SATA brand new A320 ; hard landing in Lisbon
CONF iture;
From the A320 FCOM:
"Predetermined altitude call outs are computer generated by the Flight Warning Computer (FWC) beginning at 400 ft RA. Auto call out by the FWC below minimums/decision include "100", "50", "30" then "RETARD", "RETARD" is generated if the Thrust Levers are not at idle by 20 ft (manually flying) or by 10 ft (autoflight engaged).
"AT 30 FT APPROX
"Commence a progressive flare while simultaneously closing the thrust levers; thrust levers to be at idle prior to touchdown. Allow the aircraft to touch down without prolonged float."
Further, from the A330/A340 FCOM:
"In Normal Law the flight controls provide for a conventional flare and touchdown. Normal Law provides protection for Attitude, Maneuver, Overspeed and Wind Shear. Degradation of any of the protected systems is displayed on the PFD.
"Predetermined altitude callouts are computer generated by the Flight Warning Computer (FWC), beginning at 400 ft RA. Auto callout by the FWC below minimums/decision include "50", "30", then "RETARD". "RETARD" is generated if the Thrust Levers are not at idle by 20 ft (manually flying) or by 10 ft (autoflight engaged).
"AT 40 FT APPROX
"Commence a progressive flare while simultaneously closing the thrust levers; thrust levers to be at idle prior to touchdown. Allow the aircraft to touch down without prolonged float.
"Note: If one or more thrust levers remain above the IDLE detent, ground spoilers extension is inhibited."
The FCOM operational statements for both Airbus types are abundantly clear. The thrust levers are to be at IDLE before touchdown. The "RETARD" auto callout is a reminder for this SOP.
This is how it is taught and how it is checked in recurrent training and line checks. We even had a FOQA event for "Thrust levers not at IDLE at Touchdown".
It is also taught that on a severe bounce, the SOP is to go-around, not try and rescue the landing. I have seen in flight data such bounces followed by go-arounds and they work as described. I have also seen and dealt with the very event which occurred with the SATA A320.
fantom;
It can, and has happened, at least twice. As explained in PBL's post, Airbus has modified the logic of deployment since.
Rather than my taking up bandwidth here, it's better to read the report to understand why this can happen, especially if you're training pilots, (did that too...even "interesting" doesn't describe it adequately at times!)
In post #78, PBL states (accurately), "The point is that the MLG compression is latched for 3 seconds, and ground spoilers come when thrust levers are retarded to idle. " The FCOM "Note" cited above, "Note: If one or more thrust levers remain above the IDLE detent, ground spoilers extension is inhibited", refers directly to this and so the SOP to ensure thrust levers are at IDLE prior to touchdown relates directly to this.
Some fairly significant decisions which had to be made when designing the system, one of which is the logic that confirms to certain systems that the aircraft is landing, and on the ground. This isn't as straightforward as may first seem, so one must understand this, and must understand the system.
BOAC, I'm confused by your comment in re PBL...what is there about swept wing handling that has to do with ensuring the thrust levers are retarded prior touchdown as per SOP? What have I missed?
SOP certainly not, at most recommended procedure.
"Predetermined altitude call outs are computer generated by the Flight Warning Computer (FWC) beginning at 400 ft RA. Auto call out by the FWC below minimums/decision include "100", "50", "30" then "RETARD", "RETARD" is generated if the Thrust Levers are not at idle by 20 ft (manually flying) or by 10 ft (autoflight engaged).
"AT 30 FT APPROX
"Commence a progressive flare while simultaneously closing the thrust levers; thrust levers to be at idle prior to touchdown. Allow the aircraft to touch down without prolonged float."
Further, from the A330/A340 FCOM:
"In Normal Law the flight controls provide for a conventional flare and touchdown. Normal Law provides protection for Attitude, Maneuver, Overspeed and Wind Shear. Degradation of any of the protected systems is displayed on the PFD.
"Predetermined altitude callouts are computer generated by the Flight Warning Computer (FWC), beginning at 400 ft RA. Auto callout by the FWC below minimums/decision include "50", "30", then "RETARD". "RETARD" is generated if the Thrust Levers are not at idle by 20 ft (manually flying) or by 10 ft (autoflight engaged).
"AT 40 FT APPROX
"Commence a progressive flare while simultaneously closing the thrust levers; thrust levers to be at idle prior to touchdown. Allow the aircraft to touch down without prolonged float.
"Note: If one or more thrust levers remain above the IDLE detent, ground spoilers extension is inhibited."
The FCOM operational statements for both Airbus types are abundantly clear. The thrust levers are to be at IDLE before touchdown. The "RETARD" auto callout is a reminder for this SOP.
This is how it is taught and how it is checked in recurrent training and line checks. We even had a FOQA event for "Thrust levers not at IDLE at Touchdown".
It is also taught that on a severe bounce, the SOP is to go-around, not try and rescue the landing. I have seen in flight data such bounces followed by go-arounds and they work as described. I have also seen and dealt with the very event which occurred with the SATA A320.
fantom;
Having sat next to countless newbies on their first-ever circuits (base training, we call it here) I must have seen more interesting arrivals than the average but never, as far as I am aware, have the spoilers deployed in the air.
I always understood that they simply cannot.
I always understood that they simply cannot.
It can, and has happened, at least twice. As explained in PBL's post, Airbus has modified the logic of deployment since.
Rather than my taking up bandwidth here, it's better to read the report to understand why this can happen, especially if you're training pilots, (did that too...even "interesting" doesn't describe it adequately at times!)
In post #78, PBL states (accurately), "The point is that the MLG compression is latched for 3 seconds, and ground spoilers come when thrust levers are retarded to idle. " The FCOM "Note" cited above, "Note: If one or more thrust levers remain above the IDLE detent, ground spoilers extension is inhibited", refers directly to this and so the SOP to ensure thrust levers are at IDLE prior to touchdown relates directly to this.
Some fairly significant decisions which had to be made when designing the system, one of which is the logic that confirms to certain systems that the aircraft is landing, and on the ground. This isn't as straightforward as may first seem, so one must understand this, and must understand the system.
BOAC, I'm confused by your comment in re PBL...what is there about swept wing handling that has to do with ensuring the thrust levers are retarded prior touchdown as per SOP? What have I missed?
Last edited by PJ2; 26th Feb 2011 at 17:39.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
PJ - my training was military. As the sweep angle of my aircraft increased through my career the teaching became more and more that retarding throttle/throttles too early could cause a heavy landing. The teaching was, and I'm sure still is, to retard the power WHEN it was appropriate. When I flew it, the teaching for the BAC Lightning with its high sweep was to land with power on. I can recall also being taught on the 737 much later in life NOT to retard the throttles too early, particularly at F40. I have learnt over the years to judge myself when it is appropriate to close the throttles and have never done it at a fixed height! If I have felt I needed power on to touchdown I used it.
It appears that the AB philosophy is different. Boeing also suggest closing the levers in the flare but are not so insistent. I am surprised and glad I never had to learn a new way of landing. I cannot understand the apparent stress on flaring with throttles closed. The spoilers will still extend WHEN they are closed, will they not? The increase in landing run is insignificant. To have an FOQA 'event' for this raises even more doubts in my mind. Why? Because of the software?
I guess CONF and I have similar 'memories'. Ah well. As you say "ignorance is bliss".
It appears that the AB philosophy is different. Boeing also suggest closing the levers in the flare but are not so insistent. I am surprised and glad I never had to learn a new way of landing. I cannot understand the apparent stress on flaring with throttles closed. The spoilers will still extend WHEN they are closed, will they not? The increase in landing run is insignificant. To have an FOQA 'event' for this raises even more doubts in my mind. Why? Because of the software?
I guess CONF and I have similar 'memories'. Ah well. As you say "ignorance is bliss".
Mistrust in Management
As BOAC says
The teaching was, and I'm sure still is, to retard the power WHEN it was appropriate
Regards
Exeng
SOP's
The system behaviour is predicated on a general process which PBL uses the term SOP. He is not wrong in the literal sense, which is broader than the definition you may consider wholly from aircraft operations.
There are numerous examples of such mismatch of crew actions to the expectancies built into the design:
B744 off piste, midfield, NTAA. Crew miss one ATR mode change during the approach phase (ATR originally manual but engages in speed... with appr mode change), and then miss one thrust lever when going into reverse. ATR is managing to speed in the flare, ie increasing thrust.... until such time as disconnected by selection of reverse thrust on a thrust lever.
B744 off piste, end, VTBD. Multi crew inputs, decision changes and thrust levers advanced at touchdown, disarming A/B and SP/BRK's. Thrust lever again missed... "Fore...!"
Note these are TBC product events.
Same sort of issue (mismatches) occur with high ROC altitude captures, where the AP/FD captures a path that is established from a perceived energy state that exceeds the steady state energy capability of the system, results in odd ATR behaviour and as often as not, a minimum speed exceedence event.
SOP? At the system description level.
For the A320... in the flare, (without delving into the software code...) at some point the mode of the ATR needs to inhibit a response to low speed (then again... ). This could occur at the transition point for the flight control laws (pitch) or later at the trigger point for the thrust "retard", or even could have been achieved by a time delay and WOG sensing. All would work to some extent. reviewing flight data for the type would probably allow you to verify the laws... Or try the simulator for an indication (may be an approximation). The B737's logic we worked out from the THY bingle... if it wasn't already understood.
If the operator spends all of their time within the normal envelope, then the actual functioning of a system may be quite different to what is assumed, even after reading the available (limited by SBO) literature on the system behaviour. When things go pear shaped, the exact behaviour of the systems may not be identifiable in time by the operator, or provide adequate opportunity to rectify before cascading failures occur. Add task saturation, inattention/distraction/SA-I failures, then the number of excursions that occur on runways should hardly be a surprise given how we design and teach. Just add fatigue, water, (or other adverse stressors).
There are numerous examples of such mismatch of crew actions to the expectancies built into the design:
B744 off piste, midfield, NTAA. Crew miss one ATR mode change during the approach phase (ATR originally manual but engages in speed... with appr mode change), and then miss one thrust lever when going into reverse. ATR is managing to speed in the flare, ie increasing thrust.... until such time as disconnected by selection of reverse thrust on a thrust lever.
B744 off piste, end, VTBD. Multi crew inputs, decision changes and thrust levers advanced at touchdown, disarming A/B and SP/BRK's. Thrust lever again missed... "Fore...!"
Note these are TBC product events.
Same sort of issue (mismatches) occur with high ROC altitude captures, where the AP/FD captures a path that is established from a perceived energy state that exceeds the steady state energy capability of the system, results in odd ATR behaviour and as often as not, a minimum speed exceedence event.
SOP? At the system description level.
For the A320... in the flare, (without delving into the software code...) at some point the mode of the ATR needs to inhibit a response to low speed (then again... ). This could occur at the transition point for the flight control laws (pitch) or later at the trigger point for the thrust "retard", or even could have been achieved by a time delay and WOG sensing. All would work to some extent. reviewing flight data for the type would probably allow you to verify the laws... Or try the simulator for an indication (may be an approximation). The B737's logic we worked out from the THY bingle... if it wasn't already understood.
If the operator spends all of their time within the normal envelope, then the actual functioning of a system may be quite different to what is assumed, even after reading the available (limited by SBO) literature on the system behaviour. When things go pear shaped, the exact behaviour of the systems may not be identifiable in time by the operator, or provide adequate opportunity to rectify before cascading failures occur. Add task saturation, inattention/distraction/SA-I failures, then the number of excursions that occur on runways should hardly be a surprise given how we design and teach. Just add fatigue, water, (or other adverse stressors).
Psychophysiological entity
LR - it is appropriate to note that this 'experienced' crew did not maintain power after the bounce in order to cushion the return to earth as per 'normal'. Had they done so I guess the boards would not have popped at that height which might have knocked a few 'g' off the second go.
Countless thousands of lines of code have gone into controlling things that used to be called on by micro-switches, yet the system is not intelligent enough to recognize a bounce. Frankly, I'm astonished it could be allowed to happen.
Three seconds is a long time for the MLG to register weight before a bounce, but I can well imagine after being compressed over a good range of travel by a moderately heavy landing, the return travel is hydraulically buffered enough to sum up to that time.
Thinking about it, I suppose the high compression time would be lengthened by inertia of the outer wing sections as they flexed.
Gentlemen;
I anticipated that this explanation had the potential to confuse and drive opinion to corners. If one does not fly the airplane, isn't experienced on it, or doesn't know the systems, it is challenging to offer the notion that this isn't a problem while at the same time "explaining" the airplane.
In fact, the airplane has thousands upon thousands of unremarkable, ordinary, normal landings accompanied by none of the misapprehensions or confused understandings about swept wings, holding on power til the last minute etc.
The system is more subtle than credit is given.
If one needs more thrust at the last moment it is available. One is not along for the ride if one understands and flies the airplane like the beloved B737, etc, ad hoc.
The requirement (SOP) to reduce thrust levers to IDLE prior to touchdown is normal ops. In short-term shear conditions or other conditions which trend towards going below Vls, either the autothrust will repond if the thrust levers are not in the IDLE position, or one can move the thrust levers forward (out of the autothrust regime) and obtain instant thrust. It simply is not an issue, nor is the requirement to have the thrust levers at IDLE prior to touchdown.
I hear the "yes, but's..." offered here, and they simply aren't any more relevant to this airplane than they are to standard Boeings and Lockheeds.
For example, one NEVER closed the throttles on a B727 until the flare was completed. One didn't do it on a heavier DC8 but a light one would float if one didn't push a bit on the stick after flaring too early and rolling it on and not eating up runway. Do that same thing on the L1011 and it would thunder on as pushing would raise the spoilers a bit from their "neutral", extended position. Each type is different.
I've flown Boeings, Douglases, Lockheeds and Airbusi...this simply isn't an issue, IF one understands one's airplane as it may differ from one's previous types, just like "previous previous" types did at one time in one's career...
I do not "defend" the Airbus unconditionally. It has its problems, some of which have not even been guessed at let alone mentioned by occasional contributors. What I resist is not criticism but lack of comprehension of the Airbus A320/A330 series accompanied by the unabashed willingness to comment on something that is only known from reading.
If anyone above has experience flying the A320, tell me and we can engage in fine-point discussion, but I am betting no one here (except CONF iture, who's points I have responded to), who has made the last few posts knows the airplane first-hand through being trained on it and flying it.
The airplane has its interesting features, just like every other airplane we could name here, (except perhaps the L1011-500), and one learns such features and flies one's machine accordingly. It is an airplane...nothing more. And if one is at the boundaries of controlled flight, it, like all airplanes, will be less predictable. It is "how" it is unpredictable where system knowledge and experience come to the fore.
Q.E.D.
fdr;
Re-reading your comments, I believe I understand what you're saying but unfamiliar behaviour outside normal experience can occur on all aircraft and untoward events can occur if SOPs aren't adhered to. If the A320 bounces high and the WOW has latched, closing the thrust levers to idle will satisfy spoiler deployment conditions and that is what happened in the two events I know about. Somewhat related to your thoughts on entering the "retard" mode, the ATR software has been modified such that in any bounce, the spoilers deploy more slowly, (as mentioned in the report).
I think you'll appreciate that there are a number of problems to address which have conflicting requirements, all of which occur rapidly in the landing phase.
BOAC;
There should be no confusion, and why the doubts?....it is an SOP, and because the event (for thrust levers at idle at t/d) was instituted after the TAM A320 accident at Congonhas - the event monitored split levers as well. The event is trended and would be part of on-going reports - no big thing. It's an airplane...one can "squeeze" the power off slowly while flaring, just like any airplane.
PJ2
I anticipated that this explanation had the potential to confuse and drive opinion to corners. If one does not fly the airplane, isn't experienced on it, or doesn't know the systems, it is challenging to offer the notion that this isn't a problem while at the same time "explaining" the airplane.
In fact, the airplane has thousands upon thousands of unremarkable, ordinary, normal landings accompanied by none of the misapprehensions or confused understandings about swept wings, holding on power til the last minute etc.
The system is more subtle than credit is given.
If one needs more thrust at the last moment it is available. One is not along for the ride if one understands and flies the airplane like the beloved B737, etc, ad hoc.
The requirement (SOP) to reduce thrust levers to IDLE prior to touchdown is normal ops. In short-term shear conditions or other conditions which trend towards going below Vls, either the autothrust will repond if the thrust levers are not in the IDLE position, or one can move the thrust levers forward (out of the autothrust regime) and obtain instant thrust. It simply is not an issue, nor is the requirement to have the thrust levers at IDLE prior to touchdown.
I hear the "yes, but's..." offered here, and they simply aren't any more relevant to this airplane than they are to standard Boeings and Lockheeds.
For example, one NEVER closed the throttles on a B727 until the flare was completed. One didn't do it on a heavier DC8 but a light one would float if one didn't push a bit on the stick after flaring too early and rolling it on and not eating up runway. Do that same thing on the L1011 and it would thunder on as pushing would raise the spoilers a bit from their "neutral", extended position. Each type is different.
I've flown Boeings, Douglases, Lockheeds and Airbusi...this simply isn't an issue, IF one understands one's airplane as it may differ from one's previous types, just like "previous previous" types did at one time in one's career...
I do not "defend" the Airbus unconditionally. It has its problems, some of which have not even been guessed at let alone mentioned by occasional contributors. What I resist is not criticism but lack of comprehension of the Airbus A320/A330 series accompanied by the unabashed willingness to comment on something that is only known from reading.
If anyone above has experience flying the A320, tell me and we can engage in fine-point discussion, but I am betting no one here (except CONF iture, who's points I have responded to), who has made the last few posts knows the airplane first-hand through being trained on it and flying it.
The airplane has its interesting features, just like every other airplane we could name here, (except perhaps the L1011-500), and one learns such features and flies one's machine accordingly. It is an airplane...nothing more. And if one is at the boundaries of controlled flight, it, like all airplanes, will be less predictable. It is "how" it is unpredictable where system knowledge and experience come to the fore.
Q.E.D.
fdr;
Re-reading your comments, I believe I understand what you're saying but unfamiliar behaviour outside normal experience can occur on all aircraft and untoward events can occur if SOPs aren't adhered to. If the A320 bounces high and the WOW has latched, closing the thrust levers to idle will satisfy spoiler deployment conditions and that is what happened in the two events I know about. Somewhat related to your thoughts on entering the "retard" mode, the ATR software has been modified such that in any bounce, the spoilers deploy more slowly, (as mentioned in the report).
I think you'll appreciate that there are a number of problems to address which have conflicting requirements, all of which occur rapidly in the landing phase.
BOAC;
To have an FOQA 'event' for this raises even more doubts in my mind. Why?
PJ2
Last edited by PJ2; 27th Feb 2011 at 07:13. Reason: add comments for fdr and BOAC
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
PJ
- fully understood, but I have to say that if that is problem for you, you are on the wrong aviation forum! Unless trained on the a/c that is all some posters have, and your 'resistance' would probably be better applied to a pure AB forum?
As for
- I can assure you after many years flying such the confusion is not mine. 'Once bitten, twice shy', we say.
What raises the hairs on the back of my neck (fortunately only when 'reading' about it, is the drive to automation of everything (cue my sporadic thread in 'Safety'). To have to introduce a system that 'reminds' a pilot to close the thrust levers is ludicrous, on any aeroplane. How long before we have a beeper that goes off or inhibits throttle movement when you taxy without your seat belts done up properly? As I have repeatedly said on the other thread and elsewhere, the more we 'automate' flying the less pilots are aware of what they and the aircraft systems are doing, the less 'training' that will be required until we end up with pilots who only know that they should close the throttles on landing because the aircraft tells them to. A corollary is that the more and more software loops we have to remind us to do up our flies as we exit the toilet (Mrs B does that....) the more scope there is for faulty programming, un-predicted 'gotchas'/loop inter-action until the toilet door will not open if your 'zipper' is broken. The thought of being 'flagged' in some eager young manager's office for landing with power on one day is simply stunning. I disagree that landing with throttles closed is an 'SOP' - that is a mis-use of a term. It may indeed be 'normal practice' but to suggest that NOT doing it for whatever reason is automatically a 'breach' of SOPs............................
What I resist is not criticism but lack of comprehension of the Airbus A320/A330 series accompanied by the unabashed willingness to comment on something that is only known from reading.
As for
misapprehensions or confused understandings about swept wings,
What raises the hairs on the back of my neck (fortunately only when 'reading' about it, is the drive to automation of everything (cue my sporadic thread in 'Safety'). To have to introduce a system that 'reminds' a pilot to close the thrust levers is ludicrous, on any aeroplane. How long before we have a beeper that goes off or inhibits throttle movement when you taxy without your seat belts done up properly? As I have repeatedly said on the other thread and elsewhere, the more we 'automate' flying the less pilots are aware of what they and the aircraft systems are doing, the less 'training' that will be required until we end up with pilots who only know that they should close the throttles on landing because the aircraft tells them to. A corollary is that the more and more software loops we have to remind us to do up our flies as we exit the toilet (Mrs B does that....) the more scope there is for faulty programming, un-predicted 'gotchas'/loop inter-action until the toilet door will not open if your 'zipper' is broken. The thought of being 'flagged' in some eager young manager's office for landing with power on one day is simply stunning. I disagree that landing with throttles closed is an 'SOP' - that is a mis-use of a term. It may indeed be 'normal practice' but to suggest that NOT doing it for whatever reason is automatically a 'breach' of SOPs............................
Join Date: Aug 2002
Location: Up north
Posts: 1,657
Likes: 0
Received 0 Likes
on
0 Posts
even when its calm, lisbon can be sporty....when its windy, one has to be at the top of their game....period.
Disadvantages of A/THR approaches on Airbus FBW
Quote from BOAC (Post #80):
"LR - it is appropriate to note that this 'experienced' crew did not maintain power after the bounce in order to cushion the return to earth as per 'normal'."
and from CONF_iture (Post #81):"Another example of the poor A/THR response for a late wind change. A bit of thrust under manual thrust would have helped to improve the first touchdown and avoid what came next."
As a Boeing pilot (?), what BOAC may not appreciate is the difficulty involved in doing this on Airbus FBW aeroplanes when A/THR is engaged. (Have to confess to having not read the report, but am assuming that the PF was using A/THR, which is – sadly – the SOP in most airlines.) The reason for this, he will now recall, is that the non-driven throttles (thrust levers) are both stationary in the Climb-Thrust detent throughout the approach. If left there during the flare, the thrust will normally tend to rise to correct the airspeed decay. This is usually not desirable: hence the pilot normally retards the thrust levers as he/she commences the flare.
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).
On the other hand, once the thrust levers have been closed, the A/THR is disengaged automatically, and you are in conventional manual thrust. Unfortunately, most Airbus FBW pilots get little or no practice at using this (except for taxiing), so are ill-prepared if they suddenly need an increase in thrust in the flare.
This is why CONF_iture and I advocate the use of manual thrust for non-AP (visual) approaches, provided visual contact is acquired in time to carry out the slightly fiddly, time-consuming process of disengaging the autothrottle. This involves initially retarding the throttles to a point where the thrust limit matches the existing thrust (using the N1 or EPR gauges, as appropriate to the engine type) – so as to avoid any undesired thrust change – and then disengaging the A/THR. The thrust levers then work conventionally, and are second to none in my experience. The GS-Mini system helps enormously to minimise thrust changes in windshear, provided you remember that it provides a target speed, NOT a minimum speed.
"LR - it is appropriate to note that this 'experienced' crew did not maintain power after the bounce in order to cushion the return to earth as per 'normal'."
and from CONF_iture (Post #81):"Another example of the poor A/THR response for a late wind change. A bit of thrust under manual thrust would have helped to improve the first touchdown and avoid what came next."
As a Boeing pilot (?), what BOAC may not appreciate is the difficulty involved in doing this on Airbus FBW aeroplanes when A/THR is engaged. (Have to confess to having not read the report, but am assuming that the PF was using A/THR, which is – sadly – the SOP in most airlines.) The reason for this, he will now recall, is that the non-driven throttles (thrust levers) are both stationary in the Climb-Thrust detent throughout the approach. If left there during the flare, the thrust will normally tend to rise to correct the airspeed decay. This is usually not desirable: hence the pilot normally retards the thrust levers as he/she commences the flare.
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).
On the other hand, once the thrust levers have been closed, the A/THR is disengaged automatically, and you are in conventional manual thrust. Unfortunately, most Airbus FBW pilots get little or no practice at using this (except for taxiing), so are ill-prepared if they suddenly need an increase in thrust in the flare.
This is why CONF_iture and I advocate the use of manual thrust for non-AP (visual) approaches, provided visual contact is acquired in time to carry out the slightly fiddly, time-consuming process of disengaging the autothrottle. This involves initially retarding the throttles to a point where the thrust limit matches the existing thrust (using the N1 or EPR gauges, as appropriate to the engine type) – so as to avoid any undesired thrust change – and then disengaging the A/THR. The thrust levers then work conventionally, and are second to none in my experience. The GS-Mini system helps enormously to minimise thrust changes in windshear, provided you remember that it provides a target speed, NOT a minimum speed.
Last edited by Chris Scott; 27th Feb 2011 at 13:26.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Useful, thanks Chris. On the 737 'min speed reversion' for autothrottle in 'armed' mode was similar, but disengagement of the A/T was just the click of a button - no 'fiddling' required.
Join Date: Jul 2008
Location: UK
Age: 69
Posts: 475
Likes: 0
Received 0 Likes
on
0 Posts
There of course a number of issues with this incident but......
what sort of industry professional would ground an aircraft because he may have lost his flight back or a couple of maps but continues to fly although there is missing engineering data.
I am sorry to be so blunt but you are your own worst enemies. Even only a suspected hard landing requires a tech log entry and a release to service.
what sort of industry professional would ground an aircraft because he may have lost his flight back or a couple of maps but continues to fly although there is missing engineering data.
I am sorry to be so blunt but you are your own worst enemies. Even only a suspected hard landing requires a tech log entry and a release to service.
Psychophysiological entity
PJ, It really is a forum for a broad spectrum of interested parties. Trying to sort the facts from fiction means relying on qualified people who know the type, people like you, and indeed for us dinosaurs, people who are up to speed on modern systems. The odd confused statement/question is inevitable.
I said:
It seems the system can recognize a bounce, but then does something I wouldn't do in a million years. The designers must know something I don't. Maybe, a thing or two.
You said:
Thinking aloud, I suggested two reasons why the three seconds of oleo compression requirement might well be met in even a modest bounce, but as I imply, I'm still astonished to read that anything is allowed to diminish lift during the following crucial moments.
Again, just trying to understand. A bounce can be quite a long affair in time/distance, so I assume any deployment has be a delicate balance between containing the bounce, and the fear of running out of concrete.
I said:
. . . yet the system is not intelligent enough to recognize a bounce.
You said:
If the A320 bounces high and the WOW has latched, closing the thrust levers to idle will satisfy spoiler deployment conditions and that is what happened in the two events I know about. Somewhat related to your thoughts on entering the "retard" mode, the ATR software has been modified such that in any bounce, the spoilers deploy more slowly, (as mentioned in the report).
Again, just trying to understand. A bounce can be quite a long affair in time/distance, so I assume any deployment has be a delicate balance between containing the bounce, and the fear of running out of concrete.
Join Date: Apr 1999
Location: Canada
Age: 82
Posts: 91
Likes: 0
Received 0 Likes
on
0 Posts
Very interesting discussion. PJ2 and Chris Scott have it right, every aircraft type has its own characteristics. BOAC may have a point regarding different degrees of wing sweep, I do not have his experience - all the civilian transports that I flew had essentially the same amount of sweep.
What is important is the overall character of the wing on approach. Early models all approached with a lot of trailing edge flap extended, and sometimes a not very sophisticated TE design, hence a lot of drag demanding a lot of power. Close the throttles early in those types and you dropped like a rock. Same was true of the B-727 and the DC-9-30 series, if you were right on the bug you did not want to get the power back until very close to the ground, if not already touching down. The DC-9-10 series, however, with no LE slat, would float forever so you had to retard the the power before touchdown if you wanted to stop on a limiting runway. As an aside, my mob decided to operate the "Little -9" on "Little airports" with little runways - not a lot of fun especially in the winter! Boffins!
What has happened as jet transports evolved is a much cleaner, low drag wing on approach resulting in lower thrust levels (fuel saving and noise reduction) allowing the power to be reduced in the flare before touchdown. In fact in all the AB airplanes that I flew you had to get the power back if you wanted to touchdown in the desired runway zone. When I first went on the A-320 an F/O told me that he had observed one particular captain who consistently got the best and smoothest landings - he never heard the retard call with him. That's when we began to realize that the call was not an instruction, it was telling you that you had not done something.
Never flew the B-737 but the 747 and 767 had so much wing out there that you could easily (unless really heavy) get the power back, the nose up and run it on on the back side of the tires, nice and smooth.
My two cents worth.
What is important is the overall character of the wing on approach. Early models all approached with a lot of trailing edge flap extended, and sometimes a not very sophisticated TE design, hence a lot of drag demanding a lot of power. Close the throttles early in those types and you dropped like a rock. Same was true of the B-727 and the DC-9-30 series, if you were right on the bug you did not want to get the power back until very close to the ground, if not already touching down. The DC-9-10 series, however, with no LE slat, would float forever so you had to retard the the power before touchdown if you wanted to stop on a limiting runway. As an aside, my mob decided to operate the "Little -9" on "Little airports" with little runways - not a lot of fun especially in the winter! Boffins!
What has happened as jet transports evolved is a much cleaner, low drag wing on approach resulting in lower thrust levels (fuel saving and noise reduction) allowing the power to be reduced in the flare before touchdown. In fact in all the AB airplanes that I flew you had to get the power back if you wanted to touchdown in the desired runway zone. When I first went on the A-320 an F/O told me that he had observed one particular captain who consistently got the best and smoothest landings - he never heard the retard call with him. That's when we began to realize that the call was not an instruction, it was telling you that you had not done something.
Never flew the B-737 but the 747 and 767 had so much wing out there that you could easily (unless really heavy) get the power back, the nose up and run it on on the back side of the tires, nice and smooth.
My two cents worth.
Join Date: Apr 2010
Location: New Forest
Posts: 143
Likes: 0
Received 0 Likes
on
0 Posts
Humour
To lighten the mood of the thread somewhat (after all, do you really need SOP's to tell you how to land your aircraft -if so maybe you should be re-considering your position), I offer this in reply to BOAC's #92.
Late night and dark. 757 and the forward galley lights go out if the flight deck door is opened. Disorientated passenger needs a desperate call of nature. He opens the flight deck door (pre all this nonsense) but suddenly becomes aware of his surroundings. Disregarding the (absent) instructions which always used to be displayed in Public Lavatories (and I remember asking, "Mummy, what does that mean?!!!") - "Please adjust your dress before leaving", the unfortunate passenger retreated in a hurry, slammed the door and turned around. On came the lights and there he was for all to see. Wonderful. How did Mr Boeing think of that?
Prober
Late night and dark. 757 and the forward galley lights go out if the flight deck door is opened. Disorientated passenger needs a desperate call of nature. He opens the flight deck door (pre all this nonsense) but suddenly becomes aware of his surroundings. Disregarding the (absent) instructions which always used to be displayed in Public Lavatories (and I remember asking, "Mummy, what does that mean?!!!") - "Please adjust your dress before leaving", the unfortunate passenger retreated in a hurry, slammed the door and turned around. On came the lights and there he was for all to see. Wonderful. How did Mr Boeing think of that?
Prober
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes
on
0 Posts
PJ2
I stand corrected, even if I think that that kind of stuff involving basic flying technique and airmanship should not load a SOP manual.
To be fully configured by 1000 feet on the way down or to operate the thrust reduction at a specified altitude on the way up are more kind of things that qualify for SOP designation, in my mind.
It is not what I understand. The ground spoilers will still fully deploy if thrust levers are retarded in the air following a bounce, the modification is, in the aim to reduce or cancel any bounce, that the ground spoilers will partially deploy at initial touchdown, even if the thrust levers are still in the CLB detent.
The message that Airbus sends could be the following :
No matter how badly you bring the aircraft to its initial touchdown, our automatism will make sure it stays there.
As BOAC put it : the more we 'automate' flying the less pilots are aware of what they and the aircraft systems are doing
To me that full deployment of ground spoilers in the air is a total heresy.
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 stand corrected, even if I think that that kind of stuff involving basic flying technique and airmanship should not load a SOP manual.
To be fully configured by 1000 feet on the way down or to operate the thrust reduction at a specified altitude on the way up are more kind of things that qualify for SOP designation, in my mind.
the ATR software has been modified such that in any bounce, the spoilers deploy more slowly, (as mentioned in the report).
The message that Airbus sends could be the following :
No matter how badly you bring the aircraft to its initial touchdown, our automatism will make sure it stays there.
As BOAC put it : the more we 'automate' flying the less pilots are aware of what they and the aircraft systems are doing
To me that full deployment of ground spoilers in the air is a total heresy.
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 !
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Likes: 0
Received 0 Likes
on
0 Posts
I am not sure how much I have to contribute to this thread, and in this I am at one with CONFiture and BOAC, but for a different reason.
I first published on the A320 air-ground logic eighteen years ago, and for a specific period of time a few years ago I probably understood more about the hazard analysis of it than anyone else on the planet. So when I mention SOPs about retarding throttles to idle, and am contradicted by CONFiture, I wonder how someone close to these airplanes can be so unaware of the series of continuing measures required by EASA from February 2008 on to address the air-ground transition on landing.
I would also have thought that someone who spent an entire career flying 50's and 60's era high-performance designs, as BOAC tells us he has done, would maybe understand that those intuitions and vast experience do not necessarily transfer unchanged to high-performance civil aircraft from the 80's and 90's. For one thing, the later designs are all fly-by-wire. All of them.
It is a different design paradigm, and there are issues that simply don't arise with older designs, as I thought should have been obvious to everyone after the Malaysian Airlines 2005 experience with a Boeing 777 and Qantas's experience in 2008 with two A330's, let alone the Byzantine-failure phenomenon which almost led to the airworthiness certificate of one of the most-used FBW civil aircraft being withdrawn. The term "Byzantine failure" refers exclusively to distributed-digital-system (mis)behavior. The Space Shuttle got some of it, too. Hands up those here who have a clue what it is. I thought so.
BOAC contends that an explicit requirement to retard throttles on landing is "ludicrous", and CONFiture agrees that such stuff should not need to be written down in manuals. It is well to recall that the manufacturer of this aircraft agreed with this view wholeheartedly at least until early 2008 (those who need to review this history can check the testimony of Yannick Malinge before the Brazilian parliament in August 2007 - but one does need to be able to read Portuguese). However, EASA obviously does not agree with this point of view, starting as I mentioned in mid-February 2008 and continuing until the present. Unlike BOAC and CONFiture, I imagine (no, I don't imagine, I know) that EASA has good reasons for going this line.
One can discuss the wisdom, or ludicrousness, of instructions to pilots about "basic flying technique" until the cows come home, of course, and one would be continuing to miss the point entirely. None of this instruction about throttles is *necessary*. After you point the airplane at the TDZ at a couple thousand feet altitude, it can do everything else itself very reliably, including the energy management, and it seems to me, given some recent experiences including the subject of the current thread, that we might all be better off if it always did.
What experienced pilots are trying to have a go at here is just what we try to show beginning students of computer science doesn't work (unfortunately, many of them don't learn the lesson very well). You can design a piece of SW or logic to handle situation A, almost perfectly. Let's call the result D_A. Then there is situation B. You think of how D_A can handle situation B, and end up with the resulting modified design D_(A+B). Then you think about situation C .... and so on. It doesn't work, because, by the time you have thought of about 20 different situations and calculated the time you'll need to do that, you learn that not only is the semester long finished but the sun has swallowed the earth back up. And there aren't twenty different situations, there are qualitatively hundreds of them, plus those that come up during ops that you hadn't addressed at design time.
What you do is design a system according to certain principles, and one of those design points may be that one retards throttles to idle upon or slightly before touchdown (as our experienced pilots say - and said to the manufacturer at design time - that one *obviously* does, ignoring of course residual saurial whimpering about swept-wing behavior of 50's-60's era jets).
The question which then arises is what you then do, as an airplane manufacturer or regulator, if it turns out that not everyone who flies your airplanes does what they should "obviously" do? (Let us remember that there are a half-dozen cases of it in the public record, and there are people on this list who could, but won't, tell you about the non-public record.) Well, I'd tell you what I'd be very tempted to do if I were King of Megajets Inc. I would make a great effort to introduce more automatics and the pilot-plus-large-dog concept of crew. Then we can imagine PPRuNe in thirty years with all the retired pilots reminiscing of the good old days when they were allowed to touch something.
I first published on the A320 air-ground logic eighteen years ago, and for a specific period of time a few years ago I probably understood more about the hazard analysis of it than anyone else on the planet. So when I mention SOPs about retarding throttles to idle, and am contradicted by CONFiture, I wonder how someone close to these airplanes can be so unaware of the series of continuing measures required by EASA from February 2008 on to address the air-ground transition on landing.
I would also have thought that someone who spent an entire career flying 50's and 60's era high-performance designs, as BOAC tells us he has done, would maybe understand that those intuitions and vast experience do not necessarily transfer unchanged to high-performance civil aircraft from the 80's and 90's. For one thing, the later designs are all fly-by-wire. All of them.
It is a different design paradigm, and there are issues that simply don't arise with older designs, as I thought should have been obvious to everyone after the Malaysian Airlines 2005 experience with a Boeing 777 and Qantas's experience in 2008 with two A330's, let alone the Byzantine-failure phenomenon which almost led to the airworthiness certificate of one of the most-used FBW civil aircraft being withdrawn. The term "Byzantine failure" refers exclusively to distributed-digital-system (mis)behavior. The Space Shuttle got some of it, too. Hands up those here who have a clue what it is. I thought so.
BOAC contends that an explicit requirement to retard throttles on landing is "ludicrous", and CONFiture agrees that such stuff should not need to be written down in manuals. It is well to recall that the manufacturer of this aircraft agreed with this view wholeheartedly at least until early 2008 (those who need to review this history can check the testimony of Yannick Malinge before the Brazilian parliament in August 2007 - but one does need to be able to read Portuguese). However, EASA obviously does not agree with this point of view, starting as I mentioned in mid-February 2008 and continuing until the present. Unlike BOAC and CONFiture, I imagine (no, I don't imagine, I know) that EASA has good reasons for going this line.
One can discuss the wisdom, or ludicrousness, of instructions to pilots about "basic flying technique" until the cows come home, of course, and one would be continuing to miss the point entirely. None of this instruction about throttles is *necessary*. After you point the airplane at the TDZ at a couple thousand feet altitude, it can do everything else itself very reliably, including the energy management, and it seems to me, given some recent experiences including the subject of the current thread, that we might all be better off if it always did.
What experienced pilots are trying to have a go at here is just what we try to show beginning students of computer science doesn't work (unfortunately, many of them don't learn the lesson very well). You can design a piece of SW or logic to handle situation A, almost perfectly. Let's call the result D_A. Then there is situation B. You think of how D_A can handle situation B, and end up with the resulting modified design D_(A+B). Then you think about situation C .... and so on. It doesn't work, because, by the time you have thought of about 20 different situations and calculated the time you'll need to do that, you learn that not only is the semester long finished but the sun has swallowed the earth back up. And there aren't twenty different situations, there are qualitatively hundreds of them, plus those that come up during ops that you hadn't addressed at design time.
What you do is design a system according to certain principles, and one of those design points may be that one retards throttles to idle upon or slightly before touchdown (as our experienced pilots say - and said to the manufacturer at design time - that one *obviously* does, ignoring of course residual saurial whimpering about swept-wing behavior of 50's-60's era jets).
The question which then arises is what you then do, as an airplane manufacturer or regulator, if it turns out that not everyone who flies your airplanes does what they should "obviously" do? (Let us remember that there are a half-dozen cases of it in the public record, and there are people on this list who could, but won't, tell you about the non-public record.) Well, I'd tell you what I'd be very tempted to do if I were King of Megajets Inc. I would make a great effort to introduce more automatics and the pilot-plus-large-dog concept of crew. Then we can imagine PPRuNe in thirty years with all the retired pilots reminiscing of the good old days when they were allowed to touch something.
Last edited by PBL; 28th Feb 2011 at 07:40.
Join Date: Oct 2010
Location: U.K.
Posts: 529
Likes: 0
Received 0 Likes
on
0 Posts
Another point to the Airbus, expanding on the 20ft 'retard' call. It is a guide not a rule. With a 10Kt tail wind the thrust levers can and need to be closed at 30-40 feet. Whilst with a 30 knot headwind I would keept the thrust to maybe 10 feet. For a crosswind we add an increment onto Vapp. Flight is analog not digital though the figures are an excellent guide and in normal ops correct. The other point is that the flare is not 'fire and forget' - the aircraft needs to be flown down to the ground. At the start of my airline flying I would literally cut the thrust at 20, raise the nose a cm and wait and see what happened. With time and a will to do better I have tried to adjust - though for a week a month my landings still go to sh1t - and usually when I am flying with a captain I haven't flown with before!
By the sounds of it this chap made a split second error of judgement. It happens, he is human. He is now more experienced in it than most of us. Keep learning, keep caring and keep searching for perfection.
By the sounds of it this chap made a split second error of judgement. It happens, he is human. He is now more experienced in it than most of us. Keep learning, keep caring and keep searching for perfection.