A flight safety lesson all airline pilots should read [Archive]

mary meagher

28th Sep 2014, 09:14

I have read Will Langewiesche's Vanity Fair Article "THE HUMAN FACTOR" this morning. And skimmed over the most recent 6 pages of PP comment.

A few points have occurred to me. First, I am horrified that the Captain brought his lady friend to Brazil, and in consequence, SLEPT FOR ONLY ONE HOUR the night before the accident flight. Says it all, really. Abdicated his responsibilities. Left the care of his passengers to the automation and the two nervous co-pilots.

When the pitot tubes that had not yet been replaced froze up under a bit of frost, and the automation not enjoying ASI data stepped down from doing the flying, the two young ones remaining at the controls also froze in every sense.

After two years of searching the CVR and data recorder were recovered. The unfortunate arrangement of side stick controls, that made it likely the PNF could not know what the PF was up to, let alone those buttons that are supposed to override control, did not help.

The efforts of the PF to overcome the PIO was according to the VF article solved by the captain's tardy arrival on the flight deck, when he told the PF to use the rudder to sort out the roll. Too late, alas. And of course using the rudder too strenuously can cause other problems, as happened to the Boeing departing Kennedy shortly after 9/11.

In my experience as a gliding instructor, I have found that men usually think they are better than they are! the women have less overconfidence. If the standard and experience of the average airline pilot continues to decay, with very little real flying, real takeoffs and landings, the job becomes more and more boring, monitoring the computers. Which could just as well be done from the ground, like drones. We are happy to ride on the little train that takes us to the outer terminals at Orlando with no train driver.

How long will it be before we are happy to ride on the LHR Dulles UA with no pilots to fall asleep at the controls?

Winnerhofer

28th Sep 2014, 09:23

The Gendarmerie was sent to investigate the activities of the crew over their 72hr stop in Rio.
The report was censored other than they enjoyed themselves and sightseed Rio with an Helicopter.
Questions remain:
Why was Dubois standing?
Was someone else seated on the observer seat?
Was the observer seat recovered? Seatbelts?
4th voice not of F/A?
So long as the DCVR is hashed and truncated, we're running around like headless chickens.

SeenItAll

28th Sep 2014, 09:27

And of course using the rudder too strenuously can cause other problems, as happened to the Boeing departing Kennedy shortly after 9/11.

That was a pre-FBW Airbus A300-600 whose vertical stabilizer failed from excessive rudder input, not a Boeing.

pg wing tips

28th Sep 2014, 09:54

Interesting article. After reading the first few pages I was worried that it would descend into a journalistic dramatisation of what happened, using emotive language and assumptions, but it goes on to make some interesting observations and points.

I am not sure that I agree on all of the things and indeed some of the relevance of what is written, nevertheless it is well worth a read.

For those interested, I would suggest that Bill Palmers book, Understanding Air France 447, is a much more informative and technically rigorous account of the flight, including underlying causes and the industry as a whole.

Mear kat

28th Sep 2014, 10:12

With the latest software that is now available in stand alone GPS devices.
It might be an idea if a separate GPS system with its own power supply was provided to give the crews a better spatial monitoring system.

It would allow the crew to view a 3D perspective of where the aircraft is in relation to the ground below and whats ahead.

Imagine a flight sim screen but showing the aircraft in profile against the surface of the earth which is monitoring the position and altitude in real time.
Just a thought if any makers are reading,

Petercwelch

28th Sep 2014, 15:42

Despite a couple of minor errors, a superb layman's explanation of 447 and current automation problem. Suspect we all will learn something from it!

what next

28th Sep 2014, 16:23

Layman? The author worked as a commercial pilot for twenty years.

Petercwelch

28th Sep 2014, 19:26

Written for non pilots but some info. I haven't read before. Issues clearly stated. Shouldn't some regular hand flying be required reqired outside the sims? Criuse, at altitude as well as full approaches?

flash8

28th Sep 2014, 19:38

Many of the newer generation would be hard pressed to hand fly today. In the old days pre glass clockwork on the -200 I had my hands full (and my brain) but nowadays glass and automation have sadly declined skills considerably.

As for 447, case in point, everything from basic flying skills to CRM fell apart before our very eyes. More worrying than the loss of the hull and lives should be the insight into training gaps that were succinctly distilled into a few minutes.

edmundronald

28th Sep 2014, 21:06

The article higlights the contradictory sidestick command inputs, and the lack of situational awareness of the captain on re-entering the cokpit.

Is it true that there is no display on Airbus that shows both sidestick inputs and the resulting composite input at one glance? Why not?

On double-yoke planes, I guess the input states are immediately visible.

peekay4

28th Sep 2014, 21:30

M.Mouse

28th Sep 2014, 22:28

In the old days pre glass clockwork on the -200 I had my hands full (and my brain) but nowadays glass and automation have sadly declined skills considerably.

So your hands on flying skills were top notch but your capacity to think about much else was compromised then?

I would argue that automation and modern flight deck aids mean that one has to make the effort to manually fly at appropriate times to maintain manual skills.

In the case of Air France 447 I still am absolutely staggered that a trained and experienced pilot could maintain full aft stick and think that it was an appropriate way to fly the aircraft!

I have never flown a sidestick and have nothing against them in principle but, as was proven in the AF case, it means that pilot monitoring has absolutely no idea what pilot flying might be doing. And what competent pilot would expect his/her colleague to be holding full aft stick?

edmundronald

28th Sep 2014, 23:24

There is a "side stick indication display" on the PFD, however it is only active while on the ground, and disappears shortly after rotation during take-off.

It is disabled after take-off because this indicator was only designed for control check purposes (e.g., during pre-flight) and could be misleading during flight.

There is a "side stick priority" button either pilot can press to take over the flight control from the other pilot (and/or from the autopilot).

How was the captain supposed to understand the sidestick states and plane behavior, if he just entered the cockpit and the control state is not displayed?

How could he even have been *sure* which pilot had effective control of the stick if either pilot can take it over?

Yes, he could have observed and figured it out; but the whole issue of stress and fatigue is that it degrades cognitive performance up to a point where even trained individuals cannot reason well - which is why instruments and controls should be designed for clarity.

I see a design error here. Updating the controls by adding a composite sidestick display to one of the panels is a simple technical solution - but probably an instrument update would be tantamount to admitting liability and therefore legally impossible as an initiative for Airbus to take if not mandated.

Naali

28th Sep 2014, 23:30

No, Petercwelch and also some else too. You have chosen a Professional Pilots,screen and it was not meant to be Evening News. We are really not supposed or expected to, -only find the most uncomfortable seat to sit on, and belt up for a free flight anywhere in the world. We really want to do that as professionals,as good as we all can. so You might read these threads and opinions as a gift for You. We allow You to read these,and expect that You have the same respect that we have for all who fly,or are with us. So i would wish that all who want to express opinions outside ,would find their ways somewhere else and leave these for us,to talk about.

onetrack

28th Sep 2014, 23:39

I think there's one glaring omission in the crew discussions on the CVR record as AF447 descended out of control.

At no time did any one person state, "I have control". I would have thought that would have been a critical statement, showing leadership and setting the command structure for the emergency, and stopping the confusion over who was actually in control.

The problem appears to be that no-one on that flight deck exercised any strong leadership and took full control, thereby relegating the others to supporting, advisory positions.

That one person should have been the Captain, of course. Despite just being woken and despite still trying to regain full faculties, he should have shown immediate strong leadership and actions.
He failed, and the other two crew members also failed in their duties, in like manner.

Dream Land

29th Sep 2014, 01:45

Side stick indications had nothing to do with this accident.

peekay4

29th Sep 2014, 07:13

At no time did any one person state, "I have control". I would have thought that would have been a critical statement, showing leadership and setting the command structure for the emergency, and stopping the confusion over who was actually in control.

Actually, the pilot in the left seat (Robert) did just that, at 02:11:37, when he exclaimed "Commandes à gauche!"

Unfortunately, the pilot in the right seat, probably in panic, kept his side stick at the full back position for two more minutes.

Centaurus

29th Sep 2014, 08:21

Before departure the captain had allotted the "leg" to the second officer (?) who was in the RH seat. Seemingly by default, the captain on leaving the cockpit for crew rest, handed over control to the junior pilot simply because it was his "leg." The real first officer who now replaced the captain in the LH seat, was clearly subordinate to the second officer purely because it was the second officer's "leg"

Surely, common sense should have dictated that before the captain left the other two up front on their own - especially since the weather radar showed storms brewing ahead- he should have nominated the first officer as temporarily in charge as PF since he occupied the captain's seat. As it turned out, the real first officer who occupied the captain's seat as PM sat on his arse and did nothing for fear of taking over control on the other guy's "leg" All he apparently did was to watch the second officer completely lose the plot even though it was obvious that the second officer was quickly out of his depth.

The seemingly sacrosanct policy of a captain giving a co-pilot a "leg" then being reluctant of taking back the co-pilot's "leg" for fear of causing offence, is idiotic and may have been a contributing factor in the utter confusion that followed.

darkroomsource

29th Sep 2014, 11:02

Quote:
At no time did any one person state, "I have control". I would have thought that would have been a critical statement, showing leadership and setting the command structure for the emergency, and stopping the confusion over who was actually in control.
Actually, the pilot in the left seat (Robert) did just that, at 02:11:37, when he exclaimed "Commandes à gauche!"

Unfortunately, the pilot in the right seat, probably in panic, kept his side stick at the full back position for two more minutes.

Don't we all learn it this way?
My Airplane.
Your Airplane.

So, if one says "my airplane" or "controls left", and does not hear the response "your airplane" or "controls left", then shouldn't one repeat the process, or attempt to get the other to recognise they are panicking?

(Please forgive me for posting, I am not an airline pilot, just a CPL, interested in this conversation)

BOAC

29th Sep 2014, 12:39

handed over control to the junior pilot simply because it was his "leg." The real first officer who now replaced the captain in the LH seat, was clearly subordinate to the second officer purely because it was the second officer's "leg" - if that was the situation then Air France is badly out of sync with the real world. Having 'the leg' does not transfer command, so LHS was still 'Captain' and NOT 'subordinate'.

flash8

30th Sep 2014, 03:06

We had two first officers out of sync with each other and anything that could have effectively woven a thread of commonality between them that they could have used to salvage the situation would have been priceless.

There is no logical reason I can see (not an Airbus pilot mind you) that tactile feedback between the joystick controls shouldn't be implemented and what seems significant evidence that it could only help.

Can anybody tell me if I am reasoning fairly here?

Judd

30th Sep 2014, 07:40

Having 'the leg' does not transfer command, so LHS was still 'Captain' and NOT 'subordinate'.
The first highlighted part is true but the second statement is open to interpretation that day. The captain was on rest down the back, but there was no indication who was now in charge while he slept during his rest period. Of course there may have been a company policy of who becomes in charge when two pilots other than the captain are up front. But it seems the second officer in the RH seat was really running the show as pseudo in command (his leg given to him by the real captain) as PF when everything hit the fan. Quite confusing situation especially as the F/O in the LH seat was so hesitant all the way along to take over. But why was he so uncertain?

Piltdown Man

30th Sep 2014, 09:06

The modern aircraft is now so clever, able and reliable that pilots have come to trust it. This trust runs from the manufacturer, through the airlines, their training departments right up to the line pilots. The trust, allied to control laws which prevent exceedances, means that very few crews will ever had to deal with an aircraft throwing a very basic "hissy fit". Simulator sessions will be dealing with complex failures involving hydraulic systems and their effect on LDA's, electrical system malfunctions, bleed problems and so forth.

Then, when we add prescriptive SOPs regarding autopilot usage we start ending up with pilots who forget how to fly. Then, when challenged, they revert to what they know and what has always worked for the last so many years. I regularly fly by hand, but I've witnessed first hand my own skills being degraded over time. I'm not as good as I used to be when I flew a simpler aircraft. Our SOPs (which are the manufacturer's) make the mouth music for hand flying so complicated that many choose to avoid it. I just get them wrong, but at least I'm having a bit of hands on time. But from what I can gather, I'm an exception. Our long haul colleagues rarely hand fly. They cite fatigue, SOPs, complicated airspace, etc. as reasons for not doing so.

I am also fortunate that we spend time in the sim hand flying. But I feel better with more time and more unpleasant failure scenarios. But that costs time and money, neither of which will be spent willingly until we are persuaded to do so. Maybe pilots (ladies as well as men) who fly highly automated aircraft should be forced to fly piston and turbo-prop twins part time? More basic flying is one if the answers, but I'm sure other things could be improved.

PM

BOAC

30th Sep 2014, 10:53

Cannot see that, Judd. Normally a F/O needs to be qualified as cruise Captain and I wonder if RHS was?

Add in, SENIOR F/O in LHS - de facto Captain?.

It would be nice to know who was actually designated cruise Captain on the roster.

Centaurus

30th Sep 2014, 11:24

It would be nice to know who was actually designated cruise Captain on the roster

Agree. That seems to have been kept quiet. One thing seems true, though. It surely wasn't the second officer in the RH seat.

peekay4

30th Sep 2014, 17:22

Hi,

Discussion about which of the two co-pilots had command was addressed in the AF447 Final Report issued by BEA.

Here are verbatim quotations:

At the time of the accident:

In the Air France operations manual, the Captain’s replacement was a co-pilot designated as relief pilot. Acting in this capacity, he made the necessary operational decisions for the flight in accordance with the Captain’s instructions. He stayed in the right seat and from this seat carried out the PF function. He performed tasks marked “C” in the check-lists and emergency procedures.

Changes made following the accident:

Reinforcement of the role of co-pilots

- Modification of rules for relieving the Captain in March 2010: the relief co-pilot is designated by the Captain, he sits on the left side and is PNF.

- Deployment underway of a new decision-making method: the co-pilot speaks first, before the final decision of the Captain (optimisation of decision-making, reinforcing the co-pilot’s responsibilities).

The Final Report is available below:

http://www.bea.aero/docspa/2009/f-cp090601.en/pdf/f-cp090601.en.pdf

BOAC

1st Oct 2014, 11:02

Thanks, pk - it looks as if AF had a strange concept of command! As for that last paragraph.........:eek: I hope translation has caused that.

Max Angle

2nd Oct 2014, 18:17

Surely, common sense should have dictated that before the captain left the other two up front on their own - especially since the weather radar showed storms brewing ahead-Perhaps common sense should have dictated that with "his" aircraft heading towards the ITZC and an area of highly active thunderstorms associated with it that he shouldn't have left at all but stayed on the flight deck.

BOAC

2nd Oct 2014, 19:57

Don't forget his 'girlfriend' may well have been awaiting him.

peekay4

4th Oct 2014, 21:53

I am still puzzled over that last paragraph I quoted in the Air France report... seems contrary to CRM best practices to me??

pithblot

12th Oct 2014, 23:57

Maybe a single Centre Stick, useable from both pilot seats, would be a better way in future?

cf the Australian Victa Airtourer (http://www.theairplanepictures.com/victa-airplanes/interior-design-cockpit-victa-airplane/)

BOAC

14th Oct 2014, 17:44

..........until that one fails?

FlyingStone

14th Oct 2014, 22:36

Maybe a single Centre Stick, useable from both pilot seats, would be a better way in future?

So instead of proven and redundant design of many years (dual inter-connected control columns) you want to "improve" it by putting only a single stick in the middle of cockpit?

I guess people have different definitions of term progress...

pithblot

15th Oct 2014, 00:25

I was advocating a concept and assuming the self-evident did not need explaining. The concept is putting the primary control in a position that is visible and accessible from both seats: the self-evident is that control redundancy is provided.

(Baron von Munchausen, if you are listening, I'm not seriously suggesting we replace two high bypass turbo fan engines with a single Continental 0-200. Also, just to be clear, there is no insinuation that a lawn mower starter handle should be used anywhere in an Airbus. At least not on the flight deck.)