Pitot Tubes and the Recent Air France Crash

There are at least 3 sensors and according to last developments it is ONLY one of the causes, not sure it is THE reason the plane crashed.

I'm having a hard time buying the theory also. There are several alternate ways to check air speed. And at a high altitude, there is no need to make rash decisions such as maximizing the throttle. (And even if they did, there would have to be major structural issues for that to be a problem.)

With that in mind, it dosen't make sense that a reaction of decreasing the throttle could have catastrophic results either. With contact to the ground air traffic control, the speed and direction could be easily verified.

It would be easier to buy an explanation of impact with a meteroid, which has also been suggested.

Let me throw this out as a possible cause. Composite plane materials. Do we have enough data to determine the usable life of them? Do we know what stress will do to them? Today's reports imply the plane broke apart before it hit the water.

Any comments?

I am not an aeronautical engineer nor have adequate knowledge in Aircraft Operation. However, being an instrumentation engineer, the common sense tells that there should be sufficient redundancy in the field sensors ( in this case the Pitot tubes) to ensure protection against at least a single failure. Also there must be Governor restriction to prevent unlimited throttle in the engine! As such I am not comfortable with the idea that the Pilot of the fateful ( fateless!) Aircraft increased unlimited throttle due the failure of the Pitot tube!

What are your views friends?

You asked for "our view", mine is that you do not read what was already written. This concerns redundancy of sensors. With respect to the "unlimited" command of the engine, it is a step by step reduction of automatic control and step by step transfer to the human control when the conditions rech some limits. So that every thing you thought about was alreday implemented. This makes difficult to find the explanation. For your information redundancy goes in aircraft or space controls very far up to 4 or even 5 times if it is possible. In software there are cases when the different parallel working computers were programmed by different teams with different languages in order to avoid a commonality of errors.

If you have three pitot tubes of the exact same design, and they are exposed to the exact same environmental conditions, then if one is icing, it is more than probable that all three iced over which means you really don't have any redundancy now do you?

As I understand the situation, this was probably a cascade failure. let me explain my reasoning.

What we know:

The A model Thales pitot tubes were found to be prone to icing.

The B model Thales pitot tubes that all the airlines initially started to retrofit the aircraft with were later found to be no better than the A model ones under icing conditions.

So neither Thales units were worth a crap.

There was a third (and soon to be a fourth) pitot tube that was certified for use on the airframe but it was made in the US by Goodrich and the protectionist Europeans wouldn't use it. They all used the Thales units which we have already determined were crap.

The Airbus A330 and A340 share a similar rudder design with the A320. The A320's rudder has been shown to fail catastrophically when heavily loaded from large rudder movements at high speeds. Airbus "fixed" this in software by putting a limit on the amount of rudder that could be applied at high speeds. They did not alter or reinforce the actual structure however.

The aircraft's air data unit indicated it was no longer receiving airspeed data from the Thales pitot tubes.

The aircraft was in heavy turbulence which in my opinion may have been strong enough all by itself to rip the flawed rudder off, regardless of rudder inputs.

The software "fix" that limited rudder movements was dependent on airspeed data, lacking such data, the "fix" was inoperative.

The fly by wire system does not have any tactile warning to the pilot that he or she is approaching or entering a stall. The system has no feedback mechanism to the pilot at all, so he cannot "feel" the aircraft the way a Boeing 787 (which has a fly by wire system that operates under different control laws and design philosophies from the Airbus) pilot can. Much of the Airbus control system is dependent on the AIRDU receiving airspeed data. Without that data, most of the automated safeties built into the system become non-functional.

At that speed and altitude, the margin between overspeed and stall is vary narrow, and with highly variable gusting winds aloft, maintaining the aircraft in that narrow speed envelope would be nigh impossible. I've had the experience of flying a Cessna 172 on a very mildly gusty day with a lot of thermal updrafts and maintaining stable altitude and airspeed was dang near impossible. I can only imagine how much more difficult flying in that kind of stormy weather would be, especially without any tactile feedback.

The rudder of the aircraft was found virtually intact, sheared off at it's base, just like the rudder for the A320 that crashed in New York was found intact and sheared off in the same manner. This was the crash that led to the software "fix" that prevented excessive control surface movements at high speeds. The crash was blamed on the pilot using too much rudder to fight wake turbulence from a 747 that had taken off a few minutes before. But wake turbulence is not the only form of turbulence a pilot could face, and in fact it is largely avoidable if proper takeoff spacing is maintained.

What I suspect happened:

The Pitot tubes iced over, the AIRDU no longer had airspeed data, the autopilot could no longer control the aircraft under it's control laws and dumped the situation in the pilot's lap. The pilot fought for control of the plane in very heavy turbulence and shear, the automated limits that prevented excessive control surface inputs were disabled because the aircraft did not know how fast it was going. Either the turbulence itself was strong enough to rip the control surfaces off, or the pilot inadvertently oversped, stalled, or used too much rudder to maintain control, or some combination of the four, causing a loss of the control surfaces. The aircraft then plunged to the ocean uncontrollably.

A possible contributing factor may have been the pilot's potential over-reliance on the automation of the aircraft to keep him out of trouble. He possibly failed to grasp that the safety systems he had come to rely on were no longer operative.

Therefore fundamentally the problems are one of design and design philosophy. The control surfaces are too weak to withstand severe turbulence or control inputs, probably in order to reduce weight. And the system relies too much on automated systems without giving proper independent feedback to the pilot in order to allow him to safely control the aircraft when all of the automated systems have failed. The pilot and co-pilot should be the ultimate safety system on board every aircraft. But if the pilot and co-pilot are insulated too much from the control of the aircraft, then they cannot effectively respond when the automated systems fail, which invariably they will.

you are wrong in one aspect the systems have sensitive feed-back.

That is at variance with what I have read and been told, but even if they do have stick shakers like the 787 does, I'd be willing to bet that feedback is based on AIRDU data, which lacking any functioning pitot tubes, the AIRDU had no data on whether the aircraft was entering a stall or not so it could not give that feedback to the pilot.

We are both talking about pilot tactile feedback right?

Where have you read ? I would like to use same source for information since I have different inputs. What I know is that the official investigation commission said about 1 week ago that with available informations they are not able to explain the crash and that they expect detailed forensic results and that in a few weeks the search for the boxes will start again with different means. If it would be as simple as you picture it I imagine they would have avoided the amount of money involved in this search.

What I do not know is if the different channels for data processing were identical or different. Have you any input on that? This is very important since usually when safety has to be obtained by redundancy the different channels are hardware and software different in order to avoid repetition of same errors.

I very much doubt that the "protectionism" will go so far as to sacrifice people's lives. If your sources are only from "Boeing" origins they could "coloured".

http://news.yahoo.com/s/ap/20090808/ap_on_go_ot/us_planes_speed_malfunction

http://www.dailymail.co.uk/home/moslive/article-1200132/LIVE-SPECIAL-INVESTIGATION-The-series-mysterious-Airbus-330-accidents-culminating-tragic-loss-Air-France-Flight-447.html

http://www.vision.net.au/~apaterson/aviation/airbus_critique.htm

http://www.vision.net.au/~apaterson/index.htm

http://www.guardian.co.uk/business/2005/mar/13/theairlineindustry.internationalnews

http://www.seattlepi.com/business/boe202.shtml

and I was incorrect, it was an A300 that had it's rudder break off not a A320, but they all share a similar design.

I'm sure there have been others I have read that I can't put my finger on right at the moment.

As to whether protectionism would be the basis for risking lives, I would point to the initial finding that the rudder failure on the A300 was not due to structural deficiency, but due to the pilot fighting to maintain control of the aircraft. Now, from my perspective (and I suspect the perspective of every pilot alive), maintaining control of the aircraft is of paramount importance, especially when the flight path is so close to a populated area. So that in of itself is proof that Airbus is not above sweeping major problems under the rug in order to whitewash the situation to maintain sales. Then there is the issue of Kapton wiring. Airbus was still using it up until 2005, a decade after Boeing switched to safer TKT wiring.Why? Because it was cheaper and lighter. Even today they refuse to submit their version of TKT wiring for independent testing to verify that it meets FAR 25.

How many crashes have been due to Kapton wiring over the years? dozens. But they refused to change.

Now on the new A380 they are talking about using aluminum wiring because they are overweight. Haven't we learned anything from the use of aluminum wiring in homes in the 1970's? Thousands of house fires in Texas alone are attributable to aluminum wiring. And this is what we are putting in an aircraft carrying hundreds of people every day?