autopilot will always do it better than you Tulip, you just need to know how to tell it to.

Yes, sir, Seriph. Thank you.....

Seriph,

No it won't.

'Always' do it better that is.

I just don't comprehend how you continually attribute such infallibility to autoflight systems. Are you saying that you have never had to intervene because an autopilot has been performing less than brilliantly? I'm talking modes correctly selected, as opposed to having to intervene manually as a quick way of sorting out finger trouble?

CPB

Clearly Seriph is a young(ish) guy/gal who has not been in aviation for very long. Pity the punters down the back should he/she be in charge at the pointy end when the autopilot is unserviceable. Day or night.:rolleyes:

"A little knowledge is a dangerous thing. No knowledge at all is frequently fatal."

An Autopilot / AFDS is best regarded as an extra set of hands and a big help most of the time. However it is far from perfect some of the time and needs watching. The modern equipment is becoming more and more reliable it is true, but herein lies a big trap. (and if I need to explain that then you should not be reading this.)

An autopilot will only fly better than you if :

1. You are tired, distracted or busy.
2. You are disorientated or overloaded.
3. You never learnt to fly properly in the first place.

These are then the conditions when it is best to use the A/P. All other times its' use is best left to that old fashioned term - Airmanship.

Before anyone starts about Cruise Economy / Turbulence / Departure Tracking / Workload etc etc - I refer to the paragraph above.

There is a new generation of Nintendo Kiddies out there folks - Pinball Wizards to a man (I will totally discount the females here since,without exception, all that I have had the honour to train have been great handlers). On the rails these PBWs / NKs are fine, if mechanical. Get off the rails and then watch out for the recovery. It's a worry.

MG

Capt Pit Bull, yes they do, I defy you to fly any procedure as accurately as an autopilot and if you try have any capacity left for the real world on a dark and dirty night and if you try it why?

411A - 37 yrs 16000 hrs. Trainer. I have flown all types with and without autopilots. I have never had to operate glass cockpit aeroplanes with all autopilots u/s, if I had to fair enough, but in 15 yrs with my current airline we have not had a single occurance of multiple autopilot failure. I have witnessed many near misses and crashes in simulators because guys cannot operate the aircraft and revert to manual flying. I have never seen a correctly operated flight system crash the aircraft. We have cases by the dozen where aircraft have crashed because the crew screwed up, this looks like one, engage the autopilot in hdg and alt and you won't hit the sea!

Master Green - 777's ? you're in the DC3 era.

Hasnt it occured to you all that the way to do it lays....


EXACTLY in the middle between everything you are saying....

100% Autopilot......BAD!
100% Manual.........BAD!

Try to combine both in a way that suits you best considering fatigue, situation, traffic, etc......

:cool:

Seriph,
You sure dont talk like a trainer who's been flying 37 years.Most good trainers emphasize raw data and handling skills.Pilots who spend too much time on automation(not crusie) can become detached from the aircraft they are operating.eg,L1011 crash in the Everglades.Both pilots watched that A/P take them into the swamp...No disrespect to the crew..they fell into a trap that would fool most..misunderstanding of CWS and A/P mode annunciations.
You must surely know there are maneuvers that must be flown by the pilot...Windshear/TCAS/GPWS escape maneuvers.Visual approaches are to be flown manually.
I strongly suspect you fly Airbus...then dont forget that their chief pilot died with the A/P engaged.
A/P is a tool..a good tool.Use it as such.

"I strongly suspect you fly Airbus...then dont forget that their chief pilot died with the A/P engaged. "

A lousy remark dont you think, RANIM?
:o

The autopilot flies the airoplane, you programme it and monitor. Simple really, of course you can 'fly it' like Buck Rogers but you reduce the crews efficiency and significantly increase you chances of hitting something. This has been established time after time. Apart from the few cases where temporary reversion to manual flying is required then you only do it for fun not for safety. I fly Boeings.

Seriph you are starting to bore me; yes the AP is a good tool to use for reducing workload, but in many conditions the AP stinks.

Ever in your 16000 hours flew an automated approach with too much crosswind (say 40 knots) and saw what happened?
Ever seen a 320-AP shoot 150ft through its level, because of the builtin passanger comfort?
I don't need to be in the DC3-era to see things like that happen, and I'm glad I'm there to take over from my autopilot when it just doesn't do what it is supposed to do, or to be able to take over when the conditions met are not what designers thought off on beforehand.

I go with Lodestar, 50% automated, 50% manual suits me fine.

Besides... the remark about the airbus chiefpilot who died while flying automated may be pretty macabre, but the case shows perfectly well, that if you encounter circumstances not before thought of at the design-table, the autopilot will not be able to handle it and then you should be.

P77

Bet you drive a horse and cart.

Seriph, autopilots on boeings such as the 75/76 don't even have rudder control unless you're on a final approach. if the eng fails in the cruise the thing will roll onto its back and kill every one.

Not if you apply a bit of rudder and trim it laddy.

An uncalled for remark about Nick Warner and incorrect,the a/p was diconnected.

There is nothing wrong with flying manually but remember high performance a/c are more efficiently and comfortably flown on the a/p.
I am afraid the day of the macho pilot is over.!

Seriph,

I am not disputing the virtues of using automatics.

I am simply taking issue with your statement that the automatics will always do it better.

You are joking, right?

Any Procedure? How about any landing with a crosswind in excess of 20 Kts.

What excactly does a BA captain know about the A340? And please, if you need to fly at alpha max and in order to do that the pitch attitude is less than zero, so what? After all, in a microburst of large downhill proportions than alpha max may not be positive with reference to pitch. Heave the nose up and stall in a conventional a/c? I believe that too many people comment on things that they know nothing about.

AhhhVC813, your sentence

is exactly correct.

(A certain 37-year old "trainer" with 16 000 hours and 15 years in one airline, which makes him a tender 22 when he joined..., AND who has still managed to fly "all types", comes to mind).

Personally, thanks to the people who have posted meaningful and thoughtful replies to a topic which, after all, discusses the death of 143 people.

To seriph and his spoiler mates, while it is your 'right' to be patronising, disparaging and rude to people about a subject that you successfully give the impression of knowing very little about, it might be an idea to start a new thread where you can all squabble and argue among yourselves, and come away feeling victorious about having 'won'.

Manual vs AP
 

I'm not a complete technophobe (hey I can program my VCR!), in fact I work in the tech industry, but I can say that Airbus software is almost certainly FLAWED.
Before you start, no I am not saying that has anything to do with this incident, merely that probability tells us that the sheer volume of code, and the complexity of the calculations involved for each of the permutations involved means that it could be flawed in several ways. It may be missing an essential response simply because the circumstances haven't happened yet for them to be added to the code.
It may have an error in incident response code for circumstances it DOES knows about, but that haven't happened yet. It may have flaws in code that does get executed, but doesn't have a significant or noticeable impact on aircraft activity.
But it IS flawed, I guarantee you that.
So to all you technophobes: Yes, the Pilot should always be ready (and able) to take control in the blink of an eye.
To all you technophiles: It's written by HUMANS. Humans make mistakes. Ergo it makes mistakes. Be ready for them!
Here endeth the lesson, aaaaaaaamennnnnn.
Regards,
Desertia

P.S. In the 36th century, Robots will take over and come back looking like Arnold Schwarzenegger. Is that what you want? 'Cos that's what'll happen!

As Seriph is misunderstanding autopilots, so Desertia is having his shot at misunderstanding the whole scope and breadth of modern control technology. Too much sand up the nose, mayhap?

The short answer is that - contrary to Desertia's assertion - we currently know how to make (at considerable cost) software, computers, and composite systems that are provably more reliable and more nearly 'correct' than any other component one finds on an aircraft.

To the extent that autopilots, control systems, etc. are not perfect in practical use, the problem in well-engineered designs is almost always one of incomplete SPECIFICATION of the circumstances to be dealt with and/or UNDERSTANDING of the appropriate solution. In less completely engineered ones, a lack of complete TESTING can also be an issue.

To the extent these issues exist, all are basically matters of as- yet incomplete communication between the people flying and the people designing. The full automation of aircraft is a 125-year problem, with only 75 years so far under the belt. When it's done, there will still be pilots and they'll still be saving the bacon now and then, so don't worry about that.

In a PC-dominated world it is an easy mistake to confuse the motives and abilities of those who design truly critical systems for health and safety with the products of highly visible companies, like Microsoft, that aim to profit greatly from a perpetuum of faulty consumer products requiring frequent replacement.

Like any computer...GIGO garbage in...garbage out...bad inputs either to the flight controls or the FMC can have bad results...obvious which one was the case here...:confused: :confused:

No Tulips 37 yrs flying not age. I wish. Oh and if you switch your flight directors off, you can still fly the 'birdie'

Arcniz, stick to theory, problems with auto systems are very rare. The difficulties come with the human / machine interface and was why initially Airbus had so many accidents. Now that the pilots have learnt how to use it the system works very well. As long as you are prepared to come out of the stone age.