Lonewolf_50

Thank you Chris.

OK:
That leaves me more confused than before.

I appreciate the mind set of "let's not be hasty."

flyboyike

It's hard to argue with a word like "whilst". If somebody says "while", it's no big deal, but once they throw "whilst" into the discussion, the opponents pretty well clam up. Another such word is "fraught". If somebody says something is dangerous, it's not that big a deal, but once it's "fraught with danger", the party is over.

Kudos to Wally for excellent rhetoric skills.

Chris Scott

Quote from Lonewolf_50:
That leaves me more confused than before.

Thanks? Not sure if you were referring to the piece from OK465, or mine!

Coincidentally, I was just labouring heavily to improve (edit) my post while you and flybyike were posting yours. So let me know what you think - if you've the stamina to re-read it...

OK465

You're not the Lone Ranger, Lone Wolf. That AAIB narrative has a number of confusing statements and has me scratching my head.


Here's the first two:

.

.

Confusing to me because I would think a 5000+ hour on type Captain ought to know that 'alpha lock' is not displayed on the PFD and is slat/flap related.

And a 5000+ hour on type Captain ought to know the autopilot cannot be re-engaged below VLS.

Next taken together confusing statements 3, 4 & 5:

.

.

A TCAS DESCEND RA evidently WAS indeed provoked within 1 second of A/P disengagement, but for 18 seconds the 340 remained within 200 feet of FL360 (the 330 encountered a 200 foot altimeter drop), so what's the point to be made with the 'climb vigour' and 'could well have climbed thru FL363' statement. RA occurred without either of these. Time to do some pilot stuff with the SS.

And here's the kicker (board statement):

Read A33Zab's previous post on these functionalities (alpha lock 'pulses' bright/dim on the EWD slat/flap display, alpha floor is a PFD FMA steady indication) and you can see why I'm confused. Of course I don't have 5000+ hours on type.

And as Chris Scott alludes to, the DFDR traces are almost worthless....it would be interesting to see a TLA trace, 'cause the whole thing is screwy IMO.

DozyWannabe

I'd say they identified a *possible* reason, and acknowledged this at the time. To say it's a "likely" reason is objectively inaccurate - as we have no corroborating evidence to support the assertion, and subjectively unfair - as it assumes a degree of automation dependency on the part of the crew on the basis of anecdotal reporting (i.e. "Many of us have seen this kind of behaviour").

I'm sure the anecdotal reporting is correct, but at the same time I think it's wrong-headed to try and squeeze this particular incident into a "...Magenta"-shaped mould without any real evidence to back it up. Finding repeatable patterns in a series of mishaps can be something of a double-edged sword, because while spotting and working to eliminate that behaviour will help, it won't help in the case of those incidents where, while outwardly similar in nature, that pattern wasn't involved - and can mask other problems that can be equally as dangerous.

tdracer

FAA unveils pilot training to avoid stalling airliners


FAA unveils pilot training to avoid stalling airliners

I would have added the Birgenair 757 to that list....

Chris Scott

Quote from me:
"RetiredF4 and DonH have identified the likely reason for the lack of sidestick action by either pilot. To be a bit more more specific, one or both pilots were probably trying to re-engage one of the two APs, possibly trying them both alternately. Many of us have seen this kind of behaviour: priority given to re-engaging the automatics, rather than flying the aeroplane."

Quote from DozyWannabe:
"I'd say they identified a *possible* reason, and acknowledged this at the time. To say it's a "likely" reason is objectively inaccurate - as we have no corroborating evidence to support the assertion, and subjectively unfair - as it assumes a degree of automation dependency on the part of the crew on the basis of anecdotal reporting (i.e. "Many of us have seen this kind of behaviour")."

From the AAIB Bulletin:
"The A340 commander's report
"...Shortly before the AIRPROX event he experienced moderate turbulence and noticed outside air temperature changes. Suddenly the aircraft began to climb, the Master warning soundedand the autopilot self-disengaged as the aircraft exceeded the speed limit of 0.86 Mach. The indicated airspeed dropped below VLS (the lowest selectable) as the aircraft climbed and the commander took manual control of the aircraft because neither autopilot would engage. The crew subsequently reported the incident to Shanwick on HF radio and, using their TCAS, they descended back to FL 360 in a safe area...
"The commander could not remember the sequence of warnings but he did recall being unable to re-engage either autopilot which prompted him to make manual inputs. He also remembered seeing an 'Alpha Lock' warning displayed on his Primary Flying Display."

My use of the word "likely", rather than "possible", was to specify my opinion. I cannot think of any other legitimate activity to explain the delay that we are discussing. It was not necessarily to imply what Dozy describes as "dependency"; rather that the crew prioritised re-engaging the AP during a period of time when the pitch attitude and flight profile were becoming abnormal, and clearly compromising the safety of the flight.

Clandestino

I would have definitively removed USAir 427 from the list - putting it up there with Colgan, Pinnacle and AF447 is so terribly wrong. Given dr. Wiener's track record, this just has to be something misheard.

Lonewolf_50

Chris, I put two replies into one post.

My thanks to you was for you clearing a bit of the fog out of the air.

My response to OK465, which I simply addressed to OK, was to show that even with a bit clearer air, my brain is still fogged with not getting it. His further explanation makes me feel a bit better about my own lack of seeing how it went down, given the testimony of one of the crew. Glad it all worked out, all said and done.

Sorry if I added to confusion by using OK, which may have come across as other than a reference to his PPRuNe nomme de post.

tdracer

Agreed - although it's the second recent press release on the subject that referenced the USAir 427. After the first I actually Googled it to make sure it was the same crash that I was thinking about.

DonH

Hi Dozy, Re, "I'm sure the anecdotal reporting is correct, but at the same time I think it's wrong-headed to try and squeeze this particular incident into a "...Magenta"-shaped mould without any real evidence to back it up. Finding repeatable patterns in a series of mishaps can be something of a double-edged sword, because while spotting and working to eliminate that behaviour will help, it won't help in the case of those incidents where, while outwardly similar in nature, that pattern wasn't involved - and can mask other problems that can be equally as dangerous. "

I too, would use Chris Scott's term, "likely" to describe what happened. This isn't jamming behaviour into the magenta-mold, it is recognizing factors that I, and probably Chris, (don't want to speak for you!), really do know about from having done too many simulator sessions! ;-) where that very phenomenon occurs.

Each sim session's script usually has a few very busy moments where "all hell is breaking loose" with a serious system failure, sometimes combined with a go-around from a low-vis approach.

The single chime and Master Caution are going off constantly, (and one is constantly resetting it), as the ECAM catches up with the FWC messages being displayed and re-prioritized, there may be warnings that the autoflight system has disconnected, perhaps voice-audio warnings, (such as the TCAS warnings) are occurring depending upon the scripted failure(s) - lots of times this usually occurs at/just-after takeoff or, far busier, the go-around, when one is making sure the gear is up, (if no one calls "positive rate", almost invariable the gear is left down while concentrating on other things!), the climb-thrust is set and so on.

The procedures are set out and well-trained as you will know from reading previous threads, but one is working hard, maintaining concentration, anticipating the next step while remembering/executing the drills in sequence. It is a very, very busy airplane when something major goes wrong or one is in cruise and all of a sudden the cautions and warnings come.

As I've mentioned previously, there are perhaps two, possibly three times when decisive, timely actions are required of the pilot-flying or the captain; they are the rejected takeoff, (captain only - and I do know this is controversial in some circles), stall, TCAS, EGPWS warnings and the go-around. All the other times, including engine failure-fire-damage one can & must take one's time, and by that, (it's been mentioned before), I mean 4, 5 or more seconds, to do nothing but collect one's thoughts and communicate with the PNF calling the failure and calling for the ECAM actions, etc. while taking the radios.

To your point, I would be quite certain that the A330/A340 event wasn't a "magenta-line" matter, nor is it startle; I would characterize it as re-grouping & otherwise marshaling one's resources which heretofore had been "at rest" so to speak, and while only partly due to the maintenance of cockpit discipline before launching, is an adjustment of focus while dealing with the inevitable surprise, (none of which is startle - startle is the result of a "comfortable unexpectedness" and perhaps a knowledge-complacency where "relaxed-but-ready" is that which is borne of thorough training, an abiding passion for the books and discussions and a desire to "read" in one's chosen profession which means reading accident reports, studies on current topics as well as staying in the books and knowing one's airplane....going beyond the minimum required to build depth and resourcefullness which can be called upon even once or perhaps twice in one's career and not waiting for, or complaining that one's airline isn't teaching one!

Automation is fabulous and I loved it but if that's all that one knows, one is at risk of being quickly overwhelmed, (startled) by the airplane.

gums

I like this newbie "Don". Good thots.

@ Doze

Well, as most here know, I am in the latter grope of pilots. Got my engineering stuff from USAFA and was always a curious type. I even looked into the code for the A-7D computers and then the Viper's. Got the charts and block diagrams for the Viper FLCS and had to instruct the first few folks back in 1979 and 1980. Ya gotta know your plane, and it's a poor craftsman that blames his tools for a botched approach or incident.

I was and still am a proponent of using Otto to reduce workload and to provide some time to hit the head or eat a sandwich or... Otoo also helps after a missed approach and divert to a new field, especially when in a single seater and no "PNF".

But what bugs me is the seeming over dependence upon Otto for things that the pilot in command should be on top of. I think Asiana. I think of my buddy in the Cali crash. And I think of at least one Viper pilot friend that died because the altitude hold function had a lower AoA protection than the "normal" law. But he wasn't a child of the magenta line, he was trying to get his act together after aborting a low level route.

I am confused about the alpha protect laws and implementation. My documents show that the jet won't let you go above a certain AoA except in sub-law 2 (b) iii part three ( sarc off). So the zoom climb in the other 'bus plane is confusing if the pilot(s) did not command a climb.

Gretchenfrage

DonH throws in some good thought. I'd still like to contest some.

Well, procedures might be set out, but they do not cover every situation. Therefore to put it right at the beginning of an argument seems to imply this. We need to be careful and vigilant that the same over reliance to automation is not aped with procedures. We definitely need procedures, but as with automation and sops, if there starts to be a religious thirst for all these things to replace basic skills and common sense, then we end up full backstick at FL350.

To pretend that procedures are well trained is somewhat preposterous. We all know how the bean counters cut back on anything they can in training, we all know how the operations bosses limit anything outside hand-cuffing sops, to invalidate such arguments. By simply reading the training syllabus of most airlines, you will discover that it rarely goes one little step beyond the bare minimum set up by regulators under pressure from the managers lobby.

Well put. However todays sops from industry and airlines and especially todays freshmen TRI/Es go the other way. The pandemic of "memory items" and additional "procedures" is creeping up even more after every incident. Just look at some airlines who want UAS to be a memory item, a procedure (see above) that should be well trained (see above) and readily dispatchable by a brain under sudden fire. In sim sessions i regularly, and by that i mean regularly, see UAS, rapid descents and severe damage assessments and memory items badly screwed up. Why? Because there is way too much emphasis on memory items and wording. When a brain is utterly stressed, it does not perform well. To put it under even more stress with the above is purely and simply counterproductive. I have seen newbees completely blocked because they could not recall the correct name for a checklist or stuttered for seconds to produce the correct pronunciation or term for an item. The TREs were taking notes like wild and elaborated for hours in the debrief of how important it was to know that kind of stuff. Maybe the poor AF-FO was torturing his mind to find the right wording, memory item or "absolutely correct procedure" and by doing that he was blocked-cramped pulling.
Checklist have been introduced to counter such phenomenons and to counter blockage, to give a straw to cling to. I once reverted to the checklist for a emergency descent, becasue i was really surprised in that sim session, just to be marked down for that.

On the other hand there are times where the only thing that is left is your rear side. For a few seconds, after that you have to go back to the marvels of modern aviation, because your bird has become mainly digital.

I contest the statement that todays pilots are still able to do that.
They are NOT well trained enough and NOT free enough from the limiting constraints of over reliance on automation, procedures and sops.

The moment we admit to that we can start working with these wonderful birds A and AB give us.

HazelNuts39

A similar incident: Air France A340-313, over the North-Atlantic on 22 july 2011

Chris Scott

Thanks for that link to the BEA Report into this "Serious Incident", HN39.

Judging from the Synopsis, this incident had elements in common with both AF447 and the A340 AIRPROX. It promises to make for shocking reading.

Better to take the dogs out first, methinks...

BOAC

Don't let the dogs see it, Chris....................

A quick scan through that report leaves me very puzzled as to who was PF and who was PNF and exactly what were they playing at?

Sadly again absolutely no credit to AF, selection, training and abilities in that incident. I cannot relate to the timeline portrayed in this report.

3 minutes from 'Overspeed' before ANY action is taken - then by PNF?
After 4 minutes the 'Overspeed' ceased and they were practically back at cruise Mach BUT at 6 minutes 'someone' decides to select a slower speed? AND
at 7 minutes extend the speedbrakes when almost at cruise Mach?
At 8 minutes after the 'event' they are still almost 'level' (+200')

Speedbrakes come in after 13 minutes, so I guess they had power against speedbrake for 6 minutes?

Either the world has gone mad or the report has been corrupted in translation. Look what they 'missed'

Nose-up attitude, which had reached 12° pitch up,
ˆ High vertical speed, which had reached 5,700 ft/min,
ˆˆThe flashing frame around the altitude display on the PFD when the altitude
passed 35,200 ft,
ˆˆThe position of the FD cross bars,
ˆˆAltitude until it reached more than 38,000 ft,
ˆˆThe “AP1” display which had disappeared from the FMA strip,
ˆˆThe amber ECAM “AP OFF” message which remained displayed for 9 seconds after the AP disengagement.

The report also says 'Because of the absence of CVR recording, it was not possible to evaluate CRM in order to explain the lack of monitoring of the basic parameters and the flight path.' - Do we really need that?

All this AFTER 447.

rudderrudderrat

Hi BOAC,
I think you are misreading minutes instead of seconds.

"At 1 h 00 min 27 (point n°2), ......., the PNF manually disengaged the AP by pressing the
takeover pushbutton on the side-stick. A pitch-up input on the PNF’s side-stick going as far as ¾ to stop was recorded for 6 seconds.

At 1 h 02 min 06 (point n°9),...... the aeroplane descended past 36,520 ft. The PF then became aware of the disengagement of the AP and made a pitch-down input on his sidestick."

The total amount of time between AP disconnect and PF (Captain) actually taking control was "only" 1 min 39 secs.

Is it the ergonomic design of the Airbus flight deck or AF crews?

On previous AC types, we had large red letters on the PFD telling whether the AP was engaged "CMD" or only the Flight Directors "FD" in addition to all the other alarms, bells and whistles.
Airbus simply removes "AP(1)(2)" from the top right hand corner of the PFD - it is easier to miss.

Chris Scott

Hello BOAC,

Re. the BEA Report on the AF A340 Serious Incident of 2011, should your time values be in seconds, rather than minutes?

Aso, you write:
Either the world has gone mad or the report has been corrupted in translation. Look what they 'missed'

You then go on to quote what the BEA has written on Page 14. Are you merely remarking that the crew may have gone mad, or perhaps you can explain what has been "corrupted in the translation"? At a superficial glance, their script seems to be in line with the DFDR traces in the Appendix.

Re PF vs PNF, I am inferring that the PF is the pilot (in this case, the captain) who was nominated as such at the beginning of the flight. it's always been a potential source of confusion (particularly in an airline like BEA/BA, which switches PFs for the "monitored approach"). Airbus itself used to add the terms CM1 and CM2 to avoid the ambiguity. In this case, the copilot seems to have remained the PNF (i.e., as far as we know there was no formal handover of control), but made sidestick inputs - whether intentionally or not.

BOAC

Ok - I did say "A quick scan through that report" and I can see it is me that has gone mad. It certainly did not make sense as minutes. Still an unholy mess, even in seconds.

CS - there are two sentences in the para you quote - the 'missed' is a separate sentence - there is a full stop there.

The query on PF/PNF was that PNF appears to have been operating the sidestick?? Not my understanding of the operation. Certainly disconnecting the A/P should have been PF - unless incapacitated.

OK465

Just a quick sidebar,

From the F-GLZU report:

Pedantic point concerning reports and accuracy....in the A346 & A332, the "AP OFF" ECAM message is red not amber when the A/P is disengaged with the takeover button. The red ones usually tend to get my attention.

Is it, or was it at one time, amber in the A343?....or is this a translation problem?....a tech writer problem?....or a report quality control problem?

(Some of these reports make a substantial amount of the info on PPRuNe look surprisingly studied. )