Sponsor Message:
Civil Aviation Forum
My Starred Topics | Profile | New Topic | Forum Index | Help | Search 
XL/NZ A320 At Perpignan Interim Report Out  
User currently offlineAustrianSimon From Austria, joined May 2008, 69 posts, RR: 0
Posted (5 years 4 months 4 weeks 1 day 9 hours ago) and read 5313 times:

The French BEA have released an interim report, see:

http://avherald.com/h?article=410c9cec/0009

A low speed test went wrong, alpha floor protection was aimed at but apparently did not activate and lead to a stall.

14 replies: All unread, jump to last
 
User currently offlineOsiris30 From Barbados, joined Sep 2006, 3192 posts, RR: 26
Reply 1, posted (5 years 4 months 4 weeks 1 day 7 hours ago) and read 5090 times:



Quote:
The attachments with the graphical readout of FDR parameters contains an oddity not mentioned in the report: both the left and right hand angle of attack sensors show a constant value starting from 15:05Z until impact.

Any Airbii drivers here care to enlighten us simpletons on what this could mean, what could cause it and what the implications could be of this, if there was an instrumentation malfunction in the systems that fed data into the flight control software (if this even feeds that)?



I don't care what you think of my opinion. It's my opinion, so have a nice day :)
User currently offlineKaneporta1 From Greece, joined May 2005, 739 posts, RR: 12
Reply 2, posted (5 years 4 months 4 weeks 1 day 7 hours ago) and read 5084 times:



Quoting Osiris30 (Reply 1):

Quote:
The attachments with the graphical readout of FDR parameters contains an oddity not mentioned in the report: both the left and right hand angle of attack sensors show a constant value starting from 15:05Z until impact.

Since in the report the angles of attack and bank are mentioned until the crash, and the "oddity" is not mentioned in the report, I would assume it's an error in the plot/printout rather than something that happened on the aircraft.



I'd rather die peacefully in my sleep, like my grandfather, not terrified and screaming, like his passengers
User currently offlineOsiris30 From Barbados, joined Sep 2006, 3192 posts, RR: 26
Reply 3, posted (5 years 4 months 4 weeks 1 day 6 hours ago) and read 5028 times:



Quoting Kaneporta1 (Reply 2):
Since in the report the angles of attack and bank are mentioned until the crash, and the "oddity" is not mentioned in the report, I would assume it's an error in the plot/printout rather than something that happened on the aircraft.

Entirely possible, just surprised they would release it that way. In reading the report I'm left with a ton of questions.. Has anyone else read the details on pages 20 through the top of 23. The pilot inputs and aircraft reactions don't seem to line up with each other at all (in some instances). Again anyone with input? (Kaneporta, can you shed any light on this, I'm trying to wrap my head around what's contained therein and my brain is NOT behaving today  Wink )



I don't care what you think of my opinion. It's my opinion, so have a nice day :)
User currently offlinePihero From France, joined Jan 2005, 4388 posts, RR: 76
Reply 4, posted (5 years 4 months 4 weeks 1 day 4 hours ago) and read 4906 times:
Support Airliners.net - become a First Class Member!

Quoting AustrianSimon (Thread starter):

A low speed test went wrong, alpha floor protection was aimed at but apparently did not activate and lead to a stall.

Simon, I wish you'd be more careful in your writing, especially when you have a site which could be a reference to air safety buffs : Alpha floor is only available in "Normal Law". During the test, they were not in normal law (Both FACs have been switched off earlier, leaving the plane in "Alternate".....with a pitch "Direct" upon selecting the gear down).

Quoting Osiris30 (Reply 1):

Any Airbii drivers here care to enlighten us simpletons on what this could mean

The page 33 read-out is for the whole flight, and the time-range we'd be interested in would only cover the last two millimeters of that graph.
I, too would like the AoA sensors readings during the manoeuvre, but if the stall warning was activated, the sensors were working as per necessity.
The AoA information is paramount in determining the approach-to-stall situation and the triggered protections : Alpha Prot / Alpha Floor...etc...and the stall warning if you're not in Normal Law

Quoting Osiris30 (Reply 3):
In reading the report I'm left with a ton of questions.. Has anyone else read the details on pages 20 through the top of 23. The pilot inputs and aircraft reactions don't seem to line up with each other at all (in some instances).

That aircraft was in -or very close to - a stall and the response to control inputs is to say the least very sluggish...the ailerons/spoilers would have lost most of their efficiency and for roll control, a rudder input could well be necessary (which the PF did for some long time...)
One last aspect is that apparently, close to a stall, the pitch-up moment of the engines at TOGA is higher than the elevator pitch authority ( I haven't done it in the sim so it's new to me ).

Reading this prelim, I can't help but thinking that we won't see the end of this investigation for a very long time... and a lot of anguish and arguments (It's started already, see Air New Zealand CEO interview).
Being a known cynic, I expect a lot of French and Airbus bashing.
I'll be there...

And this is the official document from the BEA site that Simon did not link :
BEA Interim Report

[Edited 2009-02-24 18:19:59]

[Edited 2009-02-24 18:21:15]


Contrail designer
User currently offlineOsiris30 From Barbados, joined Sep 2006, 3192 posts, RR: 26
Reply 5, posted (5 years 4 months 4 weeks 1 day 3 hours ago) and read 4860 times:

Pihero:

First of all thanks for taking the time to reply. Appreciated.  Smile

Quoting Pihero (Reply 4):
That aircraft was in -or very close to - a stall and the response to control inputs is to say the least very sluggish...the ailerons/spoilers would have lost most of their efficiency and for roll control, a rudder input could well be necessary (which the PF did for some long time...)

This much I could figure, as from what I can gather the entire point of heavy input occurred where the IAS was at or below 110 kts, which would (depending on the configuration) by at or below stall speed based on what little I know of the aircraft.

Quoting Pihero (Reply 4):
One last aspect is that apparently, close to a stall, the pitch-up moment of the engines at TOGA is higher than the elevator pitch authority ( I haven't done it in the sim so it's new to me ).

Well it certainly seems something unexpected happened. Looking at the hours on the PIC there was certainly no lack of experience at play. I don't suppose you'll be taking a sim ride any time soon? This would seem to be rather crucial information to have *if* such an effect is in affect (sorry for that being such a clumsy sentence).

Quoting Pihero (Reply 4):
Being a known cynic, I expect a lot of French and Airbus bashing.

As we all know I'm not an Airbus fan (although it has nothing to do with the French as a people/nation). However, when people die such bashing is immature and classless and I personal won't engage in it (for what it's worth).



I don't care what you think of my opinion. It's my opinion, so have a nice day :)
User currently offlinePihero From France, joined Jan 2005, 4388 posts, RR: 76
Reply 6, posted (5 years 4 months 4 weeks 1 day 3 hours ago) and read 4823 times:
Support Airliners.net - become a First Class Member!



Quoting Osiris30 (Reply 5):
Well it certainly seems something unexpected happened. Looking at the hours on the PIC there was certainly no lack of experience at play. I don't suppose you'll be taking a sim ride any time soon? This would seem to be rather crucial information to have *if* such an effect is in affect (sorry for that being such a clumsy sentence).

High AoA situation is not really something we train for : Stick shaker or aircraft buffet is enough and recovery is initaited at once.
I've only done full stalls during my initial training and on light aerobatic airplanes.
On the 'Bus, we try and trigger the Alpha Floor (automatic TOGA) and that's it, end of exercise !
But they were not in "normal law", they basically had a 737-type in their hands.

On normal law, the pitch trim is automatic. It's not in "alternate" or "direct" and apparently, they never used the trim wheel (the THS stayed frozen at the value it had when they went into "alternate"), thus the apparent unability to check the increasing pitch up movement with just the elevator control once they removed the side-stick from the "full forward" stop.

Quoting Osiris30 (Reply 5):
I don't suppose you'll be taking a sim ride any time soon?

In a weeks-time.



Contrail designer
User currently offlineAustrianSimon From Austria, joined May 2008, 69 posts, RR: 0
Reply 7, posted (5 years 4 months 4 weeks 14 hours ago) and read 4493 times:

Quoting Pihero (Reply 4):
Simon, I wish you'd be more careful in your writing, especially when you have a site which could be a reference to air safety buffs : Alpha floor is only available in "Normal Law". During the test, they were not in normal law (Both FACs have been switched off earlier, leaving the plane in "Alternate".....with a pitch "Direct" upon selecting the gear down).

Hello, Pihero,

the alpha floor is not my "invention". The interim report clearly states, that the New Zealand captain aimed at an alpha floor test, and the XL captain performed the test, but then pushed thrust levers to near TO/GA when the speed had dropped to 99 KIAS.

The report never mentions the Angle of Attack Sensor problems except for that one single graph in the attachments, where the BEA even made a remark that the sensors appear to be frozen.

May I suggest, you read the report carefully yourself before "educating" me about what I wrote based on a very careful examination of the official BEA report?

Just to avoid a possible problem: In the meantime I had the French version of the report cross checked for any possible differences to the English one I used. The comparism has come up with the result, that the two reports are identical. No mention of the AoA in the French report as well, and a clear mention of the alpha floor test in the French version, too.

Apart from that, where do you have the information from, that the AoA Sensors were turned off? Nothing like that mentioned anywhere.
Servus, Simon



[Edited 2009-02-25 08:19:12]

User currently offlineAotearoa From New Zealand, joined May 2005, 135 posts, RR: 0
Reply 8, posted (5 years 4 months 4 weeks 8 hours ago) and read 4144 times:

Gidday All

Looking at the report this morning, it is important to clarify the a/c was in Normal law as the low speed test was commenced. The change from Normal to Direct law happend well below the minimum speed the a/c should have stabilised at. The fact that Alpha Floor did not activate, and the the a/c effectively stalled, indicates a serious system problem was evident. I can only suspect that this was related to the fixed values eminating from the two errant AOA vanes. The report puts very little emphasis on the system issues. I can only presume that this work is on-going.


User currently offlinePihero From France, joined Jan 2005, 4388 posts, RR: 76
Reply 9, posted (5 years 4 months 4 weeks 5 hours ago) and read 3973 times:
Support Airliners.net - become a First Class Member!

Gruss Gott, Simon,

Quoting AustrianSimon (Reply 7):
Apart from that, where do you have the information from, that the AoA Sensors were turned off? Nothing like that mentioned anywhere.

What I said was that the "alpha floor" is only available in "NORMAL LAW"
Page 19, you have this excerpt :
"...At 15 h 36 min 47 s, when the airplane was stable at flight level 120, the Captain asked "you
want alternate law" and the New Zealand pilot answered "okay alternate law".
At 15 h 37 min 08 s, the autopilot was disengaged. Nine seconds later, the call-out « FAC 1,
and FAC 2 is coming now » was made and the Y/D 1 was recorded as FAULT. At 15 h 37
min 22 s, the Y/D 2 was recorded as FAULT, the control law for pitch passed from normal to
alternate and the control law for roll passed from normal to direct.

The prelim doesn't say whether this test was just a check of the degradation in flight control laws...but it seems so as the report mentions an"Alternate law test.....performed"

I apologise, a quick read made me think that they kept the FACs off.

as later in the same page...
"The New Zealand pilot then said "Low speed flight is now probably next" then described the
sequence of events for the flight at low speed. The Captain asked if his intention was to go
down to VLS and alpha prot."

From that sentence, we can assume that the initial requirement was for a gentle entry into "alpha prot", then...
"He confirmed that and said that, on reaching VLS, it would be
necessary to pull up hard to go as far as alpha floor. The Captain answered that he knew.
The New Zealand pilot continued, saying that afterwards it would be necessary to push,
disengage and re-engage..."
( meaning the autothrust ? )

That's where the gray area begins : Apparently, they were fully aware of the events' sequence in the test...had it gone fine...
But it didn't :
1/- Because they were at an altitude over two thousand feet, they didn't get the Low energy warning ( "Speed Speed" ).
2/- They let the speed drop below the planned 107 kt...down to 99 kt, at which point, they selected TOGA - manually-.
3/- The big trap : TOGA induces a fair amount of nose-up pitch moment...No problem IF they were in "Normal law" - auto trim is available and can easily cope -... but they weren't any more :two seconds later they went into "alternate", then, because the gear was down immediately to "Pitch Direct"...
As the pilot never used the trim wheel, the sidestick authority wasn't enough to counteract the engine-induced pitch-up....Result : Attitude well above 50°, speed at 40 kt....Tough to recover from...

Now the AoA sensors : They were rather obviously feeding the FACs with erroneous / frozen (?) data... but the initial stall warning at 15.45.05 seems correct... if so where did the info come from ?
But they had another two stall warnings, one seems also correct at 15.45.35 at about the same value as the first one, but the third warning was way out of normal : It kept sounding at a CAS of just under 180 kt...the print outs are not complete, I would have liked to have the acceleration graphs.

for the rest of the print-out, the "frozen" THS prevented them from getting the max out of the pitch capability of the airplane...and they were too fast.

From the above, I still think that your one-liner

Quoting AustrianSimon (Thread starter):

A low speed test went wrong, alpha floor protection was aimed at but apparently did not activate and lead to a stall.

lacks a lot in accuracy, as it could be construed as one - yet again - anti-Airbus rant, which I've seen too many of on this site.
Don't you agree ?



Contrail designer
User currently offlineTristarSteve From Sweden, joined Nov 2005, 3976 posts, RR: 34
Reply 10, posted (5 years 4 months 3 weeks 6 days 17 hours ago) and read 3764 times:

There has been a lot of criticism of the BEA in how long it took to produce this report.
I would just like to say that I think the report is excellent and worth waiting for.
BEA draw no conclusions at all, but have laid out all the facts that they have.
It will be interseting to see what conclusions they draw from the AOA readouts.

Pihero, thanks for your comments. As a mere Airbus technician I must admit my knowledge of how the various lwas interact is hazy, mainly because in 20 years on the A320, I have only once seen a Direct Law on the post flight report.

And by the way, you cannot turn off the AOA sensors, but they can fail. They are very sensitive moving devices that could easily be disabled by sticky tape, or being knocked.


User currently offlinePihero From France, joined Jan 2005, 4388 posts, RR: 76
Reply 11, posted (5 years 4 months 3 weeks 6 days 15 hours ago) and read 3584 times:
Support Airliners.net - become a First Class Member!



Quoting TristarSteve (Reply 10):
you cannot turn off the AOA sensors, but they can fail. They are very sensitive moving devices that could easily be disabled by sticky tape, or being knocked.

That's probably where the warning about protecting the "sensors" during a painting job came from, while we were all - including me - assuming that the "sensors" were the static ports.
I agree with you, Steve, about the quality of the prelim report. The BEA just dealt with facts, without any assumption as to the whys of the events.
But here on A.net, we could go a bit further into unknowns :
First the facts, as they appear in the report :

  • They performed the planned (?) low speed test, involving an entry - and an immediate exit from - into alpha floor.
  • The altitude for the test is lower than the recommended FL120-140 by all documents
  • In this exercise, they expected to see successively Vls, Vmin, alpha prot and alpha floor, marked by an automatic trigger of TOGA thrust. They even talked about the ATHR handling at that point ( "push, disengage, reengage" )
  • They never saw that succession of events... Instead, they had a "stall warning" (we can safely assume that it was the "STALL...STALL" call-out and not the "SPEED...SPEED" one would get in normal law.
  • They selected a manual TOGA thrust and then,
  • They experienced some difficulty with the flight path control, ending in a plunge into the sea.

In the report there are two major issues that the investigators have not commented upon but just mentioned as "in passing" :

  • The frozen -"frozen" as in "not moving due to a blockage" of any sort - AoA sensors, probably feeding the FACs erroneous data critical to the speed limits info displayed and the computation of the AoA protections.
    As the print-out shows the AoA1 and 2 unmoving parameters, one can also assume that the stall warnings came from the - unrecorded - stby AoA.
  • The stabilizer (THS) was automatically - and correctly, in view of the deceleration - trimmed to max (?) nose-up and stayed at that value for the remainder of the flight.
    On this subject, in "Normal Law", the auto-trim stops when the airplane enters an alpha-prot situation...and in "Direct Pitch Law" it is unavailable, the crew is reminded by a message urging them to use the manual trim wheel.
    They did enter the "Direct Pitch " mode because of the landing gear been extended... and yet, the stabiliser still showed no movement : jammed or un-utilised ?

That un-moving stabilizer, stuck at max nose-up (for want of another word) couldn't be overcome by the full nose-down demand by the PF, especially when one considers the not-so-light pitch-up moment that the engines at TOGA thrust would induce.. In other words, the sidestick authority in pitch wasn't enough to check the increasing pitch attitude of the airplane, leading to the two subsequent successive stalls.
I have left out all the reversions that could have occurred during that maneuver, because in my opinion, they all lead to a further complication of the crew situational awareness.
Because in the end, I believe that they entered a totally unexpected situation they were not either prepared or trained for, with major handling difficulties and, in my opinion, some disorientation...
A situation that was unrecoverable so close to the sea.



Contrail designer
User currently offlineTristarSteve From Sweden, joined Nov 2005, 3976 posts, RR: 34
Reply 12, posted (5 years 4 months 3 weeks 6 days 14 hours ago) and read 3520 times:



Quoting Pihero (Reply 11):
The frozen -"frozen" as in "not moving due to a blockage" of any sort - AoA sensors

The AOA sensors on the A320 are little "wings". They cannot be blocked, the air flows over them and they show the air flow relative to the fuselage. They could have been held in place by tape.
Many years ago AOA sensors had holes in them. They looked like a spike with two rows of holes and they rotated to keep the air pressure the same through each row of holes.

I am always concerned by the A320 sensors. The lower one on the left is very close to the jetty when it is on L1 door, but I have never had one damaged.


User currently onlineSXI899 From Netherlands, joined Jan 2008, 267 posts, RR: 0
Reply 13, posted (5 years 4 months 3 weeks 6 days 13 hours ago) and read 3411 times:

It certainly seems like the crew erred in performing that test at that altitude. The last recorded data is reported as "14° nose down, 15° right bank, speed 263kts, and altitude of 340ft", which to me would indicate that had there been more altitude to work with, they might have been able to recover the aircraft.

Quoting Pihero (Reply 11):
That un-moving stabilizer, stuck at max nose-up (for want of another word) couldn't be overcome by the full nose-down demand by the PF, especially when one considers the not-so-light pitch-up moment that the engines at TOGA thrust would induce.. In other words, the sidestick authority in pitch wasn't enough to check the increasing pitch attitude of the airplane, leading to the two subsequent successive stalls.
I have left out all the reversions that could have occurred during that maneuver, because in my opinion, they all lead to a further complication of the crew situational awareness.

What I'm having trouble understanding, and maybe someone can clarify it for me, when the thrust levers were moved to the TOGA position, the pitch angle was 18.6° and decreased over the next 10 seconds to 11°, and 5 seconds after that was 7°.
At that point the report states that the longitudinal input from the captain (which was at the forward stop) was cancelled, and the pitch began to increase again (reaching 57° in 20 seconds) despite the captain again making a longitudinal input to the forward stop.
If the pitch decreased in this inital period, wouldn't that indicate that there was still enough sidestick authority to overcome the pitch-up moment from the engines? Or would it just take that long for any pitch-up moment to take effect?

Quoting Pihero (Reply 11):
Because in the end, I believe that they entered a totally unexpected situation they were not either prepared or trained for, with major handling difficulties and, in my opinion, some disorientation...
A situation that was unrecoverable so close to the sea.

Agreed. I also suspect that the crew may not have adequately understood the risks involved in the manoeuvre, perhaps due to having performed it on pervious occasions without any difficulty.

All the more reason to have an agreed test schedule to fly, an indepth briefing to set out responsibilities, and above all, to adhere to what is agreed upon.

Regards,
Yorden



Any Type, Any Time, Anywhere
User currently offlinePihero From France, joined Jan 2005, 4388 posts, RR: 76
Reply 14, posted (5 years 4 months 3 weeks 6 days 11 hours ago) and read 3295 times:
Support Airliners.net - become a First Class Member!



Quoting TristarSteve (Reply 12):
The lower one on the left is very close to the jetty when it is on L1 door, but I have never had one damaged.

That is the St-By AoA vane... in all appearances the only one working.



Contrail designer
Top Of Page
Forum Index

This topic is archived and can not be replied to any more.

Printer friendly format

Similar topics:More similar topics...
Etihad A340 Engine-test Crash Report Out posted Thu Dec 11 2008 05:48:51 by Bells
Jet Blue A320 At ISP! posted Fri Oct 24 2008 08:00:29 by SXDFC
JetBlue A320 At DAB posted Sat Oct 11 2008 18:32:03 by Unmlobo
Sata (S4) A320 At JER 09-SEP-2008? posted Wed Sep 10 2008 13:56:08 by HT
IB A320 At BA Hangar LHR (16th June) posted Mon Jun 16 2008 18:28:02 by GBOAF
AF A320 At St. Maarten posted Wed Jun 11 2008 14:27:05 by Lufthansi
FCA A320 At EXT? posted Sun May 18 2008 14:57:10 by RichM
QR A320 At LAX? posted Thu May 1 2008 20:39:52 by KU747
AF A320 At SXM? posted Fri Jan 11 2008 17:36:06 by AK
Untitled JetBlue A320 At Kbdl - German Reg? posted Mon Sep 24 2007 20:31:14 by Golftango