Ok - I did say "A quick scan through that report" and I can see it is me that has gone mad:rolleyes:. 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.

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. :})

A340 AoA (protection) Law - disengagement conditions
 

The new (to this thread) BEA Report on the AF A340 Serious Incident of 2011 describes the conditions that allow AoA Law to disengage back into Normal Law (which it refers to correctly - in the context of high altitude - as "load-factor control"). The OR-gate conditions that it lists include a third condition that we seem to have been unaware of so far in our discussion of the A340 AIRPROX event of October 2000.

I raised some controversy recently when I questioned the disengagement logic of AoA Law, having been led to believe that, in the A340 AIRPROX case, once AoA Law was triggered by a short-term, phase-advanced (gust-induced) identification of alpha-prot, it remained engaged until the PF finally made a substantial push on the sidestick. This seemed to be despite the fact that - after the trigger-gust subsided - the AoA and phase-advanced AoA appear to have been below the trigger threshold for a couple of seconds until the EFCS (FBW) in AoA Law selected up-elevator to increase the AoA to alpha-prot.

Accordingly, I wrote:
"...Phase-advanced alpha-protection puts the FBW into AoA Protection Law. Meanwhile, the AoA has returned (fallen) to what it was before the gust. However, FBW uses up-elevator to increase the AoA to alpha-prot. The a/c climbs suddenly until I push the stick forward more than half-travel, OR use less-than-half forward stick for more than a second."

Well, in this new report into the 2011 incident, the BEA lists the disengagement conditions as follows (newbees, note that a sidestick is spring-loaded to neutral):

sidestick forward more than half-travel;
OR
sidestick forward less than half-travel for 1 s while AoA < Alpha MAX;
OR
sidestick neutral or forward for 0.5 s while AoA < Alpha Prot.

It seems now that, either my understanding of the disengagement conditions for the A340 in 2000 was wrong, or they were subsequently modified?

PS by EDIT

Re the report appendix, am looking forward to interpretation by those who are more experienced and better than I am in interpreting FDR traces. (Mentioning no names!) In the interim, I notice that AoA Law disengages (to Normal Law) and re-engages on 5 occasions prior to its final disengagement. So there are 6 periods of AoA Law.

The first period ended in disengagement after only 8 seconds, apparently because the sidestick was released to neutral and an up-gust ceased (or a down-gust was encountered), reducing the AoA below Alpha-Prot (see the third condition listed above) for > 0.5 s. Although this restoration of Normal Law lasted less than 3 seconds, the report suggests it would have been maintained, but for the turbulence and speedbrakes, with the sidestick neutral. During this short period of Normal Law, the pitch increased rapidly from about +5 to about +8.5. The increase in pitch rate and AoA rate was so great that, when AoA Law re-engaged, it was unable to stop the AoA briefly exceeding Alpha-MAX.

Notwithstanding my doubts, expressed a few days ago, about the logic of engagement/disengagement of AoA Law (see above), the DFDR traces in this BEA Report seem to show that the periods of its disengagement may have resulted in greater divergences from normal flight parameters than when it was engaged.

One assumes that, if AoA Law was maintained, the flight profile would approximate a phugoid? That view may be supported by the last (and longest) period of AoA-Law engagement, which started just after the FL382 apogee. During the descent, the VS increases gradually to about 4000 ft/min; then decreases gradually to zero (at FL365). It is at that point that the PF applies just under half forward-stick for more than a second, disengaging AoA law.

BOAC,

I thought that was what you meant - just checking!

As I said earlier, the "PNF" made (substantial and inappropriate) sidestick inputs, but IMO the BEA is right to continue to refer to him as PNF. We don't even know if his inputs were intentional.

Quote from OK465:
"....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?"

Must admit that red is my recollection from the A320, too, but any error is not in the translation.

Here's the French original (my emphasis):
"L’équipage n’a pas vu les différents changements d’affichage (message « AP OFF » ambre de l’ECAM, disparition de l’indication « AP1 » du bandeau FMA)."

A pity the CVR isn't available, there may have been a brief verbal exchange which would explain that. Seems out of place.

Chris, your estimation of the "chasing" of air data sensors by the A/P in such entries into turbulent air is confusing to me.

I learned a long time ago that if you were entering turbulent air that has a lot of vertical component, that you need to work on holding pitch and power for the proper "turbulent air penetration" airspeed. You could expect both A/S and altitude to vary a bit as the air mass bounces you about.
"Don't chase it!" was some of the teaching then.

Is that still current teaching?

Hello, Chris Scott, OK465, BOAC;

Since AP1 was engaged at the time of the event, and re-engaged subsequent to the recovery I think the PF was the captain. While not strictly SOP, normally a left seat PF engages AP1 and a right seat PF engages AP2.

However, the captain was eating dinner, with the retractable tray out and it is common for the PNF to temporarily take over PF duties as well as communications while the other crew member is having a meal. The AP is left in AP1, again informally.

Where something like this 'interruption in smooth flight' occurs, the "PNF" would continue the PF's duties and mind the store until the PF can put the meal and tray away and resume PF duties to deal with the event.

The Report mentions and describes the incorrect priorities and evident loss of CRM and the "command-control" process as both crew members were fussing with getting the dinner tray onto the observer's seat and getting out the PA handset for an announcement. Even as we don't have the CVR, this is clearly another breakdown in cockpit discipline, placing "Communicate" ahead of "Aviate, & Navigate". Without the CVR we dont' know if control was actually, verbally handed back to the PF, ("You have control....I have control", etc).

Further, it was the PNF's SS instinctive-disconnect button that was pressed and disconnected the AP but he clicked twice on the button which cancels the aural and ECAM warnings. From long observation, pressing the instinctive button to cancel the AP warning is not atypical behaviour; - this would have been out of habit, but here the cancellation robbed the PF of information that the AP had been disconnected and the altitude alert was also missed. The BEA Report states that such cancellation of the warning was an amber AP OFF, (OK465 you know, I do recall an amber "AP OFF" ECAM message somewhere in the dim past (perhaps on the A320?), but I can't find reference to it in either older A340 or recent A330 FCOMs. I can only find "red warnings" reference, (see below). Certainly, under the circumstances of this intentional, by the instinctive button disconnect, the warning was/is always red and the AP cavalry charge continues until the button is pressed again so the BEA Report is incorrect in its reference to the "amber AP OFF" ECAM message.

The PF who was distracted getting his meal put away, was unaware that the AP was disconnected and the airplane was under manual control. The PF remained unaware until he realized that the airplane had climbed well above its cleared altitude and when he went to set the AP to descend to their cleared altitude, did not see "AP" annunciated on the FMA, (p.5 of the Report).

The Report explains the fact that it was the "PNF" who disconnected the AP, but leaves us there as to who was flying once the PNF made the HF call to advise New York Aeradio of the level bust, while the airplane was essentially still out-of-control), nor does the Report explain why the "PNF" chose to disconnect the AP and then do what he did, which was to pitch the airplane up.

The use of airborne radar has been a topic of considerable interest since June 1, 2009 particularly. I have noticed that air carriers began publishing in their FCOMs supplementary notes on the use of current radars including various techniques of using antenna tilt and gain. It's about time.

Given history and the formally recognized need to address certain cultural issues within the carrier*, I would hope that this got a thorough internal review (I wonder if their flight data program picked this up?), because, "If the AP had not been manually disengaged, it would have remained engaged; there would not have been a significant trajectory deviation, with a gain in altitude of about 200 ft.

Note: Without the high angle of attack protection, the aeroplane would have kept its ascending trajectory until the triggering of the stall warning. p.9"

*Operational Safety Review, Dec, 2009 - the contents of, and I believe, links to this review have been previously reported on PPRuNe.

Chris,

Both incidents have been discussed on PPRuNe when the respective reports were published. I do remember that the disengagement conditions were modified after the AIRPROX incident, but I don't remember the details.

Hi DonH,

Think the report says the copilot only pressed the AP instinctive-disconnect PB on his sidestick once (top of Page 13)?

Am in full accord with your analysis of CRM aspects. Quite shocking mis-prioritisation of actions from start to finish of the incident. Hard to know where to start...

Yes, it also reflects my experience of the common underuse of wx radar, and misunderstanding of how to use it effectively to avoid flying into or over the upper parts of Cbs.


'Evening HN39,
Quote:
I do remember that the disengagement conditions were modified after the AIRPROX incident, but I don't remember the details.

Thanks, but it was news to me. The only recommendations in the AAIB Bulletin relate to RVSM procedures.

Quote from Lonewolf_50:
"I learned a long time ago that if you were entering turbulent air that has a lot of vertical component, that you need to work on holding pitch and power for the proper "turbulent air penetration" airspeed. You could expect both A/S and altitude to vary a bit as the air mass bounces you about.
'Don't chase it!' was some of the teaching then. Is that still current teaching?"

Afraid I'm the wrong person to ask about current teaching! But you raise a very valid point - in line with my own experience.

However, I think the only "chasing" by the EFCS/FBW visible in the DFDR traces is of the AoA when it is in AoA Law? I can only suggest that perhaps that is inevitable, bearing in mind the (presumably) narrow margin between Alpha-Prot and Alpha-MAX at cruise Mach - not to mention the stall. (That is: I'm out of my depth here, call an aerodynamicist!)

I doubt that, at normal cruise speeds and AoAs, Normal Law would be "chasing" the AoA, or even the Nz that it has been commanded to achieve by the sidestick position.

Aeroperu 601 with blocked static ports worse than AF447
 

Although the post to which you were replying incorrectly stated that the Aeroperu 603 crash was caused by the aircraft stalling on account of the wrong PITCH was incorrect, the aircraft did finally stall.

As pointed out in my book Air Crashes and Miracle Landings--60 Narratives the Aeroperu pilots flew out over the sea for safety (and presumably because there would be no mountains there did not use their radio altimeter). Not realizing that the altitude of 9,700 ft confirmed by ATC was merely that given out by their transponder in turn dependent on the barometric altimeter (not working properly due to the blocked static ports) flew into the sea.

The aircraft bounced and rose 200 ft before STALLING, inverting and falling back into the sea. Although one of engines had ingested some water on initial impact with the sea and was impaired, it might have been have been a different story had the engines not already been throttled back due to a spurious overspeed warning and had no time to spool up.

As you rightly said, blockage of the static ports is even more serious than blockage of the pitots. The AF447 pilots could see that after a considerable climb (causing the a/c to stall) they were falling downwards which is more than the Aeroperu pilots could. In the circumstances they (Aeroperu pilots) did well.

ChrisScott
In posts abot A340 incidents I want to clarify some misunderstanding about FBW control laws.
Alpha Lock function cannot be triggerred in clean configuration. If anything it was alpha floor protection and experienced pilot should be able to recall if it was engaged because he would have manually disconnected it.
In normal law from Valpha prot to alpha Max side stick commands alpha directly which is desirable because it gives the pilot direct control over AOA for recovery.
Alpha prot is not same value when below Vls and when at cruise speeds. In cruise it is about 5 degrees.
Once in alpha prot with stick neutral AOA does not increase but is decreased to Valpha prot. In this case phase advanced means it triggered before 5 degrees may be 2or 3 degrees. So FBW will try to maintain that and not increase as you suggest.
Manual flying skill is not a panacea for automated aircraft problems. Manually flying Airbus is very easy but it is more important to understand automation thoroughly. in this case the experienced pilot had he understood the auotmation he would not have wasted precious time trying to engage autopilot which does not engage when the protection is active.
It is necessary to study each line and between the lines of Autoflight system and Flight control system. Merely Flying 4 sectors manually everyday does not necessarily make you a better Airbus pilot. When the speed is below V alpha prot you should know that side stick is commanding alpha there is no need for any further display.

Thanks to Chris, Don, and vilas. Makes a bit more sense now, with all of that presented. :ok:

I'm having flashbacks to the line in "A Few Good Men" when Tom Cruise is asked, "Is that clear?"

And he says, "Crystal".

Well I'll tell ya folks, even if you pitch up from 35000 feet and 265 KCAS to 20 degrees nose high with idle thrust, the lowest Mach you will see before 'prot' finally gets the nose down is about 0.58....no alpha floor.

And 0.67 was the lowest Mach recorded in the near miss.

Someone please 'splain' to me how it's possible to get alpha floor activated above 0.53. :)

Hello vilas,

Thanks for listing many of the areas we have been discussing in relation to the A340 AIRPROX and (since yesterday) the AF A340 Incident, and summarising your understanding of them.

I particularly welcome your explanation of the Alpha Lock and Alpha Floor functions in relation to the captain's statement to the UK AAIB after the A340 AIRPROX incident. Some of us are still trying to confirm that Alpha Floor could be engaged at a speed greater than M0.53. Do you have any information?

Some of your points, however, contain significant differences from my understanding. I shall now quote them in the order of your post, and respond.

"In normal law from Valpha prot to alpha Max side stick commands alpha directly..."
I think you are referring to AoA (Protection) Law, not Normal Law? ***

"Once in alpha prot with stick neutral AOA does not increase but is decreased to Valpha prot."
I think that would be true if AoA (Protection) Law was triggered as the AoA was increasing through Alpha-Prot, but not necessarily true if the trigger criterion was phase-advanced Alpha-Prot, i.e., below Alpha-Prot.

"...phase advanced means it [is] triggered before 5 degrees may be 2or 3 degrees. So FBW will try to maintain that and not increase as you suggest."
How could it do that? I think you may be misunderstanding the principle of phase-advanced Alpha-Prot. Even if you do understand it, I think it may be helpful at this point to share my own (tentative) understanding in case it helps some of our readers. No doubt better brains than mine will correct me if necessary.
"Phase-advanced" stall-protection existed, to my knowledge, nearly 50 years ago on the VC10 and BAC 1-11. On Airbus FBW, when the AoA is increasing rapidly in Normal Law, the EFCS/FBW needs to anticipate the exceedance of Alpha-Prot and trigger/engage AoA (Protection) Law in "advance". So it is constantly extrapolating what the AoA may be a short time after the present. We do not know the precise figure of that time-period, but it is probably between 1 and 2 seconds. That means that the precise AoA at the moment of engagement is a function of the rate-of-change of AoA (i.e., delta-alpha), as well as the current value of Alpha-Prot. Delta-alpha obviously depends on a number of variables, including recent/current pilot inputs and gusts. Therefore, for the EFCS to use the engagement AoA as the target for the duration of the period of AoA Law would seem unlikely and inappropriate.
I stand to be corrected, but my money is on the EFCS targeting current Alpha-Prot - which is variable for the reasons you have implied.

"When the speed is below V alpha prot you should know that side stick is commanding alpha there is no need for any further display."
True but - sitting comfortably here on terra-firma - I would simply remind you that the speed indications on the PFD, excellent as they are, may be quite difficult to interpret when you and they are jumping up and down in severe turbulence?


*** (added by edit)
Apologies, vilas! I now see from DonH's post (below) that you are right. Although the AAIB Bulletin into the A340 AIRPROX incident of 2001 refers to it as "AoA (Protection) Law", the FCOM extracts from both 2001 and 2010 that DonH displays clearly refer to it as the "High Angle of Attack Protection" mode of Normal Law.

Chris Scott, Re, "Think the report says the copilot only pressed the AP instinctive-disconnect PB on his sidestick once (top of Page 13)?"

Yes, agree thanks - I had misread the para. on P13, believing that the button had been pressed a second time.

To clarify for OK465 regarding the "AP OFF" ECAM message, some collective in-depth research through historical FCOMs has revealed nothing in the way of an amber ECAM message - the "AP OFF" ECAM message is red, no matter what conditions, (instinctive disconnect on the SS, pushing the AP1/2 button, moving the SS when AP engaged, aircraft self-disconnect) the AP was disconnected under.

- edited to add: OK465, Re, "Someone please 'splain' to me how it's possible to get alpha floor activated above 0.53. http://images.ibsrv.net/ibsrv/res/sr...lies/smile.gif"

Alpha-floor is an A/THR mode - (in that the A/THR is provided with AoA information and invokes Alpha-floor when other conditions are satisfied). Alpha floor is inhibited above M0.53 because, (I'm guessing here!) there probably isn't sufficient thrust to respond alone to an alpha-prot event, (meaning, the pitch must also be reduced to unload the wing/reduce AoA).

Alpha-prot on the other hand is a pitch flight-law reversion from Nz (load factor) to AoA and is not inhibited at any time, (but, logically, isn't available outside of Normal Law). This may explain why the recent "A340-2011" event moved into, and out of "Alpha-prot" several times.
-end edit

HN39, Re, "I do remember that the disengagement conditions were modified after the AIRPROX incident, but I don't remember the details. "

It appears as though the disengagement conditions were changed some time after June 01 2001 and prior to July 01 2010:

http://batcave1.smugmug.com/photos/i...RB8j9ft-XL.jpg

High AoA Protection mode - disengagement conditions (cont'd)
 

'Evening DonH,

See you've finally confirmed by the 2010 FCOM the third condition that was in the BEA Report of 2011 into the AF A340 Incident - but absent from the AAIB Bulletin about the A340 AIRPROX.

Thanks for posting the relevant extracts of a 2010 FCOM and a 2001 FCOM. We can infer that the 3rd condition/criterion may have been added as a result of the A340 AIRPROX event, although the AAIB's recommendations did not include it.

As one who has recently been arguing for easier disengagement conditions than those that seem to have applied at the time of the A340 AIRPROX, I must reiterate that the AF A340 Incident of 2011 does little to support that contention, as I admitted yesterday.

Re. Alpha Floor inhibition
Do you have access to FCOM 2.22.30? Does it mention an inhibition at speed > M0.53?

Hello Chris Scott, Re, "Re. Alpha Floor inhibition
Do you have access to FCOM 2.22.30? Does it mention an inhibition at speed > M0.53? "

Yes, M0.53 is mentioned in relation to an A/THR Reversion Mode, (but not a change from Nz to AoA law - that is a different matter, I think): - from 2.22.30 in an FCOM:

I have to say that the differences between "Alpha-prot" and "Alpha-floor" have eluded me all these years, perhaps even still !

In addition to the FCOM descriptions of both, I think there is some imprecise use of terms in the FCOM, specifically the word, "protection". The word is used in both contexts to describe "prot" and "floor" and so is informal, whereas the terms alpha-prot, (or "α prot") and alpha-floor are formal terms.

I think the words, "prot" and "floor" are being understood as inter-changeable but am thinking that they are not. I think it is significant that "alpha-floor" is discussed only in ATA22 (Autoflight) but "alpha-prot" is not. Alpha-prot is only discussed in ATA27 (Flight Controls), with "alpha-floor" only being used in the context of, and with reference to, "alpha-prot".

Put differently, I think one can be in "alpha-prot", (a pitch-law mode) without being in alpha-floor, (an A/THR mode), but one cannot be in alpha-floor without first having "gone through" alpha-prot. The schematic in the graphic above even shows this.

So you can be in "alpha-prot" at any Mach number, but alpha-floor mode is restricted as described in the para. from the FCOM.

The important key to understanding this and using all this information from a pilot's pov is to remember what Chris Scott has outlined in the discussion on "disengagement of the AoA law". Both the AirProx and the A340 2011 Events were controllable either by leaving the stick alone, (but that leaves the airplane to "select" Max AoA in a phase-advanced scenario and is therefore not a good option), or to take control to gently push the stick forward to take the airplane back to Nz law (because, as has been observed previously, one is 'unloading the wing', so to speak) and deal with the airprox on the one hand, and sort out the mess the crew of the A340 - 2011 made for themselves on the other.

Alpha-Floor inhibition above Mach 0.53
 

Thanks Don,

That nails it in line with what OK465 has been telling us all along?

My (perhaps simplistic) understanding has always been that Alpha-Prot and Alpha-MAX relate to the flight control laws, whereas Alpha-Floor is purely an A/THR mode, albeit an extremely powerful one at low altitudes.***

AFAIK, Alpha-Floor is always higher than Alpha-Prot. As you know, it has always been necesssary to inhibit Alpha-Floor below a certain height on landing (IIRC, 50R on the A320, but I see 100R on the big 'buses), although it was widely thought that one fellow captain did not take that into account during his cavalier fly-by at Habsheim...

Quote from OK465:
Well I'll tell ya folks, even if you pitch up from 35000 feet and 265 KCAS to 20 degrees nose high with idle thrust, the lowest Mach you will see before 'prot' finally gets the nose down is about 0.58....no alpha floor.

YES...! :ok:


***(by edit)
You may not be aware that AI originally introduced Alpha-Floor on the A310 in 1983. The A310 has conventional flying controls (except for the roll-spoilers, which are FBW). So, on the A310 and A300-600, you have to push very hard if Alpha-Floor engages...

Hello Chris;

Re, "That nails it in line with what OK465 has been telling us all along?"

Yes, I think so! So I probably mis-read his comment about 'splainin' how alpha-floor can occur below M0.53! - it was a statement, not a question! Ahh, me.

Please do not spread disinformation, that captain was definitely not relying on A-FLOOR to get the thrust on.

DonH has now confirmed that earlier comment

Terms
 

@ DonH
AlphaFoor is strictly A/THR and only triggers within certain parameters.
It also triggers based on a set of variables not just indicated airspeed.
No relation to AlphaProts in flight controls other than Normal Law must be active.

Valpha prot is the AOA threshold from where the side stick will command AOA directly instead of load factor demand. Valpha max is the maximum AOA that flight control computers will let the side stick achieve and/or hold. With AP is off and side stick neutral if the aircraft is pitched up beyond Valpha prot due to disturbance it will try to pitch down to Valpha prot and not pitch up to alpha Max. If the pitch up is rapid then Phase advance will bring this threshold to a lesser value. The effect of activation of this protection mode is that trim freezes at entree point and the aircraft will stabilize at that AOA when stick free, But Pilot's action in A340 case would be same push forward to stop climb and that would bring flight controls out of protection mode to load factor demand which was not done and instead attempt was made to engage AP which does not engage when in protection. So this incident is not attributable to FBW characteristic.

Quote from vilas (my underscoring emphasis):
"With AP is off and side stick neutral if the aircraft is pitched up beyond Valpha prot due to disturbance it will try to pitch down to Valpha prot and not pitch up to alpha Max. If the pitch up is rapid then Phase advance will bring this threshold to a lesser value."

In an otherwise accurate and succinct analysis, I think you may still be confusing the phase-advanced threshold of an AoA-mode engagement with the target AoA once AoA mode is engaged. For the reasons I stated in my earlier reply to you, with neutral sidestick that target can only be Alpha-Prot itself; not the AoA value at its phase-advanced engagement, which has no further relevance.

Chris
I am in agreement with you that after initial lowering of threshold subsequently it will stabilize at actual V alpha prot because the purpose of phase advance is to preempt the change over point. I was explaining the normal change over at Valpha prot.

Chris
Somewhere you have said "So my question remains: why does it not revert to Normal Law once the AoA falls to a safe value?" I take it you mean normal AOA (Vls)
Let me attempt to answer that. Basically the high angle of attack protection is provided to prevent stall while dealing with wind shear or terrain issues where you are likely to deliberatally invoke the protection. When the angle of attack has fallen to a safe value (I presume it as less than alpha Max) any attempt to increase speed beyond Valpha prot by lowering the pitch is left to the pilot as conscious decision. If it was automatically done without assessing the situation it would have been unsafe.
You will remember that when you change from NAV to HDG, CLB changes OPCLB but in descent it does not change to OPDES but to present VS. The reason is same. In CLB or OPCLB the thrust is in THR CLB but in DES it need not be in thrust idle and in such a case change to THR IDL as it would in OPDES will pitch the aircraft down increasing ROD. So this is left to the pilot as a conscious decision.

I don't think Chris was intentionally misinforming - let's face it, the story has been misremembered and re-told in slightly different ways so many times over the years - especially in the media - that deviations like that are pretty much the norm. It's pretty much only pedants like me and you who are militant about getting it right! :8

Indeed - however your comment here:

seems at odds with what was actually changed. It used to be that to cancel AoA protection - in which the nose comes up - one had to push the nose back down, which seems reasonably intuitive to me. The change meant that leaving the stick in neutral for a certain period of time would also return you to Normal Law - again, not exactly rocket science.

It may just need a little help from atmospheric disturbances. Without those, it would just maintain alpha-prot in a phugoid motion which may be damped.

P.S. I think you mean Nz-law. The 'normal' protections are included in Normal Law, and may not exist in Alternate Law.

Agreed. Picky, picky, picky... ;)

Quote from Dozy Wannabe:
"I don't think Chris was intentionally misinforming - "

I don't think anyone will ever know for sure, so - more out of respect for Conf_iture than the gentleman concerned - I amended my post this morning to reflect fact, rather than supposition.

Hi vilas,

I like your explanation of why AoA mode used to remain engaged until a sidestick was displaced forward (as in the first two disengagement conditions listed in the FCOMs posted by DonH). The interesting development is that, sometime after the A340 AIRPROX incident of 2001, the introduction of the third disengagement condition seems to have been completed across the fleet.

The third disengagement condition is not unlike what I was proposing a few days ago in reaction to the A340 AIRPROX. (That was prior to HN39 drawing our attention to the BEA Report into the AF A340 Serious Incident of 2011.) I was concerned that a short-term up-gust had triggered AoA mode by phase-advance, and AoA mode had initiated the undesired climb with neutral sidestick which the crew were (admittedly) slow to correct.

Having studied the traces provided by the BEA of the AF A340 incident, where the third disengagement condition operated five times, I am reviewing my opinion. The first disengagement into normal (Nz) mode was, as you know, shortly followed by a significant exceedance of Alpha-MAX. It seems probable that the wing effectively stalled at that point, which probably would not have happened but for the temporary disengagement of AoA mode.

Chris - the ultimate question at the end of the day is "how much hand-holding is a good idea to provide?". It's reasonable to assume that Alpha Prot and Alpha Floor as designed were in the main intended to assist a pilot who is already commanding nose-up with the stick, such that the pilot's command is followed to the best of the aircraft's ability while the aircraft remains flying. In this airprox incident we have a situation where one or both of the alpha protection modes were triggered by transient weather conditions, which were outside of the original spec.

Contrary to legend, the protections were designed to assist a pilot who, for whatever reason, has been forced to take the aircraft to the edge of the safe operating envelope - counteracting handling mistakes is and always was a secondary design consideration.

In this case we have an incident during which - for whatever reason - a protection mode has actually had a negative impact on the safety of the flight, hence the change in logic.

Chris &Dozy
Some more thoughts on the A340 incident:

The report states “ The over speed warning(“OVERSPEED”) was triggered(3) andthe “Master Warning” warning light came on. The crew stated that they were very surprised by this warning. Mach reached 0.87.” The warning is set to trigger at MMO+.006 i.e.M.866 and it triggered correctly why was the crew surprised? The report says “during the event the high speed protection was not activated”. This is surprising as it should have activated and according to A340 FCOM “The autopilot disconnects when high speed protection goes active”.

The report says” F-GLZU recorded data showed that the AP would have remained connected if it had not been manually disengaged”.I don’t see any guarantee of this as high speed protection should have activated or was about to activate and that would have disconnected the AP.

Then the PF disconnected the AP and applied ¾ back stick which put the aircraft in high AOA protection.

“From the start of thepitch-up input until point n°9, the high angle of attack protection was activated several times”. This statement has been misunderstood by some bloggers as automation causing unnecessary pitch up to Valpha MAX. This is not what happens in AOA protection rather it only can cause pitch down to Valpha prot with stick neutral. However if AOA is left at Valpha prot then the mode of pitch control may keep changing from load factor to AOA mode as the AOA changes due to disturbances. That is not same as pitching up and down. The only thing is if left alone, in AOA mode it will stabilize at alpha prot speed. In any case Pilot’s action should have been to push the nose down and recover the speed. So automation has not played any adverse role in this.

This has nothing to do with 'returning to Normal Law'.
Your approximations generate confusion on a complex System that on top of it keeps changing with time.
As 'it is not exactly rocket science' why do you get it wrong then ... ?

Sorry Chris but I just can't understand your thinking here as part of the procedure to present the Airbus at high AoA is specifically to inhibit A/THR to prevent Alpha Floor to spoil the demonstration.
How can you suggest the guy was waiting for Alpha-Floor to kick in when his initial intention was to prevent it to interfere in the first place ?

DozyWannabe

"It used to be that to cancelAoA protection - in which the nose comes up - one had to push the nose backdown, which seems reasonably intuitive to me. The change meant that leaving thestick in neutral for a certain period of time would also return you to NormalLaw - again, not exactly rocket science".
I am sorry but it appears that you have completely misunderstood normal law and protections. Normal side stick fore and aft movement is a load factor demand and computers position elevator and stab to give you that. Pitch control through load factor demand has an advantage in that aircraft response remains same at all speeds. High angle of attack protection which begins at Valpha prot the pitch control is changed to directly achieve through AOA change and not through load factor demand. Valpha Max is maximum AOA aircraft can be pitched upto. When the angle of attack increases deliberately as in case of wind shear and GPWS issues or through atmospheric disturbance the pitch control is changed to AOA mode since that is more appropriate for the situation, this does not mean aircraft pitches to alpha max, rather it tries to pitch down to alpha prot which is the first low speed threshold and pilot has to overcome it by backward pressure if he wants more pitch up.. If you release the stick aircraft will pitch down and maintain that AOA (Valpha prot). If the contingecy is over and you want return to normal speed you push the stick forward as you would do in any aircraft and the pitch control changes back to load factor demand. You have to remember that only method of pitch control is changed aircraft doesn't pitch up on its own. During all this aircraft remains in normal law. In the incident we are discssing the protection mode had no negative impact whatsoever. It is definitely not rocket science but you have not understood it at all.

With the sidestick left in neutral the airplane will pitch up on its own -

-- in AoA law at alpha=alpha-prot if the speed is greater than Valpha-prot
-- in Nz law when the pitch attitude is greater than the pitch which will maintain constant airspeed - AoA increases to maintain Nz=1 when airspeed reduces.

HN39
With the sidestick left in neutral the airplane will pitch up on its own -

-- in AoA law at alpha=alpha-prot if the speed is greater than Valpha-prot
-- in Nz law when the pitch attitude is greater than the pitch which will maintain constant airspeed - AoA increases to maintain Nz=1 when airspeed reduces.
No way. I think you should read flight controls chapter. It is high angle of attack protection. It begins at alpha prot. A protection can only be called protection when it prevents the pilot from doing it. In high speed protection if it pitches down it is not a protection it is destruction. Similarly in high AOA protection if AOA increases what kind of protection is that? Your first case, the speed is greater than Valpha prot you are not at all in any protection zone and side stick commands load factor so why would it go to alpha prot? If you intentionally keep pitching up it pitches up to alpha max and no more. if you release the SS it will not maintain alpha max but pitch down to alpha prot.

AoA mode - engagement and disengagement criteria
 

vilas,

I agree with nearly all of your latest thoughts about the AF A340 Incident. You make some valid points, including the apparent discrepancy in relation to the overspeed protection. The latter had disengaged the AP in the A340 AIRPROX of 2001, but seemed already overdue to do so on the AF A340 when the backward movement of the F/O's sidestick manually disengaged it. Like you, I find the BEA statement that you quote to be confusing.

Where our respective understandings do still differ, however, is in what may happen after AoA-mode engages by phase-advance resulting from a brief up-gust.

It's worth noting here that the extra (3rd) disengagement condition available on the a/c in the AF A340 incident of 2011 caused the first 5 of the 6 reversions to Nz mode, for which the sidestick remained neutral. In the case of the A340 AIRPROX incident, where the 3rd condition had not been incorporated, AoA mode consequently persisted throughout the main event - even after a small push-forward on the sidestick for less than a second - until instantly disengaged by a selection of more than half-forward sidestick.

Quote (my emphasis):
"This is not what happens in AOA protection rather it only can cause pitch down to Valpha prot with stick neutral."

I disagree here, as I tried to explain in my last reply to you. A brief up-gust can cause the phase-advanced value of AoA to exceed alpha-prot briefly. Actual AoA can remain below alpha-prot, and when the up-gust ceases the phase-advanced value will return to the real AoA. Meanwhile, however, AoA mode has engaged. Under the present-day three-condition disengagement logic, AoA mode would disengage after 0.5 sec if both sidesticks were neutral. Under the two-condition disengagement logic in the A340 AIRPROX of 2001, it would remain engaged with neutral sidesticks. So in the latter case the EFCS will have to increase the AoA to Alpha-Prot, which will probably involve a pitch-up. This is what I think happened in the AIRPROX case, and why I questioned the disengagement criteria.

One of the vulnerabilities of my argument in the A340 AIRPROX case is that, AFAIK, we don't have any estimates for what would have happened to the pitch (and, therefore, the flightpath) if the AoA mode had reverted to Nz, delivering 1.0 G for the long period that the sidesticks remained neutral.

Quote:
"In any case Pilot’s action should have been to push the nose down and recover the speed. So automation has not played any adverse role in this."

Your first sentence is correct for both events. Your second sentence is arguable in the A340 AIRPROX case, as I've explained above. In the AF A340 case, however, the F/O's early action (pulling the stick) was - like AF447 - what started the zoom-climb. Once he had released the stick, the AoA mode was free to disengage each time the AoA happened to fall below Alpha-Prot for half a second. Each time, Normal mode targeted 1.0G, which on 5 occasions led to re-engagement of AoA mode. As previously noted, the first re-engagement was too late to prevent an exceedance of Alpha-MAX (perhaps in a gust).

These two events may present conflicting arguments for and against the third disengagement condition. I suspect that AI engineers may have been revisiting AoA-mode engagement and disengagement criteria since 2011.

Vilas,

I think you will find that HN39 understands flight mechanics as well as any writing in these pages.
Although I agree with much of what you write, I think you nevertheless have some things wrong:

Alphaprot is NOT the same as normal law, so when you wrote:

this was not correct.

I think you are failing to distinguish between angle of attack and pitch. When HN39 writes
he is properly relating the flight dynamics. With stick neutral, alphaprot will control the aircraft to a constant AoA, to whit the alphaprot for the applicable Mach Number. It controls AoA; it does NOT control pitch. If the airspeed at this time is greater than that appropriate to 1g at alphaprot then the aircraft will develop a normal acceleration and start to change its flight path. Unless corrected by a change in pitch this will result in a change in AoA. But the system is trying to control to a constant AoA so it must apply a pitch up to compensate. This, I think, is what HN39 is alluding to.

Quote from Dozy Wannabe (my insertions in square brackets):
"In this [A340 AIRPROX] case we have [had] an incident during which - for whatever reason - a protection mode has actually had a negative impact on the safety of the flight, hence the change in logic."

Hi Dozy,

Prior to our attention being drawn to the AF A340 case, I caused some controversy here by opining that the disengagement criteria for what we are now calling the AoA mode (of Normal Law) should be relaxed. I'm now having second thoughts, because the new (third) criterion MAY have allowed the AF A340 to stall during the second engagement of AoA mode. This followed a brief (3-second) resumption of Nz mode, during which the a/c pitched up by about 4 degrees.