SATA brand new A320 ; hard landing in Lisbon

[overun]

l know nothing of the Airbus, but l do know a little of aircraft.

What ironic nomenclature manager came up with the "retard" call ?

Sorry, l have to go to bathroom now, albeit too late.

[PBL]

Overrun,

a joke making the rounds after a relatively recent high-profile event on another continent: the countermeasure is to make the call louder. I guess that was ultimately seen as just too rude.

[BOAC]

B&B - I'm not sure why a '10 kt tailwind' affects the 'need' for throttle closure?

overun - possibly a 'retard'? Let us again ask the 'experts' which human sense vanishes first in a stressed situation?

I think this has been done to death. My views on piloting are that I would choose when I do something like closing the throttles and not have some dreamed up/modified/confusing software/subsequent 'SOP' tell me. Yes, that is, apparently, old-fashioned.

As all who have dabbled in writing lines of code know, the odds are high that when you go in and change a line or two to 'correct' some discovered anomaly you introduce others (Merton's 'Law of Unintended Consequences'). Whoever decided that you should have the throttles closed on landing must be very experienced on tail-draggers. The ONLY reason for a competent pilot to have the throttles closed on or after landing on a 'modern' nose-wheeled aircraft is to allow the aircraft to be slowed (as Congonhas found). It does not need to be an 'SOP'. If it does then the aircraft is wrongly engineered.

Of course, when we move to a totally automatic aircraft (or 'Captain and dog' for those who recall that little prophetic 'joke') the software will guarantee to cope with a bounce/tyre burst etc etc will it not - or will the 'Captain' exceptionally be allowed to touch something even if bitten?

The only question would be who is on who's lead as the crew walk up to the kennel?

[BlackandBrown]

With ATHR in (normal ops) the aircraft will fly at Vapp Target all the way down. With a tail wind this will be Vapp ( + no 'gust' value) because there is a tailwind in this instance. As we enter ground effect the tail wind will typically reduce giving an increase in our airspeed. Either way we will have a higher groundspeed. To counter this, closing the thrust levers and using the flare to control the speed will help ensure a touch down in the right place and with a 'normal' touch. If you retard the thrust levers at 20 the aircraft will float - I haven't done a landing where it hasn't or seen a captain do one where it hasn't. If you try and fly it to the ground carrying the excess speed you will probably land overly firm and possibly bounce. Either way it probably wont be controlled. I have found closing the thrust levers later, in a headwind, works quite nicely for the same reason. I am not massively experienced so I could be a million miles off but it is what I have been taught and it seems to work for now until I gather more experience in different circumstances.

The reason I offered 10 knots as a tailwind is because this is the limit of the aircraft.

[overun]

BOAC.

l am possibly a " Retard" but there again, l really wouldn`t know would l ?

l have to rely on my betters to guide me on that one.

lf the speed is right, whatever the aircraft is, it shouldn`t float or bounce.

Or am l wrong ?

[overun]

PBL.

Thanks for that, l didn`t know.

[BlackandBrown]

Yes you are wrong.

[BOAC]

overun - I think you mis-understood - I was answering your "What ironic nomenclature manager came up with the "retard" call ?", not 'classifying' you

[fdr]

qualitatively hundreds of them: of pilots and pets

Alargeplanemaker Industrie found that their rendition of FBW certainly had more failure modes than they had time to validate, and thereafter, in service, new and novel system behaviour came to light, including sensor input errors, hardware errors, and cascading faults that resulted in abnormal system behaviour following a hardware fault. (Harley-D would have patented such lousy performance and turned it into icons of tasteful design).

PBL notes that there are fundamental issues related to validating system design, and there is the issue. If the system design requires the pilot and dog to safeguard it's performance as it is rationally not achievable to develop a truly fault tolerant failsafe system, then the mere act of placing the least reliable component in monitoring duties (pilot not the dog...) in such an out of the loop position ensures poor performance. This is not going to elegantly resolve itself in the short term. Any system that places the pilot further away from the active cognitive demands of control is going to end up with the dog being the most situationally aware and reliable component of the program, (other than distractions such as food, ear rubs, spot marking and "Squirrel!").

As we seem to be at an impasse on reducing "Human Error" in system design and operation, seems a sad state of affairs to consider that "Canine Error" is easier to chew on. Would the dog not have the same issues though of inadequate dog rest, poor pay scales, low self esteem, Man-Machine-Canine interface issues, inadequate sanitation facilities, and airline food fit-for-humans-only.

I can state from experience that dogs get somewhat confused in barrel rolls, and just hate negative g, but of course that is entertaining at least, when they have big ears. The wonder is that a dog will still chase after the plane to get on board with their pack leader. On the other hand, Caesar Millan would have a new business in behaviour modification of the dogs, probably cheaper than a full evaluation of a software code's/hardware design's potential for aberrant behaviour.

Actually, I think that the dog whisperer could have some pointers for current airline management to stop them being bitten by their staff, and at least would have a chance to work on the incontinence issues, although I suspect the erosion in the industries remuneration to the extent that pilots need to work until they have 2 feet in the grave may be a factor in any increased prevalence in incontinence in our Chuck Yeager's.

Perhaps, crews are just p****d off...

[Loose rivets]

It's like reading something from Surreal Land.

[BOAC]

fdr - then, of course, the management will spot the 'pay to bite' opportunities and off we go again. There will be no need for the 'pilot' to utter the words, "sit" "stay" as the software will do this, with rising frequency and volume until the command is followed, with triply redundant sensors and 'voting' algorithms following the reaction.

I struggle to see how airline management will adapt to 'reward training', however. The electric collar can readily be adapted to combat fatigue (sorry, 'dog tired'................) of course.

LR - how now?

I bet even now the software guys are smacking their lips and drafting code. All my fault

[windytoo]

Woof. Woof.?!?

[Clandestino]

Ladies and gentlemen, I have watched this highly entertaining thread with some amusement and I found myself unable to resist the temptation to participate and spoil the fun.

Regarding the aerodynamics: notwithstanding her swept wings, A320 is quite a happy floater. If flown a bit above Vapp its runway gobbling capacity is on pair with PA-28, factored for size. I have done about 500 landings as PF in her and watched the roughly same number, in various WX and rwy conditions, and never observed the need to land her with anything except idle thrust. However, if handled with average manual flying skills, it can be safely landed with thrust above idle. It's against the SOP but it wont harm you if you've done everything else properly (I was told that and checked it in the sim, if you wonder). Ground spoilers are helpful to keep her on the ground but they're not really essential if you keep your touchdown soft enough not to bounce. I'm digressing a bit but I must mention that the hardest part of my conversion form 320 to Q400 was learning to land with power on - something I haven't done for previous 4000 hours. If you haven't seen her, Q400 is straight winged thus it is pretty certain that wing sweep is not the only factor affecting the landing characteristics of an aeroplane.

It is only natural for pilots to try to defend the coleague as much as possible and search for the culprit in machine and but sequence of this mishap started at first touchdown, with 752 fpm RoD when mainwheels met the tarmac and I see no sugestion that machne made the pilot do that. Even if it wasn't for second touchdown, it would warrant heavy landing inspection.

It wasn't first case it happened on A320 either. I remember exactly the same sequence described in Flight somewhere in mid-90ies, Air India was involved, if I'm not mistaken. Hard landing with thrust levers out of idle, WOW latched, bounce, levers retarded, ground spoilers, smack.

Inflight deployments of ground spoilers did happen on other aeroplanes, like DC-8 and DC-9. Ground spoilers logic on 320 is pretty simple, with not much place for agonizing over complex codes or byzantine faults; wheels are spinning, struts are compressed, trust levers are at idle, therefore we're on the ground and are stopping so we could use some boards to assist with the braking effort. Now, some could scream that we need another input to really check whether aeroplane is on ground, eg. radio altimeter. Good thinking - errr, not. Introducing one more input parameter would needlessly complicate things and introduce more potential for failures. Now that something for competent risk analysts like PBL to dwell upon and judge whether the new solution creates more problems than it solves.

Regarding the computers: all of you worrying about being replaced by computer, wondering why the ATHR computer doesn't better anticipate the need to change to thrust or being shocked by inability of SEC to recognize the bounce, stop now! Real intelligence is needed to accomplish any of the three aforementioned tasks and as far as it is known there is no intelligent computer anywhere in the world nor it is reliably forecasted that it will be made anytime soon. 99% of the line flying is routine and routine is something that computers excell at. The remaining 1% remains insurmountable obstacle for electronics and it is where we earn our pay or get killed.

Originally Posted by CONFiture

To write that a go-around must be initiated following a high bounce, is not enough.

A note should clearly warn that retarding the thrust levers after a bounce could trigger the full deployment of all ground spoilers in the air !

I'm not sure what procedure is nowadays, when I was converting to A320 in early 2007, procedure I was taught to TOGA TEN after bounce - TLs in TOGA detent (ground spoilers would auto stow) and maintain pitch 10°, accept the second bounce if it happens. Follow normal go-around procedure only when rate is absolutely positive. Regarding the need for note, perhaps you're right but then those whose interest in their aeroplane did not end with the FCOM had very good chances of hearing about previous landing mess ups.

Originally Posted by PJ2

or one can move the thrust levers forward (out of the autothrust regime) and obtain instant thrust.

I see what you mean, but taken out of context it can be a bit misleading. To clarify: you can get increase above thrust commanded by ATHR by putting TLs between CLB and TOGA detent. However there's nothing instantaneous about it: CFM56s have lots of inertia, V2500s even more so and 60 tons of aeroplane is not to be regarded lightly. Burst of power at last second might save botched landing in Twotter, chances of same procedure succeeding on 320 are extremely slim. Heavier the aroplane more the forethought is required to fly it.

[Meikleour]

BOAC: as an addition to Chris`s excellent posting - what many people on here who have not been "Airbussed" may not realise is this .......... with the ATHR engaged, when the Thrust Levers are retarded in the flare manoeuvre, the thrust delivered is not actually reduced until the Thrust Lever Angle passes the point which coincides with the actual thrust being delivered. Only after that is the thrust actually reducing. Typically this occurs after about 50% lever travel. Not the same situation at all to a more conventional driven lever system. The Airbus guidance to close the thrust levers prior to touchdown is more to do with preventing the thrust from creeping up during the flare itself. So, with a flare guidance of 20ftRA then the thrust is barely off prior to touchdown or should I say impact!!

To the poster who claimed never to have seen the spoilers deploy on bounces on base training surely this was more to do with them not usually being armed for touch and goes. (spoken from the perspective of a BTC on Airbuses for seven years)

[PJ2]

Clandestino;

Quote:
Originally Posted by PJ2
or one can move the thrust levers forward (out of the autothrust regime) and obtain instant thrust.

Originally Posted by Clandestino

I see what you mean, but taken out of context it can be a bit misleading. To clarify: you can get increase above thrust commanded by ATHR by putting TLs between CLB and TOGA detent. However there's nothing instantaneous about it: CFM56s have lots of inertia, V2500s even more so and 60 tons of aeroplane is not to be regarded lightly. Burst of power at last second might save botched landing in Twotter, chances of same procedure succeeding on 320 are extremely slim. Heavier the aroplane more the forethought is required to fly it.

On the contrary, the procedure works well if and when a slight increase in energy is all that is required. Of course, the procedure would necessarily have to be permitted by your FCOM and that is what provides the context for understanding and use.

Normally, if the ATHR is engaged, the speed will not decrease below Vls but there may be circumstances (quartering, slightly increasing tailwind) which may require a small, momentary increase in thrust and setting the thrust levers above the CL detent is one way - disconnecting and flying manual thrust all the way to touchdown is another.

Not all flight crew training manuals describe the procedure nor is it necessarily an SOP for all operators. While the procedure is a valid one, each airline publishes its own SOPs using Airbus publications as final guidance. In other words, I shouldn't think that an airline would permit something not permitted by Airbus but may restrict their own crews from something permitted by Airbus.


Here's an example of a description of the SOP described above:

If all thrust levers are set to beyond the CLB detent, when A/THR is active, the flight crew manually controls thrust to the Thrust Lever Position.

The FMA displays MAN THR in white, and the A/THR is armed.

As a reminder, LVR CLB flashes on the FMA. This technique is most efficient, when the aircraft speed goes significantly below the target.

When the aircraft speed or acceleration is satisfactory, the thrust levers should be brought back to the CLB detent.

This re-activates the A/THR.

SPEED DROP ON APPROACH: RECOMMENDED RECOVERY TECHNIQUE

Note: When using this technique during approach (e.g. to regain VAPP), the thrust levers may be moved past the CLB detent, but not beyond the MCT.

In most cases, it is not necessary to go beyond MCT, as the PF could inadvertently advance thrust levers all the way to the TOGA stop, and thereby engage go-around mode.

Meikleour, superb, helpful comments.

PJ2

[Loose rivets]

My comment:

It's like reading something from Surreal Land.

was pertaining to the link above (in post 113). Not to the thread in general.




I know it's an aside, but having come from an era of 8 seconds to 80% thrust, I'm left with imagining a Hanna Barbera manipulation of gravity to allow the aircraft to sit at 10degrees nose up, waiting for clean wings and enough power to go around.

"Accept a second bounce." Okay, if it bounced in the first instance, then I guess it's likely to have significant airspeed. But that won't always be the case.

[PJ2]

Loose rivets;

I'm left with imagining a Hanna Barbera manipulation of gravity to allow the aircraft to sit at 10degrees nose up, waiting for clean wings and enough power to go around.

If I may, let's express these notions with some care, avoid hyperbole and try to understand the airplane.

The aircraft is certified for the maneuver and there is an SOP for it. There is no "10degrees nose up, waiting for clean wings and enough power to go around". One description of the SOP in a high bounce is to apply TOGA thrust, maintain current pitch attitude, accept the possibility of a second touchdown and leave the aircraft configuration as it is until a definite rate of climb is established then complete the standard go-around procedures. There are other words but these are the essentials. I've seen it in flight data, (twice), and that's how it went. The airplane is certified for the possibility of a touchdown during a go-around, (caters to the CATIII approach).

We need to keep firmly in mind, (as recently reminded by Safety Concerns) the other, more important aspect of this event, which has occurred a number of times (seen it), and that aspect is the dispatching of the aircraft after a cursory examination of the airframe after the captain's report, a dispatching which occurred as a result of an inability to access/read/interpret/act upon the available ACARS/AIMS data in the LOAD 15 REPORT which led to permitting commercial and scheduling factors to influence and prioritize decisions. That is an organizational issue with far more serious implications than the plusses and minuses of the design philosophy of the Airbus A320.

[Chris Scott]

Thanks Meikleour, for picking up my argument (#94), and carrying it further. I always advised that the only way to select idle from A/THR was to close the thrust levers rapidly, without trying to modulate the reduction in thrust: the FADECs do that for you. It's impracticable to find the angle at which the reducing thrust limit (as you retard the levers) matches the existing thrust when you are eyes-outside in the flare. Accurate modulation of thrust changes by the pilot can only be achieved in manual thrust; and this has to be established much earlier, at a safe height.

Quote from PJ2 (#119):
"Normally, if the ATHR is engaged, the speed will not decrease below Vls but there may be circumstances (quartering, slightly increasing tailwind) which may require a small, momentary increase in thrust and setting the thrust levers above the CL detent is one way - disconnecting and flying manual thrust all the way to touchdown is another."

This is the technique I referred to in my post (#94):
"The difficulty arises in the case of significant airspeed loss just before the flare, before A/THR is disengaged. There is a crude logic which enables the PF to apply a burst of thrust by pushing the levers forward for a very short time (assuming he/she does not elect to go around)."

The question is: why does the pilot need to intervene? The A/THR should already have recognised "speed below target and falling", and have called for extra thrust. If it is a gross speed deficit (say, > 5 kt), the FADECs should be accelerating the engines as fast as possible. But the new target thrust is likely to be well below CLB, and will be reduced as soon as IAS approaches target.

In the procedure PJ2 describes, on the other hand, the pilot is calling for thrust in excess of CLB, albeit momentarily. The trick is: how long is that? The curve of thrust versus time is (roughly) exponential, so there is a significant risk that target IAS will be well exceeded before the levers are returned to the CLB detent AND the big fans have had a chance to slow down again. That would be likely to necessitate a go-around, particularly on a short runway in a beam wind, which might otherwise have been unnecessary.

It would be interesting to hear how successful PJ2 and others who have used this expedient have been in avoiding overkill. I would also like Airbus to explain if and why, with a serviceable A/THR, it should be necessary for the PF to intervene, except to go around. In the early years, many of us commented on sluggish autothrottle response, and it seems to be little better today. Bernard Ziegler should have provided us a better one, particularly as he insisted on static (non-driven) throttles.

But manual thrust, assisted by GS-MINI, is a delight for visual approaches...

Chris

[PJ2]

Chris;

Your question, and your caution are very well placed. Indeed, one must be very careful with the procedure.

The autothrust system maintains speed well enough that action shouldn't be required. The question arose as to how the pilot might obtain more thrust if in the pilot's judgement such is required; - a momentary selection above the CL detent can be made, but you are absolutely right...it can be over-done so must be used with great caution - a second or two would be all that would be needed, and that is from my experience on the A320 and A330. Even in the last 20ft, there is plenty of residual thrust as the thrust levers are closed, (precisely because of the behaviour of the FADECs as you have described) and the engine response, at this point beginning the flare, is much quicker than has been assumed.

In fact, in researching the question, I couldn't find the procedure in some manuals familiar to me so at least some have decided it isn't necessary for normal ops. Perhaps the risk of inadvertently going further, into TOGA, was deemed greater than the small benefit. In my experience, the autothrust system does an extremely good job.

[CONF iture]

In fact, in researching the question, I couldn't find the procedure in some manuals familiar to me so at least some have decided it isn't necessary for normal ops. Perhaps the risk of inadvertently going further, into TOGA, was deemed greater than the small benefit. In my experience, the autothrust system does an extremely good job.

PJ2
The first time we have heard about that procedure was in 2004, I believe.
It was the subject of a FCOM BULLETIN called AIRCRAFT HANDLING IN FINAL APPROACH

The interesting part is, that procedure has proved to have more drawbacks than advantages, so, as discretely as it could be I must say, Airbus has decided in 2009 to not recommend to use this procedure any longer.

If the A/THR performance is not satisfactory, the flight crew should take over and control the thrust manually.