Announcement
Collapse
No announcement yet.
Air Berlin pilots suspended after unnecessary go-around at Düsseldorf Airport
Collapse
X
-
I suppose it's valid given how many $/min it takes to operate an airliner...the little excursion probably cost a few $.
And maybe they were not legally high enough when they left the 'runway environment' and started 'flying over crowds of people'?Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.
-
I have mixed feelings about this... the incident was supposedly referred to the BFU for investigation and I wonder if they had anything to do with the suspension.
The key thing is: this isn't a go-around, it's a fly-by. Since airline pilots do not train for fly-by's, it's probably not something you want them to do with passengers aboard.
What I see on the video appears to be a safe, well-executed maneuver in VFR conditions (with stall protections in effect), but what concerns me, as always, are the stealth factors.
From a systems standpoint, the key difference between a fly-by and a go-around lies in the interdependencies between the activation of TO/GA and the activation of GA autopilot and navigation modes. If the thrust levers are not placed into TO/GA, certain nav cues are not activated, the FD is not providing GA guidance, the modes are not SRS and GA-TRACK, the F-PLN is not go-around, etc. If the pilots are expecting these things, that could trigger the old cascade into 'what's-it-doing-now'.
Assuming this was well-planned and briefed beforehand, which hopefully included the pax (ATC clearance was requested) and the pilots knew the unusual systems ramifications and made provisions for nav, then I don't really have a problem with this.
But, since we don't get to choose our pilots, and because we have come to learn that there are still cowboys out there with outstanding stick-and-rudder skills who don't know their aircraft well enough, and because we have come to learn that many accidents are precipitated by degraded situational awareness caused by the aircraft doing something different than their expectations, often during unusual circumstances, often in blended-automation, often at low altitudes where there is no tolerance for error, I think it's best to officially frown upon this sort of thing. It introduces the potential for error.
But it's also my impression that AB pilots are top-notch and hopefully this suspension is only a short, ceremonious thing. They still have eight days left to get reinstated...
Comment
-
Originally posted by Evan View PostThe key thing is: this isn't a go-around, it's a fly-by. Since airline pilots do not train for fly-by's...
Let's see....add power.
Manage pitch and bank and attitude and airspeed (and AOA if Gabriel has any authority over you).
Assure that the aircraft is not descending towards a touch down. Adjust flaps, raise gear
Climb to a 'safe altitude' Turn, if needed, to a new heading.
Navigate as appropriate.
Fly the plane.
For a go-around, it might be a surprise!
For a fly by, it's not a surprise.
Which one is probably safer per billion incidents of each?
I know...MY procedure may not apply to that PARTICULAR airplane.Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.
Comment
-
Originally posted by 3WE View PostMr. Black and white, please list the procedures for a go-around and the procedures for a fly by.
There is no procedure for a fly-by. Nobody expects airline pilots to do fly-by's. A go-around can be executed in managed guidance, where almost nothing bad ever happens. A fly-by requires* either selected guidance or manual flight (afiak without flight directors) where all the unthinkable errors seem to go down.That's the safety issue I see.
In the right hands on the right day, fine. When you plunk down in seat 33A, are you confident that you are in the right hands? Or should we play it safe?
* I CONCEDE THAT THERE MIGHT BE A WAY TO DO THIS IN MANAGED GUIDANCE THROUGH THE FMGC BUT I HIGHLY DOUBT IN WOULD BE DEEMED WORTH THE TROUBLE.
Comment
-
Originally posted by Evan View Poststealth factors systems standpoint interdependencies TO/GA GA autopilot and navigation modes thrust levers are not placed into TO/GA nav cues not activated FD GA guidance modes not SRS GA-TRACK F-PLN not go-around
AP ==>> Click click
AT ==>> Click click
FD ==>> Clack
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
Comment
-
Originally posted by Evan View PostThen what?
Keep speed above minimum and maximum for the config, check speed before changing the config, use pitch and power in a coordinated way to achieve the desired speed and climb performance, use bank to achieve the desired heading.
What is the procedure to hand=fly a circle to land approach? Or a traffic pattern?
Well, the same.
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
Comment
-
Originally posted by Gabriel View PostThen fly the airplane.
Keep speed above minimum and maximum for the config, check speed before changing the config, use pitch and power in a coordinated way to achieve the desired speed and climb performance, use bank to achieve the desired heading.
What is the procedure to hand=fly a circle to land approach? Or a traffic pattern?
Well, the same.
I'm referring to the known fact that risk of pilot error is amplified by transitional phases of flight and non-standard, unpracticed and improvised maneuvers.
It should be easy enough (the AB crew did a fine job as far as I can tell), but what if one of those stealthy* human factors leads to error? (*stealthy means you don't foresee it happening to you)
For example, what if you decide to do a nice, slow fly-by using auto-throttle and then you turn off the runway heading at around Vref + 5, bank significantly, pitch up slightly to achieve a shallow climb AND (being go-around -minded) retract flaps to 15, and thus, between the bank and the configuration change, you significantly raise the stall speed and the AoA before a modest increase in power has become effective thrust?
If this was a go-around and procedure was followed, you would have first advanced the levers to TO/GA and remained wings level until you had positive climb, and then you would now have correct guidance to navigate the pre-determined F-PLN go-around.
As I said, I have mixed feeling about this. The AB incident seems to have been done in a safe, expert manner so no harm done. But what if it hadn't been, or what if the next crew that tries this isn't so expert or prepared?
I think that is what the company or the BFU is saying here: leave the showboating to the non-revenue flights. Better safe than sorry.
And I seem to remember anothor Airbus that tried this and ended up in the trees...
Comment
-
Evan, doing a flyby is easy. All the mistakes that you mention can happen in any phase of manual flight. The transition from descent to climb doesn't pose any especial difficulty in manual flight, not any more than transitioning from descent to level flight, for example. Can it be done wrongly? Yes, ans the Air France accident you mention demonstrates. Take offs, approaches, landings and even cruise flight can be done wrongly and end up in fatal accidents.
The Air France pilot, unsurprisingly now in the hindsight, was so bad pilot that even today he believes that the all-mighty computer wanted to land against his desires to climb.
A pilot should not go below Vref, if they go things can go wrong, in a flyby (as it happened in Air France) or in an approach (as it happened in Asiana and Turkish) or in cruise (as it happens to Pinnacle).
But a flyby is much easier, for example, than an approach, because you have more room for power, speed, vertical speed, etc... In an approach you don't want to go below Vref (in fact Vapp) but also you don't want to go faster than that. In a flyby, anything above Vref and below the speed limits for the flaps setting is ok. In an approach you want a specific vertical speed, that which keeps you in a 3deg approach slope that ends exactly in the TDZ of the RWY. In a flyby, any non-negative vertical sped si good enough. And so on.
Flybys are not common at all, but they are still more common than you might think. They are performed so the control tower can check potential damage or gear status, in uncontrolled airports they are sometimes performed to check that the RWY is clear, it is performed in ceremonial ways when an airline retires a type from the fleet, in airshows where an airline displays their new type, etc...
An how much did that scare that passengers? (you can jump to 4:45)
Another example
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---
Comment
-
Originally posted by Gabriel View PostBut a flyby is much easier, for example, than an approach, because you have more room for power, speed, vertical speed, etc... In an approach you don't want to go below Vref (in fact Vapp) but also you don't want to go faster than that. In a flyby, anything above Vref and below the speed limits for the flaps setting is ok. In an approach you want a specific vertical speed, that which keeps you in a 3deg approach slope that ends exactly in the TDZ of the RWY. In a flyby, any non-negative vertical sped si good enough. And so on.
But let me ask you instead... why do YOU think the AB pilot was suspended over this? What do you think their reasoning was?
Comment
Comment