Airbus crash/training flight
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
How come the AoA sensor seals let in water?
- The absence of consistency in the rinsing task in the airplane cleaning procedure, and in particular the absence of protection of the AOA sensors, during rinsing with water of the airplane three days before the flight. This led to the blockage of the AOA sensors through freezing of the water that was able to penetrate into the inside of the sensor bodies
Join Date: Jan 2005
Location: France
Posts: 2,315
Likes: 0
Received 0 Likes
on
0 Posts
Nigel,
"My" AoA sensors are little vanes, connected to a thoroughly sealed rotating sensor, such as a pot or a synchro.
You could have washed them all day with a hose, no water would have gotten in...
The A320 ones are obviously different, if water can remain in them for three days.
Hence my question.
CJ
"My" AoA sensors are little vanes, connected to a thoroughly sealed rotating sensor, such as a pot or a synchro.
You could have washed them all day with a hose, no water would have gotten in...
The A320 ones are obviously different, if water can remain in them for three days.
Hence my question.
CJ
Trash du Blanc
Join Date: Mar 2001
Location: KBHM
Posts: 1,185
Likes: 0
Received 0 Likes
on
0 Posts
Can you really think it correct for airline crews to go out and test the equivalent of the "stalling speed" without having a good idea of what that speed should be, depsite it being presented in a table in the schedule, and still flying (IIRC) some 15K+ below it wondering when things might start to happen
I do these tests on a regular basis. They are done at a minimum of 12,000 feet, and we calculate the shaker speed beforehand. We begin the deceleration about 40 knots above target speed, and we cannot trim nose up below about 25 knots above target speed. That way there is a big pitch-down moment as soon as we relax a little.....
Join Date: Jun 2009
Location: europe
Posts: 6
Likes: 0
Received 0 Likes
on
0 Posts
Dear Christiaan,
rinsing implies to me a soft shower with water. During flight the rain does also not hit the case/sealing in 90° angle due to the position of the sensor on the fuselage.
So how about the sealing, if high water pressure is used from close distance? It might be possible then.
What I ask myself is, if the sensor doesn't have a heating? I found one here
http://www.goodrich.com/portal/goodr...%20Systems.pdf
from Goodrich, which has a standard case heating, but optional vane heating. Is it similar to the one used by Airbus 320?
Another question is: if the water is really inside the case, then it can hardly evaporate. Did the crew experience a nightmare, that could have occured any time later with passengers on board?
And for the sake of terminology: I guess the crew didn't see the maneuver as a "test flight", but as a demonstration/training flight and such didn't expect anything to go wrong. That is also highlighted in the BEA report.
So we have to ask, what would have happened if the AoA sensors have given correct values? Most probably nothing special would have happened and the crew still alive.
There is a good summary of the BEA final report on the aviationherald
http://avherald.com/h?article=410c9cec/0015&opt=0
regards twinotter
rinsing implies to me a soft shower with water. During flight the rain does also not hit the case/sealing in 90° angle due to the position of the sensor on the fuselage.
So how about the sealing, if high water pressure is used from close distance? It might be possible then.
What I ask myself is, if the sensor doesn't have a heating? I found one here
http://www.goodrich.com/portal/goodr...%20Systems.pdf
from Goodrich, which has a standard case heating, but optional vane heating. Is it similar to the one used by Airbus 320?
Another question is: if the water is really inside the case, then it can hardly evaporate. Did the crew experience a nightmare, that could have occured any time later with passengers on board?
And for the sake of terminology: I guess the crew didn't see the maneuver as a "test flight", but as a demonstration/training flight and such didn't expect anything to go wrong. That is also highlighted in the BEA report.
So we have to ask, what would have happened if the AoA sensors have given correct values? Most probably nothing special would have happened and the crew still alive.
There is a good summary of the BEA final report on the aviationherald
http://avherald.com/h?article=410c9cec/0015&opt=0
regards twinotter
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
CJ...
I don't want to get into a technical debate over matters I am no expert in. The report goes into very fine detail, and is clear:
NoD
I don't want to get into a technical debate over matters I am no expert in. The report goes into very fine detail, and is clear:
- There is a procedures to "wash" the Aircraft - it was not followed
- It involves protecting the AoA vanes - they were not
- Excess pressure was used, for too long, at the wrong angle
- Your vanes might be fine, but can they really stand up to a pressure wash on them as close quarters?
- The report is clear that water cannot, in normal use, get in.
- In fact, the seals etc. were probably what held the water in there for 3 days
NoD
Join Date: Jan 2005
Location: France
Posts: 2,315
Likes: 0
Received 0 Likes
on
0 Posts
"It involves protecting the AoA vanes - they were not"
That's precisely my question.... are the AoA sensors on the A320 in question the classic AoA vanes, or something "new" and different, with pressure ports located in the appropiate locations?
And as ever, this was just a contributory factor.
CJ
Join Date: Jun 2009
Location: europe
Posts: 6
Likes: 0
Received 0 Likes
on
0 Posts
@nigel o d "normal aeroplane"
No, it doesn't make it a normal aeroplane!
The report highlights the fact, that it was more a training/demonstration flight than a testflight. Thus the crew did not expect anything going wrong and waited for the protection to come (with wrongly calculated speeds derived from frozen AoA sensors).
Moreover it changed from normal law to direct law, switching auto pitch control off.
The crew did not have a normal aeroplane, but an incorrect functioning aeroplane!
What would have happened, if the AoA sensors worked as they should? I guess nothing special, despite low altitude and all the other factors.
The report highlights the fact, that it was more a training/demonstration flight than a testflight. Thus the crew did not expect anything going wrong and waited for the protection to come (with wrongly calculated speeds derived from frozen AoA sensors).
Moreover it changed from normal law to direct law, switching auto pitch control off.
The crew did not have a normal aeroplane, but an incorrect functioning aeroplane!
What would have happened, if the AoA sensors worked as they should? I guess nothing special, despite low altitude and all the other factors.
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
CJ...
As an A320 Capt, I should probably know what the sensors are... but not too sure
They are proper vanes, 3 in all, and vulnerable to jetty's. All 3 in the nose area.
The report has more detail on them than I suspect any Tech Manual we, or even the engineers, have. I also have heard of no significant issues with them previously... and this accident seems a rather "perfect storm" of factors that combined to a tragic outcome.
NoD
As an A320 Capt, I should probably know what the sensors are... but not too sure
They are proper vanes, 3 in all, and vulnerable to jetty's. All 3 in the nose area.The report has more detail on them than I suspect any Tech Manual we, or even the engineers, have. I also have heard of no significant issues with them previously... and this accident seems a rather "perfect storm" of factors that combined to a tragic outcome.
NoD
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
twinotter...
An "organisational failure" - the test they performed was for Test Pilots/purposes only. Hardly the aircraft's fault they embarked with this basic misunderstanding.
Pretty fundamental mistake to make during Test/Check flying. Why on earth do you perform the test if you do not expect it to go wrong?
Sorry? When did the crew ever calculate the speeds? (A calculation required to be made by the schedule).
As per design... and exactly as would be expected if case of problems.
The purpose of the flight was to determine if they had a 'normal aeroplane' and the basic rule of Test/Check flying is to assume you have an 'incorrect functioning aeroplane'.
As you say, nothing special. But if you are willing to stake your, and other people's, lives on the fact the system works, there seems little point in checking it in the first place.
The fundamental point you seem to miss is they performed a test that was designed to check if the AOA system was working correctly. Airbus specified (for Test Pilots) a min altitude to perform this test, in case the system was not working correctly. Surely if you are to test the system, you assume it will not be working? Or at least, not rely on it working to save your life...
Regrettably, all this accident proved was that the Airbus Test Schedule was correctly written, in purpose, Crew Qualifications, and Min Alt? To go back to the 737 equivalent issues, whilst there were some issues, the crew qualifications and Min Alts were followed, and the outcome was thus an "incident", not "accident".
NoD
The report highlights the fact, that it was more a training/demonstration flight than a testflight
Thus the crew did not expect anything going wrong and waited for the protection to come
(with wrongly calculated speeds)
Moreover it changed from normal law to direct law, switching auto pitch control off.
The crew did not have a normal aeroplane, but an incorrect functioning aeroplane!
What would have happened, if the AoA sensors worked as they should? I guess nothing special, despite low altitude and all the other factors.
The fundamental point you seem to miss is they performed a test that was designed to check if the AOA system was working correctly. Airbus specified (for Test Pilots) a min altitude to perform this test, in case the system was not working correctly. Surely if you are to test the system, you assume it will not be working? Or at least, not rely on it working to save your life...
Regrettably, all this accident proved was that the Airbus Test Schedule was correctly written, in purpose, Crew Qualifications, and Min Alt? To go back to the 737 equivalent issues, whilst there were some issues, the crew qualifications and Min Alts were followed, and the outcome was thus an "incident", not "accident".
NoD
Join Date: Apr 2002
Location: FUBAR
Posts: 3,348
Likes: 0
Received 0 Likes
on
0 Posts
Sorry to be judgemental, particularly towards dead brethern, but what the hell were they thinking to envisage they could just "slip" something like this in (almost as an aside ) on base leg ? ? ?
A touche bloody optimistic, even on something less complicated than a Scarebus IMHO.
A touche bloody optimistic, even on something less complicated than a Scarebus IMHO.
Join Date: Apr 2008
Location: 42 34N, 3 02E
Posts: 83
Likes: 0
Received 0 Likes
on
0 Posts
Look chaps, I know that A320 has auto heat on the AoA probes. Why didn't the ice melt? Why did water enter the AoA seals? My car's got a better (heated) system on the windsceen washers for goodness sake! And I'd take it back if it leaked in the car wash.
Join Date: May 2010
Location: not here
Posts: 34
Likes: 0
Received 0 Likes
on
0 Posts
It is hard to conceive the image of an A320 airframe with unprotected sensors pressure-washed by a kid blithely unaware he was acting in direct contravention to the manufacturer's instructions.
But it was done, and water entered the sensors.
Someone mentioned failure of the sensors' ability to withstand forced input of water. It was suggested the failure brings into question the ability of the A320 sensor seals to repel lashings of rain in a tropical storm.
Perhaps you are not familiar with the pressure coming out of a gurney.
I have a gurney at home, a Karcher. The high pressure water's delivery can strip paint. I have stripped paint with mine, marveling at the sycthe-like action of water under pressure. One's skin would come right off the feet - if you were dumb enough to point the stream at your unprotected toes.
Sadly, what appears to have happened with this accident is the alignment of a 'perfect storm' of events. That term, coined by Sebastian Junger, has already been used, but allow me to shamelessly repeat it for it's a good one, and very apt.
The wash should have been conducted properly, protecting the seals; it was not. The pilots should have known their target airspeeds for stall, given they were about to demonstrate it for real; they did not. The airplane should never have been placed in that condition; but it was.
Much as I prefer Boeings, which I've flown for 18 years, there appears to be nothing wrong with the design of the A320 sensor seals. Nor is there a fault with the stabilizer design. I don't like it, but that doesn't make it a faulty design.
If the driver of a road vehicle spun the wheel hard over at 100kph, the vehicle would most likely be out of control and quite possibly crash. It would not be the fault of the manufacturer of the vehicle anymore than it can be said Airbus are at fault for the actions of the pilots of this airplane, who took the airframe outside its design parameters.
So who is at fault? And who could have done this test? Who could have known the limitations of the stall and the A320 systems dealing with stall such that the pilot would have the requisite skills to recover?
Sorry, but the clear answer to that is any properly trained line pilot. Stall recovery is not the sole province of the test pilot. We are all trained to recover from a stall, purposely entered or otherwise.
The actions of the pilot in this accident appear to have been incorrect.
For that, you cannot blame the washer of the airplane; you cannot blame the manufacturer of the airplane; you cannot blame the seals of the sensors, and you cannot blame the design of the stabilizer, even though I agree it could be better.
One of my first jobs was flying a single engined piston for a pastoral company in central Queensland. We did a lot of things with that airplane, none of which were cattle mustering. I was not authorized for cattle mustering. Though I had an aerobatic rating, I had not been trained for the mustering of cattle.
Quite sensibly then, the Area Manager told me, a young pilot, not to allow the company's many Station Managers to pressure me into mustering, or to let them take the controls and 'have a look around' if we were on a flight to check their individual stations' distant bores.
When I left that job, my first and very memorable job, for bigger and better things, a young 21 year old pilot took over. In the first week he was there he succumbed to pressure from one of the Station Managers. The smoking hole was found the following day. That was over 22 years ago, and yet I am struck by the similarities with this unnecessary A320 accident.
Aviation is populated by many latent dangers. Many of us would like to ignore this reality. Doing so comes at a price. Recovery from a non-normal event of attitude requires a deft hand when encountered. Wilfully entering a non-normal flight regime without being fully prepared is, to my mind and from my experience, naivete or wilful negligence.
It is never both.
I've no time for sentimental notions of a fictional 'band of brothers'. The pilots of this A320 were not 21 years old. They knew better. If they didn't know better, they should not have been charged with the responsibility they had. It's as simple as that. Fallen comrades is bollocks. So is blaming Airbus. Damn the French, but not for this.
HTFU.
But it was done, and water entered the sensors.
Someone mentioned failure of the sensors' ability to withstand forced input of water. It was suggested the failure brings into question the ability of the A320 sensor seals to repel lashings of rain in a tropical storm.
Perhaps you are not familiar with the pressure coming out of a gurney.
I have a gurney at home, a Karcher. The high pressure water's delivery can strip paint. I have stripped paint with mine, marveling at the sycthe-like action of water under pressure. One's skin would come right off the feet - if you were dumb enough to point the stream at your unprotected toes.
Sadly, what appears to have happened with this accident is the alignment of a 'perfect storm' of events. That term, coined by Sebastian Junger, has already been used, but allow me to shamelessly repeat it for it's a good one, and very apt.
The wash should have been conducted properly, protecting the seals; it was not. The pilots should have known their target airspeeds for stall, given they were about to demonstrate it for real; they did not. The airplane should never have been placed in that condition; but it was.
Much as I prefer Boeings, which I've flown for 18 years, there appears to be nothing wrong with the design of the A320 sensor seals. Nor is there a fault with the stabilizer design. I don't like it, but that doesn't make it a faulty design.
If the driver of a road vehicle spun the wheel hard over at 100kph, the vehicle would most likely be out of control and quite possibly crash. It would not be the fault of the manufacturer of the vehicle anymore than it can be said Airbus are at fault for the actions of the pilots of this airplane, who took the airframe outside its design parameters.
So who is at fault? And who could have done this test? Who could have known the limitations of the stall and the A320 systems dealing with stall such that the pilot would have the requisite skills to recover?
Sorry, but the clear answer to that is any properly trained line pilot. Stall recovery is not the sole province of the test pilot. We are all trained to recover from a stall, purposely entered or otherwise.
The actions of the pilot in this accident appear to have been incorrect.
For that, you cannot blame the washer of the airplane; you cannot blame the manufacturer of the airplane; you cannot blame the seals of the sensors, and you cannot blame the design of the stabilizer, even though I agree it could be better.
One of my first jobs was flying a single engined piston for a pastoral company in central Queensland. We did a lot of things with that airplane, none of which were cattle mustering. I was not authorized for cattle mustering. Though I had an aerobatic rating, I had not been trained for the mustering of cattle.
Quite sensibly then, the Area Manager told me, a young pilot, not to allow the company's many Station Managers to pressure me into mustering, or to let them take the controls and 'have a look around' if we were on a flight to check their individual stations' distant bores.
When I left that job, my first and very memorable job, for bigger and better things, a young 21 year old pilot took over. In the first week he was there he succumbed to pressure from one of the Station Managers. The smoking hole was found the following day. That was over 22 years ago, and yet I am struck by the similarities with this unnecessary A320 accident.
Aviation is populated by many latent dangers. Many of us would like to ignore this reality. Doing so comes at a price. Recovery from a non-normal event of attitude requires a deft hand when encountered. Wilfully entering a non-normal flight regime without being fully prepared is, to my mind and from my experience, naivete or wilful negligence.
It is never both.
I've no time for sentimental notions of a fictional 'band of brothers'. The pilots of this A320 were not 21 years old. They knew better. If they didn't know better, they should not have been charged with the responsibility they had. It's as simple as that. Fallen comrades is bollocks. So is blaming Airbus. Damn the French, but not for this.
HTFU.
Last edited by big white bird; 17th Sep 2010 at 04:22. Reason: grammar
Join Date: Aug 2002
Location: Strategic hamlet
Posts: 115
Likes: 0
Received 0 Likes
on
0 Posts
My car's got a better (heated) system on the windsceen washers for goodness sake! And I'd take it back if it leaked in the car wash.
An even higher pressure version called the waterjet is used to cut through metal.
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
WP
If you really want to know, rather just make uninformed anti-Airbus comments, read the report. That is, after all, the subject under discussion 
NoD
Why didn't the ice melt? Why did water enter the AoA seals? My car's got a better (heated) system on the windsceen washers for goodness sake! And I'd take it back if it leaked in the car wash.
NoD
Join Date: Apr 2008
Location: 42 34N, 3 02E
Posts: 83
Likes: 0
Received 0 Likes
on
0 Posts
@NOD
I'm actually an airbus fan! See my location. Nothing in the report explains how water penetrated 3 seperate sensors, and no-one's yet commented or explained why the sensor anti ice did not deal with the ice build up. Until I (and others here) understand this, I think it's perfectly legitimate to ask the community these questions.
I'm actually an airbus fan! See my location. Nothing in the report explains how water penetrated 3 seperate sensors, and no-one's yet commented or explained why the sensor anti ice did not deal with the ice build up. Until I (and others here) understand this, I think it's perfectly legitimate to ask the community these questions.
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
WP...
I think report does discuss this in some detail (p37-47, 87-90)... They only know water did penetrate 2 of the 3 sensors. It penetrated because it was pressure washed in contravention of the instructions. The water penetrated right into the sensors, which is not the area heated. The sensors passed the certification requirements v rain, albeit there were some comments.
Bottom line however, is the Airbus does not need these sensors to fly. Ultimately they are stall warning devices. The accident stemmed from these sensors being iced up, the crew deciding to (or at least close to) stall the aeroplane, doing so at an unsuitable height , the crew being unable to recover from the stall.
NoD
I think report does discuss this in some detail (p37-47, 87-90)... They only know water did penetrate 2 of the 3 sensors. It penetrated because it was pressure washed in contravention of the instructions. The water penetrated right into the sensors, which is not the area heated. The sensors passed the certification requirements v rain, albeit there were some comments.
Bottom line however, is the Airbus does not need these sensors to fly. Ultimately they are stall warning devices. The accident stemmed from these sensors being iced up, the crew deciding to (or at least close to) stall the aeroplane, doing so at an unsuitable height , the crew being unable to recover from the stall.
NoD
Join Date: Aug 2000
Location: Farnham, UK
Posts: 323
Likes: 0
Received 0 Likes
on
0 Posts
WP
It is all in the report.
The third sensor did not fail - It explains that the No3 Sensor was working. It gave the stall warning.
The report gives data abour two tests. One had high pressure water sprayed directly at the sensor - water was found inside the sensor. Another test had water sprayed in wash pattern - i.e. to and fro across the sensor. This sensor was found to have no water inside.
As far as anti-ice. The AoA blade is heated so that the sensor can resisit accumalted ice. The internal mechanisms are not heated. The tests also showed that once ice forms and then melts, this residual water still impeded the sensor functionality.
Of note is a series of pictures showing the PFD speed displays for the flight with and without the AoA frozen. The pics show a very large range between Vls & Alpa prot. It just looks so odd and you would have hoped that someone would have noticed. It they had calculated the speeds as per the test schedule then it would have been very obvious of the difference between FAC & FMGC calculated speeds.
It is all in the report.
The third sensor did not fail - It explains that the No3 Sensor was working. It gave the stall warning.
The report gives data abour two tests. One had high pressure water sprayed directly at the sensor - water was found inside the sensor. Another test had water sprayed in wash pattern - i.e. to and fro across the sensor. This sensor was found to have no water inside.
As far as anti-ice. The AoA blade is heated so that the sensor can resisit accumalted ice. The internal mechanisms are not heated. The tests also showed that once ice forms and then melts, this residual water still impeded the sensor functionality.
Of note is a series of pictures showing the PFD speed displays for the flight with and without the AoA frozen. The pics show a very large range between Vls & Alpa prot. It just looks so odd and you would have hoped that someone would have noticed. It they had calculated the speeds as per the test schedule then it would have been very obvious of the difference between FAC & FMGC calculated speeds.
Last edited by Thunderbug; 17th Sep 2010 at 08:10.