Jetstar malfunction link to Air France Atlantic crash

[oz_mark]

To be pedantic, that must be problems with certain manufacturers. A pitot tube works by measuring the pressure difference between moving air and the atmosphere (still air). There can only be one type of Pitot Tube, otherwise it isn't a pitot tube. [/pedantic point]

Pitot tubes on aircraft have to deal with icing, water drainage and so on. So things can certainly happen to them in flight

 

[Flying Fox]

Can someone confirm whether they retreived the Flight Data Recorder and coughpit Voice Recorder from the AF flight?

My last memory has it that they were still lost.

 

[FL360]

Air France Flight 447 - Wikipedia, the free encyclopedia

Yes, the black box is still not found, it is a very big and deep ocean, indeed...

Can someone confirm whether they retreived the Flight Data Recorder and coughpit Voice Recorder from the AF flight?

My last memory has it that they were still lost.

 

[Slats7]

nonews have finally updated thier photo of a jetstar plane (no more B717) - pity they show an A320 when it's an A330 affected by this. The caption implies it's an A330 as per the article.

Jetstar malfunction link to Air France Atlantic crash | Travel News | News.com.au

They enjoy showing a computer generated image of a QF787 for QF.

 

[Flying Fox]

So given that the CVR & FDR are still missing, how do they know that the pitot tubes are the cause of error?

 

[BAM1748]

They don't for sure, but the plane was automatically feeding maintenance information back to the airline maintenance facility in France as it flew along. Plus the pilot radioed in he was having trouble with his air speed censors just prior to the crash.

There was also the fact AF started changing all its pitot tubes within two weeks of the crash and the day after the union raised it with the airline and I think Airbus issued an alert about it as well following the crash.

Hence one of the suggestions following this accident was feeding the CVR direct back to a land based storage system (but hugely expensive at this stage to contemplate I would think)

Cheers,

Matt

 

[medhead]

You know this is all this discussion is suggesting computer error more and more, IMO. Sure they replaced pitot tubes from one manufacturer that might have been more prone to icing or blocking or some other machanical problem. But that is just treating the symptom. A pitot tube is not a sensor, it is an device that can be used to make a measurement. It's like a ruler. The sensor is the thing that is stuck on the end of the pitot tube to actually make the measurement.

If icing or some blocking of the pitot tube is to blame, that tells me that whoever programmed the sensor to measure the pressure difference and calculate the airspeed etc. does not understand the modes of failure for a pitot tube. The fact that this might have happened and that they are still taking in the numbers as real, suggests they (the computer or the programmer) don't know when to ignore the readings.

It is a sad indictment of the standard of education in the world today - total lack of critical thinking. They went to the moon using basic physics, critical thinking and a roll of duct tape. Today, they don't know to tell the computer to ignore the pitot tube when it says the speed is mach 3.

 

[markis10]

. It is a sad indictment of the standard of education in the world today - total lack of critical thinking. They went to the moon using basic physics, critical thinking and a roll of duct tape. Today, they don't know to tell the computer to ignore the pitot tube when it says the speed is mach 3.

Your missing a key bit of knowledge here, you cannot ignore an airspeed indication when you dont have an alternative reading available and you are flying an airplane that has a supercritical wing, that is one that operates in a a narrow range of airpseeds to provide lift at cruising altitude. Some aircraft operate with a spread of just 20-40 knots between having lift and stalling at altitude, and the pitot system is actually a pitot-static system with mutliple sensors including more than one pitot tube and one static port on most aircraft. I think its simplifying things a bit too much to blame a pitot tube, as with most incidents its mutliple factors coming into play that produce the final result.

 

[medhead]

I think its simplifying things a bit too much to blame a pitot tube, as with most incidents its mutliple factors coming into play that produce the final result.

Yes, as do I. But many here seem to be ready to blame the pitot tube, and point to the replacement of pitot tubes from one manufacturers. I'm trying to say that it can't just be the tube but the measurement made with that dvice and the interpretation of those results.

Your missing a key bit of knowledge here, you cannot ignore an airspeed indication when you dont have an alternative reading available and you are flying an airplane that has a supercritical wing, that is one that operates in a a narrow range of airpseeds to provide lift at cruising altitude. Some aircraft operate with a spread of just 20-40 knots between having lift and stalling at altitude, and the pitot system is actually a pitot-static system with mutliple sensors including more than one pitot tube and one static port on most aircraft.

No, I'm not missing knowledge at all. You must ignore an airspeed indication that is obviously wrong, that is known to be wrong and that is wrong because of a mechanical issue with a mechanical device. This is the point, it is absolutely stupid to try to give information that is wrong and should be known to be wrong, when the wrong information can lead to accidents for exactly the reasons you have listed. There are many occassions when knowing that you don't have certain information (critical or not) is better than being told the wrong information.

e.g. your being told that you are at high alititude and at low speed, lower the nose increase speed - BANG whoops you were actually 500ft above the sea.

Or BANG whoops you were actually rather fast and lowering the nose and increasing engine input means you exceed the design speed and your wings just ripped off.

 

[markis10]

No, I'm not missing knowledge at all......... Sure they replaced pitot tubes from one manufacturer that might have been more prone to icing or blocking or some other machanical problem. But that is just treating the symptom. A pitot tube is not a sensor, it is an device that can be used to make a measurement. It's like a ruler. The sensor is the thing that is stuck on the end of the pitot tube to actually make the measurement..

With all due rspect you are missing some knowledge by your statement comparing Pitot Static systems with rulers, rulers still work if they are iced up or flexed in shape, pitot systems require a defined shape of a specific nature and ram air pressure to work. I share your concern about poor computer work though, if we are going to design an aircraft that can only work by computer than we better make sure the code is 100% correct.

To be pedantic, that must be problems with certain manufacturers. A pitot tube works by measuring the pressure difference between moving air and the atmosphere (still air). There can only be one type of Pitot Tube, otherwise it isn't a pitot tube. [/pedantic point]..

This Air France explanation of events helps set some of the facts straight that have been posted here, not all oils are oils so to speak and not all pitot tubes are pitot tubes when they are installed in an environment they are not designed for, as they directly react with the fuselage airflow in the vicinity of their location they are installed, so a tube designed for a cessna 172 wont work well on an A330 where there is flight in known icing conditions but both are still pitot tubes:

Chronology

1988 – February 2001
At the outset, all Air France and Air Inter short- and medium-haul A320s were equipped with Pitot probes manufactured by BADIN CROUZET, which subsequently became SEXTANT AVIONICS and then THALES AVIONICS.
In 1999, due to the risk of water ingress and inconsistent speed data during heavy rain and/or icing conditions, Air France decided to replace these probes by THALES AA probes in compliance with a recommendation from Airbus. Even before the Airbus recommendation became mandatory on 4 December 2001, all the Air France A320s had been equipped with THALES AA probes on 8 February 2001.
August 2001
Following the fluctuations and/or inconsistency in aerodynamic speed indications on the long-haul A330 and A340 reported by some airlines, the French Civil Aviation Authorities published an Airworthiness Directive (AD) per type of aircraft to impose the replacement of the ROSEMOUNT P/N 0851GR Pitot probes, either with the GOODRICH P/N 0851HL probes, or by the SEXTANT (THALES) P/N C16195-AA probes. This operation had to be completed by 31 December 2003.
The authorities did in fact attribute these incidents to the presence of ice crystals and/or quantities of water which exceeded the specifications of the P/N 0851GR Pitot probes, manufactured by ROSEMOUNT, which was acquired in 1993 by GOODRICH.
In accordance with the Airworthiness Directive, the SEXTANT (THALES) P/N C16195-AA (= THALES AA) probes, already fitted on the A320s, were installed on Air France’s A340 fleet.
As from December 2001, Air France received its first A330s, already equipped with THALES AA probes so that this was the only type of probe on its Airbus fleet of aircraft.
No inconsistency in speed indications was reported on the Air France A330s and A340s until May 2008.
September 2007
Airbus issued service bulletins which recommended, on an optional basis and devoid of any context affecting aircraft airworthiness, the replacement of the THALES AA probes fitted onto all the A320/A330/A340s with a new THALES P/N C16195-BA model (= THALES BA).
This model was said to improve the performance of the probe by limiting water ingress during heavy rain and reducing the risk of probe icing.
After examination, Air France’s technical teams decided to modify the Airbus A320 fleet which had experienced incidents involving inconsistency in speed indications at low altitude during heavy rain. They decided to replace the probes on the A330/A340s with the new models only when a failure occurred, as these aircraft had experienced no incidents involving inconsistencies in speed data.
May – August 2008
The first incidents involving inconsistent speed data were reported on Air France A340s, although no such incident had been reported beforehand.
Air France immediately questioned Airbus on the cause of these incidents and on the corrective measures to be taken.
September and October 2008
A great deal of discussion took place with Airbus’ technical teams. Two new incidents were then reported by Air France.

Airbus replied that:

• the presumed cause of these incidents was the formation of ice crystals which iced up the airspeed probes;
• the new THALES BA model had not been designed to address the problem of probe icing and could not therefore provide a significant improvement;
• the probes in place were perfectly compliant with and even exceeded the regulatory requirements in terms of airworthiness and flight safety.

November 2008
Subsequent to different follow-up messages from Air France’s technical departments, Airbus amended its service bulletins of September 2007 in a memo dated 12 November 2008, cancelling the recommendation of September 2007. The revised version no longer recommended the installation of THALES BA probes to reduce the risk of icing.
A meeting took place between the technical managements of Air France and Airbus, where the incidents involving inconsistent speed data were discussed at length. Air France asked for a rapid solution. Airbus once again confirmed that these incidents were caused by the icing over of the probes, that the THALES BA probes did not address the problem and that the probes in place were compliant with the airworthiness and flight safety requirements.
Air France asked if it were possible to replace them with GOODRICH probes. Airbus noted the request from Air France and indicated their desire to verify the feasibility.
February 2009
Faced with Air France’s insistence on finding a solution, wind tunnel tests were conducted by Thales and Airbus on the behaviour of the THALES BA tubes.
March 2009
At the end of March 2009, two new operating incidents were reported, including the first on an A330. This brought the total number of incidents to nine, eight on the A340 and one on the A330. Once again Air France appealed on several occasions to Airbus, who replied by confirming the presumption of probe icing, referring the airline to a maintenance procedure and to checks on the probes.
April 2009
In a letter dated 15 April 2009, Airbus changed its position: the THALES BA probe was not designed to solve the problem of probe icing but tests conducted by Thales revealed a significant improvement in its performance compared with the AA model. Given the limitations of the wind tunnel tests, Airbus suggested a trial on Air France planes to check whether the improvement would be confirmed in a real-life situation.
Without waiting for these tests, Air France decided immediately to replace the THALES AA probes by THALES BA probes on all its Airbus A330s and A340s.
An internal document launching the replacement procedure was issued by Air France on 27 April 2009. The probes were ordered from Thales. The start of the modification process was scheduled as soon as the parts were delivered, on the basis of several aircraft per week as from 1 June 2009.
May – June 2009
In May 2009, Air France asked THALES to speed up the delivery schedule for its probes. They were delivered starting on 26 May 2009. By 12 June, all the Airbus A320s, A330s and A340s operated by Air France were equipped with THALES BA probes.
July – August 2009
A study carried out by Airbus and presented to EASA prompted this agency to make it mandatory as from 7 September 2009 (Airworthiness Directive), as a precautionary measure, to equip the aircraft in all A330 and A340 fleets with at least two GOODRICH PN 0851-HL probes. Consequently, in compliance with the service bulletin issued by Airbus on 30 July 2009, the entire fleet of A330s and A340s of Air France has been equipped with these probes since 7 August 2009.
23 September 2009
EASA (European Aviation Safety Agency) issued an Airworthiness Directive (A/D) applicable on 23 September 2009, asking all operators of Airbus A330/340s fitted with Goodrich 0851HL Pitot probes bearing certain part numbers, to verify them.
Air France wishes to point out that all its aircraft concerned by this
A/D had already been checked, as requested, between 5 and 9 September 2009. This check was carried out on the basis of technical data provided beforehand by Airbus.

It would appear the above has been taken offline at present so I have posted the cached version that appeared here: http://corporate.airfrance.com/en/news/af-447/pitot-probes/index.html

 

[harvyk]

if we are going to design an aircraft that can only work by computer than we better make sure the code is 100% correct.

And there is the problem...

No computer code can be 100% correct. Unless your talking about the simplest of programs there will always be bugs \ faults \ errors which where not expected. I think one of the problem with the Airbuses is that they do try and take the human out of the loop (computer knows best), whilst Boeing have the computer there to aid the pilot (human knows best).

That said I have no problems flying on A330's...

 

[FL360]

Now this thread has become aviation enthusiast corner, I like it…

Call me simple minded.. I am sure there are other ways the system can reference the airspeed ? For example using the GPS true speed readings ? Sure the indicated airspeed decreases as altitude increases but there is an relationship between air pressure, airspeed and true speed.

Sure I think the plane has like three pitot tubes, what will happen if ALL of them are clogged with ice, before de-icing gets rid of them ?

Now since AF447 is the first fatal A330 accident, may be it is suggesting that the testing process for certifying a plane is not detailed enough (did not put the test plane through all conditions it might expect) …. they can’t put the test plane through ALL metrological conditions which the plane might experience, sure they can test the plane near the maximum and minimum temperature and pressure.. but they can’t put the test plane through (or near) ALL kinds of storms it might pass through / go near during its life time….

 

[markis10]

Now this thread has become aviation enthusiast corner, I like it…

Call me simple minded.. I am sure there are other ways the system can reference the airspeed ? For example using the GPS true speed readings ? Sure the indicated airspeed decreases as altitude increases but there is an relationship between air pressure, airspeed and true speed.

GPS's dont provide true airspeed as they have no way of measuring the wind force being applied to the aircraft, they measure ground speed, that is the speed of the aircraft after its affected by a tail or headwind in terms of passage over the ground. You need other instruments to determine the air vector and pressure to ascertain true airspeed/indicated airspeed/ calibrated airspeed or equivalent airspeed.

 

[medhead]

With all due rspect you are missing some knowledge by your statement comparing Pitot Static systems with rulers, rulers still work if they are iced up or flexed in shape, pitot systems require a defined shape of a specific nature and ram air pressure to work.

With respect, I'm not missing anything. A ruler is a mechancal device that allows a person to measure distance. The quality of the measurement is dependant on the quality person making the reading, that is not to imply that it is hard to measure distance.

Same, same with a pitot-static tube, it is a mechancial device that allows the measurement of various air pressures or pressure differences. The quality of those measurements is dependant on the person, or gauges or computer, doing the measurement. That measurement can also be impacted by inherent faults that might occur with a pitot-static tubes, unlike a ruler. So yes, it is harder to get a correct measurement than with a ruler. but otherwise the pitot tube is nothing more than a device for making a measurement, like a ruler.

btw, It was the best analogy I could think of at the time.

Thanks for the history lesson. Very informative.

 

[markis10]

With respect, I'm not missing anything. A ruler is a mechancal device that allows a person to measure distance. The quality of the measurement is dependant on the quality person making the reading, that is not to imply that it is hard to measure distance.

Reminds me of the time I gave Air China an instruction to make a turn at 10 DME, their readback indicated a lack of comprehension and on futher quesitoning they asked me what was DME, I replied "distance measuring equipment" and asked them if they had it, the answer was "yes, we have a ruler :shock:"

Needless to say the clearance was changed to something that made more sense to both parties.

 

[harvyk]

GPS's dont provide true airspeed as they have no way of measuring the wind force being applied to the aircraft, they measure ground speed, that is the speed of the aircraft after its affected by a tail or headwind in terms of passage over the ground. You need other instruments to determine the air vector and pressure to ascertain true airspeed/indicated airspeed/ calibrated airspeed or equivalent airspeed.

Whilst that is true, you could in theory use the speed reading provided by the GPS to "guestimate" your airspeed. Whilst if your in that position you'd probably want to land ASAP, you could in theory keep the plane in the air...

Of course you'd have to know that your IAS wasn't accurate, eg it stating your doing 0 kt's, or your doing 2000 kt's would be a good indication it's at fault. but if it's stating your IAS is say 280kt's and your actual airspeed is 240kt's a pilot may not pick that up until it's too late...

 

[medhead]

This Air France explanation of events helps set some of the facts straight that have been posted here, not all oils are oils so to speak and not all pitot tubes are pitot tubes when they are installed in an environment they are not designed for, as they directly react with the fuselage airflow in the vicinity of their location they are installed, so a tube designed for a cessna 172 wont work well on an A330 where there is flight in known icing conditions but both are still pitot tubes:

I think what you mean is not all pitot tubes are A330 pitot tubes.

 

[FL360]

Yes, when I suggest using GPS to guess approximate airspeed is for the situation when the pitot tube systems reports strange results, such as all 0 kts or 2000 kts.... if that case does happen I think giving the system a second source of airspeed indication is beneficial....

Whilst that is true, you could in theory use the speed reading provided by the GPS to "guestimate" your airspeed. Whilst if your in that position you'd probably want to land ASAP, you could in theory keep the plane in the air...

 

[markis10]

Yes, when I suggest using GPS to guess approximate airspeed is for the situation when the pitot tube systems reports strange results, such as all 0 kts or 2000 kts.... if that case does happen I think giving the system a second source of airspeed indication is beneficial....

The air vector information is derived from the GPS or INS data being compared to the data provided by the pitot/static system and sent for AMDAR use, if the the Pitot/Static system is suspect then you have no way of determining the current air vector affecting the aircraft, so your approximate airspeed could have a margin of error of up to + or -150 knots.

 

[harvyk]

Yes, when I suggest using GPS to guess approximate airspeed is for the situation when the pitot tube systems reports strange results, such as all 0 kts or 2000 kts.... if that case does happen I think giving the system a second source of airspeed indication is beneficial....

The problem with the GPS is that it gives ground speed, not airspeed. The two are quite different speeds.

GPS speed is the aircraft in relation to the ground below (eg how much of the earth has passed by underneath the aircraft in a given time). Airspeed is the aircraft through the air that is supporting it (eg how fast the air is as it passes under the wings). Due to high alt winds, jetstreams and other factors these two are quite different.

If the GPS speed drops too low, all that will happen is it'll take longer to get to your destination airport.

If the airspeed drops too low (even if the GPS speed is unchanged) the aircraft falls out of the sky...