Emirates B777 gear collapse @ DXB?
Join Date: Oct 2004
Location: York
Posts: 737
Likes: 0
Received 0 Likes
on
0 Posts
Ken
Believe it or not, I have 'been around the block' a bit myself too. It's absolutely fundamental that if the a/c isn't doing what you want it to, then you MUST intervene! We all must posses the skill to take over and fly manually. Unfortunately it's also true that for us LH pilots, those skills, even with the best will in the world are likely to become a little rusty!
I think we can agree on all that?
However
Pulling on the yoke will just disconnect the autopilot won't it? On the odd occasion you'd allow it to be engaged?
In the words of Sergeant Wilson. "Are you sure that's wise"?
(Sergeant Wilson roll eyes!)
SOMETIMES it will be. SOMETIMES it won't make much difference, but SOMETIMES it'really not smart!
Believe it or not, I have 'been around the block' a bit myself too. It's absolutely fundamental that if the a/c isn't doing what you want it to, then you MUST intervene! We all must posses the skill to take over and fly manually. Unfortunately it's also true that for us LH pilots, those skills, even with the best will in the world are likely to become a little rusty!
I think we can agree on all that?
However
I believe pilots should be trained to always manually establish the climb by pulling on the yoke and pushing on the thrust levers while pressing the TOGA switch when doing Go Arounds/Rejected Landings in Boeing products
In the words of Sergeant Wilson. "Are you sure that's wise"?
(Sergeant Wilson roll eyes!)SOMETIMES it will be. SOMETIMES it won't make much difference, but SOMETIMES it'really not smart!
Join Date: Oct 2004
Location: York
Posts: 737
Likes: 0
Received 0 Likes
on
0 Posts
RAT5
But the TOGA buttons SHOULDN'T have been pressed here! The thrust levers should have been PUSHED forward!
The WRONG procedure was used for this particular manouvre!
As you say, back where we started!
But the TOGA buttons SHOULDN'T have been pressed here! The thrust levers should have been PUSHED forward!
The WRONG procedure was used for this particular manouvre!
As you say, back where we started!
Many views in this thread are naturally influenced by outcome bias and the assumption that crews will understand a range of alternative situations (see ref).
Discussions like this use 'post-mortem' thinking and often conclude that an accident can be avoided by considering the most salient feature; the close proximity of human activity - pilots, 'probable cause', cause ~ effect, that all errors are preventable. Blame and train.
In order to avoid or reduce the severity of similar accidents we need 'pre-mortem' thinking, to consider 'what if', what consequences, and what assumptions have been made.
E.G.
Why use the RAAS 'smart landing' add-on, this provides positional awareness, but what is assumed about its performance and safety benefit.
The system only alerts in the air, except where the 'in the air' changeover might occur on the ground, WoW logic.
RAAS manuals recommend "Confirm aircraft position and initiate go-around if appropriate." But what is appropriate, can the pilot establish position with better accuracy than the machine ... is the machine really that smart.
What alerting distance has been chosen, why, what assumptions have been made vs likely runway lengths, weight / wind conditions.
Operators can choose to mandate a Go-Around for any alert, use the GA SOP, it will work in the air, but if on the ground then a new procedure is required; a rejected landing, but is this already used for a bounced landing, are these situations and the SOPs the same; complexity, confusion, is it reasonable for the pilot to sort this out after an alert.
What training is given, system knowledge and assumptions. Is GA training usually given with an engine failure, requiring prompt action, which may not be consistent with a less hurried situation in a rejected landing which is more like a takeoff.
And so on ... and then what has been overlooked.
Oh, What if a pilot upgrades from a RAAS aircraft to a ROPS equipped aircraft.
Comments on the above from 737 or 'E' aircraft already using RAAS, or even from the ROPS operators.
Ref Errors in Aviation Decision Making.
Discussions like this use 'post-mortem' thinking and often conclude that an accident can be avoided by considering the most salient feature; the close proximity of human activity - pilots, 'probable cause', cause ~ effect, that all errors are preventable. Blame and train.
In order to avoid or reduce the severity of similar accidents we need 'pre-mortem' thinking, to consider 'what if', what consequences, and what assumptions have been made.
E.G.
Why use the RAAS 'smart landing' add-on, this provides positional awareness, but what is assumed about its performance and safety benefit.
The system only alerts in the air, except where the 'in the air' changeover might occur on the ground, WoW logic.
RAAS manuals recommend "Confirm aircraft position and initiate go-around if appropriate." But what is appropriate, can the pilot establish position with better accuracy than the machine ... is the machine really that smart.
What alerting distance has been chosen, why, what assumptions have been made vs likely runway lengths, weight / wind conditions.
Operators can choose to mandate a Go-Around for any alert, use the GA SOP, it will work in the air, but if on the ground then a new procedure is required; a rejected landing, but is this already used for a bounced landing, are these situations and the SOPs the same; complexity, confusion, is it reasonable for the pilot to sort this out after an alert.
What training is given, system knowledge and assumptions. Is GA training usually given with an engine failure, requiring prompt action, which may not be consistent with a less hurried situation in a rejected landing which is more like a takeoff.
And so on ... and then what has been overlooked.
Oh, What if a pilot upgrades from a RAAS aircraft to a ROPS equipped aircraft.
Comments on the above from 737 or 'E' aircraft already using RAAS, or even from the ROPS operators.
Ref Errors in Aviation Decision Making.
Join Date: Jun 2008
Location: Cambridge UK
Posts: 192
Likes: 0
Received 0 Likes
on
0 Posts
Apparently company SOP was that the pilots abort/reject the landing when getting a RAAS alert. No option.
That's the point here. These pilots had apparently been trained/conditioned to be system operators and
procedure monkeys. They were neither trained nor permitted to be thinking pilots.
That's the point here. These pilots had apparently been trained/conditioned to be system operators and
procedure monkeys. They were neither trained nor permitted to be thinking pilots.
Were the SOP writers remiss in not mentioning the TOGO differences between a go-around and a rejected landing?
They certainly had ample time to explore the possible outcomes.
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,407
Likes: 0
Received 0 Likes
on
0 Posts
I am in full agreement with 4468 on his posts 1695 and 1698. There are two GA procedures and they are different. You must brief the procedure for GA after touch down since. Since you practice normal GA at minimum and automation covers it brief by saying standard GA. When forced to do something unusual chance of error is higher as happened here. For those who find briefing one minute extra which doesn't cost anything is too much how can they justify practicing hours in SIM for the same purpose. And if you question how many things will you brief then how many things will practice in two hours?
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
But the TOGA buttons SHOULDN'T have been pressed here! The thrust levers should have been PUSHED forward!
The WRONG procedure was used for this particular manouvre!
The crux of my thinking is this: you are under stress and select an automatic function e.g. press TOGA and expect, in error, the TL's to advance. Being suspicious of these new fangled bells and whistles you follow through and discover PDQ that somehow, but not to wonder about, the electrons didn't link up as expected and the dastardly TL's did not do as instructed but went AWOL. You give them a clip round the ear and insert your digit up their backside. Hey presto the engines spring into life and resume their normal function. All in the twinkling of an eye. Problem solved. To be discussed on the ground/bar.
The thought of many is that what is missing from too many pilots is having blind belief in what you have selected will indeed happen, and during critical phases your hands are in the wrong place incase it doesn't. This suspicious nature would have saved the day. It's not that he shouldn't have pressed the TOGA, it is that when he did and it didn't do what he wanted then he should have done what was necessary. IMHO the tactile feedback of lack of movement would be far more valuable than an FMA display telling you it hadn't responded.
I've nowt more to say on the matter, but it might be an on going ramble for Clegg, Compo & Foggy over a drawn out cuppa in the cafe. Nora Batty will be the adjudicator and she'll publish her report "oh do stop messing about and get on with it" in about 3 years.
The WRONG procedure was used for this particular manouvre!
The crux of my thinking is this: you are under stress and select an automatic function e.g. press TOGA and expect, in error, the TL's to advance. Being suspicious of these new fangled bells and whistles you follow through and discover PDQ that somehow, but not to wonder about, the electrons didn't link up as expected and the dastardly TL's did not do as instructed but went AWOL. You give them a clip round the ear and insert your digit up their backside. Hey presto the engines spring into life and resume their normal function. All in the twinkling of an eye. Problem solved. To be discussed on the ground/bar.
The thought of many is that what is missing from too many pilots is having blind belief in what you have selected will indeed happen, and during critical phases your hands are in the wrong place incase it doesn't. This suspicious nature would have saved the day. It's not that he shouldn't have pressed the TOGA, it is that when he did and it didn't do what he wanted then he should have done what was necessary. IMHO the tactile feedback of lack of movement would be far more valuable than an FMA display telling you it hadn't responded.
I've nowt more to say on the matter, but it might be an on going ramble for Clegg, Compo & Foggy over a drawn out cuppa in the cafe. Nora Batty will be the adjudicator and she'll publish her report "oh do stop messing about and get on with it" in about 3 years.
Join Date: Aug 2014
Location: New Braunfels, TX
Age: 70
Posts: 1,954
Likes: 0
Received 0 Likes
on
0 Posts
It's absolutely fundamental that if the a/c isn't doing what you want it to, then you MUST intervene!
KenV, "... cannot conceive of a pilot doing a take off on autothrust .... or go-around ... etc" (#1671)
Disagreements, opinions are influenced by differing abilities to imagine* situations and understand that some 'unimaginable' ones, against all our beliefs and experiences will be encountered.
Disagreements, opinions are influenced by differing abilities to imagine* situations and understand that some 'unimaginable' ones, against all our beliefs and experiences will be encountered.
Pulling on the yoke will just disconnect the autopilot won't it? On the odd occasion you'd allow it to be engaged?
In the words of Sergeant Wilson. "Are you sure that's wise"? (Sergeant Wilson roll eyes!)
SOMETIMES it will be. SOMETIMES it won't make much difference, but SOMETIMES it'really not smart!
In the words of Sergeant Wilson. "Are you sure that's wise"?
(Sergeant Wilson roll eyes!)SOMETIMES it will be. SOMETIMES it won't make much difference, but SOMETIMES it'really not smart!
The only time I've ever flown "hands off" is during the 150 ft or so of the cat stroke off of aircraft carriers. And that is because my airplane is attached to a catapult and it does not matter what I do; that airplane is going off the end of the flight deck no matter what. But even then, I place my throttle hand immediately behind the throttles to make sure they don't slide back during the cat stroke. In A-4s I put my right hand in a fist in my gut so when the stick traveled aft during the stroke I would not get smacked by it and also to make sure my hand would be inches away from it as soon as the stroke ended. On the F/A-18 my right hand grabbed a handle on the right side of the canopy, but as soon as the cat stroke ended, that hand was on the stick, irrespective of what the automatics were doing.
Join Date: Oct 2015
Location: Tranquility Base
Posts: 77
Likes: 0
Received 0 Likes
on
0 Posts
We are finally getting to the main points in the discussion. Lots of good contributions in my opinion.
1. Moving thrust levers seem not to help realising that thrust is not coming. So we could get rid of them altogether. As an industry we need to get the basics back in shape, and that means to manipulate thrust by hand when you fly by hand. Athrust only when on autopilot or in very special circumstances.
I would bet that the PF's speed scanning was slow due to yearlong flying with Athrust, so he was late in adding thrust. Additionally he did not have a habit of pushing TL forward in a manual go around situation. We need manual flight to equal manual thrust.
2. The discussion urges that the PF used the wrong procedure. A situation where the applicable procedure changes within milliseconds (weight on weels) so drastically is fundamentally wrong. Either you should train all go arounds (inkl. WoW ones) with manual TL advancement and TOGA-buttons right after (which would automatically be the case when manual flight would mean manual thrust).
Or Boeing should redesign their design so that mode selection is integrated into TL movement, so you have one single procedure designed into the machine and not this trapping inhibition of the regularly used buttons.
3. This accident would not have happened if they would not have gone around (I do not want to speculate if they went around due to RAAS alert or just due to flare beyond TDZ). That is crass, but something to look at.
Flight data analysis shows that once in a while there is a long landing. It does not become an incident because the runway offers enough margin. The SOP-disregard of touching down a bit long does not lead to anything adverse, the system is robust.
When you now introduce the SOP that upon a RAAS (or ROPS) alert, a crew absolutely must go around, you will increase the amount of low or weight on wheels (touch after GA decision) go arounds by a big factor.
Low or weight on wheels go arounds are one of the most challenging manoeuvers. You might need to disregard config warning, you need to make sure thrust is there (on Boeings), you need not to rotate too quickly (tailstrike risk), you might need to sit out a bounce with a safe pitch, and only after all that you then start the real normal go around.
That means if we introduce RAAS (or ROPS) as a mandatory to follow tool, we abolutely need to make sure crews are capable of low or weight on wheels go arounds because we will have many more of them. Otherwise we create a bigger problem than we solve.
The A380 diversion out of MAN to LHR comes to my mind due to software bug in ROPS leading to false alert. It was clearly stated in the discussion that the crew concerned would have had to face a talk with the boss if they would have exerted normal judgement by disregarding the wrong warning and just landed on the second approach. Trained monkeys are never good in a cockpit, it creates new risks.
1. Moving thrust levers seem not to help realising that thrust is not coming. So we could get rid of them altogether. As an industry we need to get the basics back in shape, and that means to manipulate thrust by hand when you fly by hand. Athrust only when on autopilot or in very special circumstances.
I would bet that the PF's speed scanning was slow due to yearlong flying with Athrust, so he was late in adding thrust. Additionally he did not have a habit of pushing TL forward in a manual go around situation. We need manual flight to equal manual thrust.
2. The discussion urges that the PF used the wrong procedure. A situation where the applicable procedure changes within milliseconds (weight on weels) so drastically is fundamentally wrong. Either you should train all go arounds (inkl. WoW ones) with manual TL advancement and TOGA-buttons right after (which would automatically be the case when manual flight would mean manual thrust).
Or Boeing should redesign their design so that mode selection is integrated into TL movement, so you have one single procedure designed into the machine and not this trapping inhibition of the regularly used buttons.
3. This accident would not have happened if they would not have gone around (I do not want to speculate if they went around due to RAAS alert or just due to flare beyond TDZ). That is crass, but something to look at.
Flight data analysis shows that once in a while there is a long landing. It does not become an incident because the runway offers enough margin. The SOP-disregard of touching down a bit long does not lead to anything adverse, the system is robust.
When you now introduce the SOP that upon a RAAS (or ROPS) alert, a crew absolutely must go around, you will increase the amount of low or weight on wheels (touch after GA decision) go arounds by a big factor.
Low or weight on wheels go arounds are one of the most challenging manoeuvers. You might need to disregard config warning, you need to make sure thrust is there (on Boeings), you need not to rotate too quickly (tailstrike risk), you might need to sit out a bounce with a safe pitch, and only after all that you then start the real normal go around.
That means if we introduce RAAS (or ROPS) as a mandatory to follow tool, we abolutely need to make sure crews are capable of low or weight on wheels go arounds because we will have many more of them. Otherwise we create a bigger problem than we solve.
The A380 diversion out of MAN to LHR comes to my mind due to software bug in ROPS leading to false alert. It was clearly stated in the discussion that the crew concerned would have had to face a talk with the boss if they would have exerted normal judgement by disregarding the wrong warning and just landed on the second approach. Trained monkeys are never good in a cockpit, it creates new risks.
Join Date: Sep 2014
Location: Canada
Posts: 1,257
Likes: 0
Received 0 Likes
on
0 Posts
As an industry we need to get the basics back in shape, and that means to manipulate thrust by hand when you fly by hand. Athrust only when on autopilot or in very special circumstances. ... We need manual flight to equal manual thrust.
Join Date: Oct 2015
Location: Tranquility Base
Posts: 77
Likes: 0
Received 0 Likes
on
0 Posts
Permanent use of Athrust kills speed scanning, as has been shown now in many totally unnecessary accidents.
If circumstances are giving you such a workload that by using manual thrust in manual flight you are overloaded, you should not be using manual flight either in that moment, since anyways you are most probably only following the flight director.
If you are overloaded using manual thrust while flying manually on the last 3000ft of approach, you should find another job.
I deliberately make a blanket statement, we need to reset the starting point from where we differentiate.
If circumstances are giving you such a workload that by using manual thrust in manual flight you are overloaded, you should not be using manual flight either in that moment, since anyways you are most probably only following the flight director.
If you are overloaded using manual thrust while flying manually on the last 3000ft of approach, you should find another job.
I deliberately make a blanket statement, we need to reset the starting point from where we differentiate.
Join Date: Oct 2004
Location: York
Posts: 737
Likes: 0
Received 0 Likes
on
0 Posts
Ah HOTAS
According to Wikipedia, a concept first pioneered on the Lightning. (That's the English Electric version, not the slower F35B!)
Ok, here goes:
Your flying a coupled Cat 3A approach into EGLL. It's 200m in fog. At decision height, when you don't see the required visual references, are you REALLY going to push the thrust levers forward and pull the yoke back, as your FIRST actions?
At that moment, you are 50' above a runway you can't see. You've just disengaged the autopilot, and depending on the sequencing of your actions, you could well be holding the thrust levers against the drive motors, until you push TOGA.
In the words of Sergeant Wilson: "Are you sure that's wise"?
We're flying a passenger jet here. We don't really want to be exploring the aerobatic envelope 'a la' HOTAS F18!
But we can still fly these aircraft with elegance, as long as we know what we're doing! For that, we just need quality training, followed by frequent rehearsal!
I'm afraid I disagree with 1201alarm who says this is a 'challenging manouvre'. It's just a take off with manual thrust, and an unusual flap setting! I know, because my employer insists I practice it.
According to Wikipedia, a concept first pioneered on the Lightning. (That's the English Electric version, not the slower F35B!)
there are very very few circumstances where the pilot flying the airplane instead of the automatics "is really not smart". If you can come up with a scenario where it's "really not smart" for the pilot to fly his airplane, I'd love to hear it.
Your flying a coupled Cat 3A approach into EGLL. It's 200m in fog. At decision height, when you don't see the required visual references, are you REALLY going to push the thrust levers forward and pull the yoke back, as your FIRST actions?
At that moment, you are 50' above a runway you can't see. You've just disengaged the autopilot, and depending on the sequencing of your actions, you could well be holding the thrust levers against the drive motors, until you push TOGA.
In the words of Sergeant Wilson: "Are you sure that's wise"?
We're flying a passenger jet here. We don't really want to be exploring the aerobatic envelope 'a la' HOTAS F18!
But we can still fly these aircraft with elegance, as long as we know what we're doing! For that, we just need quality training, followed by frequent rehearsal!
I'm afraid I disagree with 1201alarm who says this is a 'challenging manouvre'. It's just a take off with manual thrust, and an unusual flap setting! I know, because my employer insists I practice it.
Join Date: Jan 2008
Location: South Bucks
Posts: 46
Likes: 0
Received 0 Likes
on
0 Posts
Not sure that this can be answered but here goes: if the gear hadn't been retracted, would it have been possible to stop the aircraft in the remaining distance from the point of second touch down (impact) or would an over-run have been unavoidable?
Join Date: Sep 2014
Location: Canada
Posts: 1,257
Likes: 0
Received 0 Likes
on
0 Posts
Permanent use of Athrust kills speed scanning, as has been shown now in many totally unnecessary accidents.
Join Date: Oct 2015
Location: Tranquility Base
Posts: 77
Likes: 0
Received 0 Likes
on
0 Posts
peekay: there are several million flights a year that land safely. So we should just keep on going as we did and not aim for any improvement anymore.
Right? Not? Your argument is flawed, I thought of course about that, but nevertheless came to my conclusion.
There are many airlines that allow manual thrust, some even make it mandatory under manual flight. It does not lead to accidents. Speed control on the knot is not crucial, +10/-5 is enough precise. However avoidance of speed-not-control like in AMS, SFO or now DXB is crucial!
Right? Not? Your argument is flawed, I thought of course about that, but nevertheless came to my conclusion.
There are many airlines that allow manual thrust, some even make it mandatory under manual flight. It does not lead to accidents. Speed control on the knot is not crucial, +10/-5 is enough precise. However avoidance of speed-not-control like in AMS, SFO or now DXB is crucial!
Only half a speed-brake
There is no argument in peekay's post. C'mon, he even spells "may". It is a warning, or an educated observation if you will, that a pendulum swings both ways.
A valid and necessary approach for those who understand safety instead of just abusing the word for leverage. (that's a general comment, not related to your post, alarm, in any way)
A valid and necessary approach for those who understand safety instead of just abusing the word for leverage. (that's a general comment, not related to your post, alarm, in any way)
Last edited by FlightDetent; 20th Sep 2016 at 22:59.