How safe is (airbus) fly by wire? Airbus A330/340 and A320 family emergency AD

[Kimon]

@ Hazel Nuts

Yes, AOA, Pitots and Stats blocked.

[CONF iture]

Where did you get such information Kimon ... ?

[Cool Guys]

A couple of posters seem to be using 737 rudder hard over issue to lessen the severity of this AB issue. The 737 rudder issue was obviously a very serious issue with lives lost etc but it should not be used as any form of excuse for another safty concern. There is a trace of the A verses B debate in this which should have little relevence when debating serious safty issues.

[Kimon]

@ CONF iture

Sorry got my wires crossed:
Airspeed was working normally though αlpha PROT was in an unusual place and thus pitot and static not blocked.

[HazelNuts39]

Originally Posted by Chris Scott

Based on what little we know, it would be remarkable if the BEA did not ensure there was a formal investigation. It would be interesting to know the minimum vertical-g during the pitch-down, and whether anyone was hurt. IIRC, serious injury or substantial damage = accident, which would mandate an investigation. in the absence of either, it would still be bizarre for them not to publish some kind of bulletin.

Based on what little we know, just to give an idea, my estimate of minimum vertical-g during the pitch-down:

[Chris Scott]

Eva Air minimum normal-g during push over

Thanks (as always) HN39,

Your figure seems to be about +0.35g? Even allowing for the approximate nature of the data you're working with, does it seem probable that no one went ballistic? (No pun intended.)

[HazelNuts39]

@Chris Scott:

Yes, IMHO that would seem probable.

[DozyWannabe]

From the EAD (emphasis mine):

An A330 aeroplane experienced a blockage of all Angle Of Attack (AOA) probes during climb leading to Autopilot (AP) disconnection and activation of the alpha protection (Alpha Prot) when Mach number increased.

If in climb phase, would we not be talking about a significantly different scenario than cruise, given that it would be starting from a higher positive pitch and thrust setting?

[Lyman]

Would it also be likely below 25,000 feet, involving a potential 'BUSSt'?

Does the Alpha prot bounce the A/P? Does BUSS put the a/c in Direct?

IOW, how does the flight logic know the probes are frozen to deselect autopilot?

Doesn't Direct, once active, stay active for the duration of the flight?

Dozy?

[Chris Scott]

Those who are not following the thread on probe icing may nevertheless wish to see this post by A33Zab, quoting David Learmount:

http://www.pprune.org/rumours-news/5...ml#post7618295

[HazelNuts39]

Originally Posted by DozyW

If in climb phase, would we not be talking about a significantly different scenario than cruise, given that it would be starting from a higher positive pitch and thrust setting?

I don't think it would be significantly different. Other information indicates the entry into alpha-prot occurred near top of climb. The flight path angle would then not be so large that it would change the scenario significantly. Furthermore, the V/S of 9400 fpm together with the likely TAS and the pitch of -12 degrees yields an AoA close to the zero-lift AoA of -1 degree, IOW corresponds to close to zero 'gee'.

Originally Posted by Lyman

Does the Alpha prot bounce the A/P? Does BUSS put the a/c in Direct? IOW, how does the flight logic know the probes are frozen to deselect autopilot?

Alpha-prot bounces the A/P. The system does not know that the erroneous AoA is due to blocked sensors.

[737Jock]

Chris Scott, that article on the other thread is incorrect. You wrote:

Those who are not following the thread on probe icing may nevertheless wish to see this post by A33Zab, quoting David Learmount:

I posted the following on the other thread:

Totally agree, the AFM TR should be followed!
But alleged 'total loss of pitch control' is besides the truth.

The article is actually incorrect in that sense. The OEB is structered around recognizing 3 different scenarios, however if you do not recognize these scenarios and alpha prot engages, you WILL loose all pitch control until you turn of 2 ADR's to force alternate law. The proof is in the pudding:

The A320 memory items from the OEB (my bold):

At any time, if the aircraft goes to an unmanageble pitch down attitude despite continuous deflection of the sidestick in the full backward position (in case the flightcrew missed the below symptoms or delayed the application of one of the below procedures):
One ADR........Keep on
Two ADR's......Turn off

So if you recognize the symptoms on time, stabalize the flightpath and keep one adr on.
If you are too late, alpha prot will engage and you will loose all control over pitch which can only be resolved by the memory items.

The A330 crew lost all control, but they regained it by turning ALL adr's off which forced alternate law.


So there you have it, airbus themselves say you can loose all pitch control! It's really not helpful to try to convince people otherwise. David Learmount is unfortunately incorrect.

In order to be in line with the OEB this is what it should read:
The crew turned off all three air data reference (ADR) units. This action took the aircraft out of normal flight law into alternate, which de-activated the stall-protection system. Then the pilots levelled the aircraft and reviewed the situation and decided to divert. During descent, Airbus notes, the angle-of-attack vanes became unstuck once more.

[Lyman]

HazleNuts39

Yes, the logic is not there to decide to switch off the AutoPilot...

The paper is disingenuous, for it announces first A/P switched off, then AlphaProt..

It could be a mistake, but it is glaring.

The crew and the airplane must wait for Mach to trigger the Protection.

It is like an "Ambush", a Gotcha..... Not unknown in this aircraft, its talent for the surprise....

A corollary... Bonin applied aft stick, and accomplished a frustrating lack of response (in his mind). We dismiss his actions as silly, but when the aircraft itself mimics Bonin, (though opposite in Pitch direction), we defend the machine, though it is quite obviously hazardous in its performance?

Both aircraft EVA And AF are victim to similar scenario... One lived to tell it in person...

It takes prejudice to dismiss the potential for disaster in the one, and accept with disapproval the actual accident accomplished in the other....

330 -1

Bonin -1

A Tie

[Chris Scott]

Hi 737Jock,

My purpose in providing a link to A33Zab’s post, in which he quoted David Learmount’s Blog, was – as I said - simply to draw attention to it: not to endorse David’s report. Perhaps I’m not the only one who is occasionally finding difficulty in following two PPRuNe discussions running more or less in parallel on the same subject, although this thread has a wider remit than the other one.

David Learmount’s piece is dated December 7th, and would have been written in good faith with the information available at the time. However, as also happens often on PPRuNe, he has put the description of the event in quotation marks without specifying the source. It certainly appears that his source reported the main events in the wrong order.

The news to me is that it seems to confirm the probes unjammed later, presumably during the descent to their alternate airfield.

You say:
“David Learmount is unfortunately incorrect.
In order to be in line with the OEB this is what it should read:
The crew turned off all three air data reference (ADR) units. This action took the aircraft out of normal flight law into alternate, which de-activated the stall-protection system. Then the pilots levelled the aircraft and reviewed the situation and decided to divert. During descent, Airbus notes, the angle-of-attack vanes became unstuck once more.”

Sounds good to me.

[CONF iture]

Learmount is in direct touch with Airbus ... ?

• 11300ft in the climb
• 250kt
• minus 12C
• AOA probes stuck at 5 deg
• then FL310

but where are all the other relevant data ... starting with the date of the event ?

As noted by 737Jock, the article is misleading and has IMO a taste of damage control.

"Airbus says this is the only know occurrence of this type, but it is reviewing the design of its heated sensors and their resistance to icing."

That's not what I'm looking for. I want Airbus to develop the most simple procedure to give back control to the crew in the shortest time when the protections do silly things.

[jcjeant]

starting with the date of the event ?

I suggest ... 18 November 2012

[TTex600]

"How safe is (Airbus) fly by wire?"

FBW, apparently quite safe.

Protections, statistically safe.

Could be far safer.

The 320 was and is a techno marvel, but continued development would appear to be lacking.

TTex, not a Boeing fan either. I'd rather sweat in a DC9.

[DozyWannabe]

Originally Posted by TTex600

The 320 was and is a techno marvel, but continued development would appear to be lacking.

That's not strictly true. Development and refinement of the FBW systems has continued significantly in the 24 years since the aircraft went into service, but there are limits to how much can be altered with the A320 itself without breaking type commonality. Hence the A320NEO programme, presumably.

As Chris Scott alluded to earlier, the fact that the original A320 was subject to later and more stringent certification requirements than the original B737 - along with the fact that it was designed around a more modern control and avionics setup - should mean that more significant changes can be applied to the NEO than were possible with the NG.

That said, so much preparatory work went into R&D of the systems that ended up in the original A320 (much of which significantly pre-dated the official project inauguration in 1982) that the groundwork it laid out was correct and proven (in engineering terms) to a degree that was unprecedented at the time. Also, because of the digital nature of the flight systems, a lot of work went on "under the hood" in the years between then and now that would not be obvious to those who weren't inclined to dig deep enough to find out.

TTex, not a Boeing fan either. I'd rather sweat in a DC9.

Interesting how people gravitate to different types. Maybe I'm of the wrong generation, but the DC9 always struck me as a bit flimsy in appearance and backward in engineering terms. This could also be coloured by my soft spot for the BAC 1-11 (which I thought was a lovely design and built like it was hewn from granite but let down by the choice of engine).

[TTex600]

Originally Posted by DozyWannnabe

Interesting how people gravitate to different types. Maybe I'm of the wrong generation, but the DC9 always struck me as a bit flimsy in appearance and backward in engineering terms. This could also be coloured by my soft spot for the BAC 1-11 (which I thought was a lovely design and built like it was hewn from granite but let down by the choice of engine).

So any anecdotal comment/opinion regarding Airbus demands an immediate response from you demanding facts to back up the assertion, yet you have no problem offering your own opinion about the DC9.

Can't say that I'm surprised.

[DozyWannabe]

Not at all sir, just offering a light-hearted comment! The birds we prefer (and the reasons we love 'em) are a personal thing for the most part, and I respect your choice even if I don't necessarily agree.

I swear I'm not the dour contrarian some take me for...