Air France jet vanishes

Isn't the carbon fiber itself lighter than water? And, yes - the upper part of the stab looked to be pretty intact, the only damage being the snapped off root and the piece at the lower trailing edge.

So, it could have had some air space inside that was not compromised.

Anyway, it obviously DID float.

 

The tailplane, flight control surfaces, etc are watertight to prevent water from entering and freezing at high altitude and causing structural damage. This goes for aluminum framed aircraft, as well, although some large pieces are not watertight and have drain holes at the bottom to let water flow out. On the F-111, we had a flap vane made out of aluminum that would occasionally dislodge and fall off the aircraft. On the occasions this happened at Eglin AFB in Ft Walton Beach, FL, the flap vane would float like a cork in the Bay. We had a couple returned to us by folks who had picked them up in their boats.

I can see no reason why anyone would be pounding on the rudder controls at altitude, regardless of what the aircraft was doing, under normal circumstances. If the Airbus reverted to normal control laws at some time during a loss of directional control, however, you can see someone applying recovery controls where even a medium rudder input would overstress the vertical stabilizer.

On supersonic aircraft, water gives you an opposite problem. If there is water in the vertical stabilizer, it will boil at supersonic speeds and cause structural damage. A tap test is done to ensure no water in the aft control surfaces. On one F-111 where an inadequate tap test was done, the entire leading edge of the vertical stabilizer, incorporating the HF antenna, ripped off the aircraft.

Taz
Terry Phillips

 

On the 777
The entire fin is hollow but broken up into multiple sections.
A) From tip to base, where the the rudder hinges to the fin there are multiple access panels to the actuator systems. Some of the panels cover nothing but airspace.There are dividers between each of the panels making these sections airtight.
B) The leading edge of the fin is segmented with bullnose sections, each virtually airtight.
C) The main structure of the fin (aluminum on 777) is hollow as well , mostly one chamber, segmented by ribs etc. This chamber is closed off at the bottom by a hatch.
D) Carbon Fiber floats.
E) In the right shape even steel floats, (laws of displacement)

The Airbus fin should be similar in design and construction due to the physical requirements of the structure.
I have listened to some discussions between the Engineers in my area and looked at some comparative illustrations.
There are some important differences in how the base of the fin is reinforced around the main attach fittings on the Airbus versus Boeing.

 

Interesting, and thanks, all, who chimed in.

CW

 

I think it's the other way around. The turbulence was pounding on the rudder/fin. I can only imagine.

 

Spasso- Could be. That would be some really rough air, but then that area has the most violent thunderstorms in the world.

I see we posted at the same time and said just about the same thing. The vertical stabilizer mount on the Airbus does look suspect. A lot of stress on the points where the composite structure meets the through-bolts.

Taz
Terry Phillips

 

New pic.........
Image Unavailable, Please Login

 

The Rudder Travel Limiter on A330s limits rudder pretty severely @ speed -- the limit is about 4 degrees at cruise. This is compared to 30 at takeoff. So the pilots and/or turbulence (or some combination of both) wouldn't have to be hitting it too hard -- high speed + more travel + buffeting could be enough to do it.

I should add, this is a downside of 'fly by wire' -- with computers limiting rudder travel, pedal travel is likely unaffected and feel unchanged, while the rudder can move more freely than before. So similar pedal pressure can yield dissimilar rudder results, but no difference in 'feel' for the pilots.

 

Taz, years ago when I was working there were many comments by the structures guys at the lack of structural depth in the 48 section ( tail sect.)of the Airbuses and I have to wonder if that isn't why the vertical tail on the A330 series is bolted on with a flange which is the worst type of connection for a tall structure that is subjected to extreme leverage. There is no way to disperse the loads into the structure to which it is mounted when there is a stab. center box right under it. Your comments on water collecting in flight surfaces brings back our working on the SST and the studies that exposed too many Boogiemen lurking in the structures exposed to many weather phenomena. Stuff that the layman never thinks about.
Switches

 

Just saw where the French pilots aren't going to fly 330-340 until they replace the sensors.

Art

 

I believe one of the ACARS messages indicated that the control system went into "Alternate Law" which has fewer protections, although I don't know which protections are changed.

 

This may not be real:

http://blog.seattlepi.com/aerospace/library/ACARS_AF447.pdf

If it is, note the first ACARS message: "RUD TRV LIM FAULT"

 

Missed a few old posts but you guys were a bit harsh on the 75 crew out of Lima. The static ports and not the pitot tubes were covered with clear tape. I'm sure most FO's including me would have missed it on a walkaround at night. The problem was not that the airspeed was unrealiable, but rather the airplanes warning systems. Soon after take off from Lima the warnings gave both stall and overspeed warnings. All static ports including the standby were covered for maint. I've heard the CVR in training very scary... The airplane is telling you completely different things. The altimeter is showing way off and that is sent via transponder to the center and center tells these guys to decend to try and get them back... All the while the plane has the shaker and clacker going off at the same time with master warnings going off... all at night over the ocean!!!! Suprising they kept it up as long as they did. They did a hell of a job considering the sensory overload and wrong info from ATC.

Now as to speculation on Air France... If you look at the messages sent to Airbus, this is my take. The A330 has had past problems with losing the ADIRS and Flight control computers. Loss or error of the ADIRS (AIR DATA INERTIAL REFERENCE SYSTEM) could send messages indicating air speed disagreement. My hunch is the pitot tube issue is a red herring. Not sure why it might freeze just because they flew through a TS. Those three seperate tubes are all heated and would fry your hand if you touched one.

My though is that there was an electrical problem caused by something (perhaps lightning perhaps not). The Flight control computers malfunction and an ADIRS computer or both go down. Normal law degrades to Alternate law. Another message sent. You lose some protections but no big deal just on the loss of control laws. If further degradation occurs you fall to Direct Law. Still not a big deal really, all Boeings and all other airplanes fly in Direct Law at all times, i.e., no computer controled protections for bank, pitch, stall, etc.

The cascade of electrical problems caused the flight control computers, the ELACs and SEC to fail. There is a messages sent about spoiler control failure (SEC is Spoiler control computer) (ELAC is elevator and aileron control computer).

This is when the real problem happened. There is precedent for this happening in A330s I believe it has happened two times. The airplane control law would then degrade to Mechanical Backup. At this point the sidestick is useless. Your only controls left to fly the plane is the rudder and the Trimable Horizontal Stab (trim wheel next to the thrust levers). It is tough enough to fly it in the sim in calm air and keep the airplane straight and level in mechanical backup. In severe turbulance like they were flying through, I would say it would be nearly impossible.

So if you look at the messages and timing of the messages- the break up occurred about one minute after the total failure of the flight control computers. My guess is that the pilots fought for a minute until they lost control and the airplane exceeded it structural load capacity.

Just my perspective and I base most of it on the timing of the messages that were sent from the airplane, and that I'm type rated on the 320 bus with very similar systems to the 330. I hope they find the FDR.

 

That ACARS printout looks like it could be right. The Rud Trv Lim Fault is a Flight Control Computer (F/CTL) function and is most likely related to the failure to the multiple failure associated with losing the F/CTL computers. The auto pilot disengaged at this time, as well as the auto flight computer, both flaged on the pilots PFDs - as well as going to Alternate Law warnings.

Its interesting to not that two minutes later there ADR or Air Data Reference is in disagreement. This would be the pitot/static instruments controled by the ADIRS computer. The timing of this malfunction is why I think that the disagreement is the effect of failures of systems rather than the cause of the related problems.

Not to sound too much like a cook, but the public won't get on an airplane if they think it will fall out of the sky because of a computer glitch. They will if they think an instrument was manufactured incorrectly and it has been replaced. I think that is why they are changing out the pitot tubes.

 

great posts... thank you.

 

Thank you for your input. I think what I would really like to see is recovery of the boxes, evaluation and answers. Speculation is healthy when based in fact but at this point it's anyone's guess.

They aren't going to quit on this one until they find them.

 

That is all that we have right now, guessing and trying to figure out what happened on what we have learned but not on what we know. You guys are way ahead of me on all this stuff but I have to think, like Spasso, that until they recover the FDR we will only speculate on the cause of this incident.
Switches

 

Many thanks. I agree wholeheartedly we will not know even a good part of the story without the FDRs.

We come back to Airbus and Air France pointing fingers at each other. Big bucks involved either way. Much worse for Airbus because there is so much at stake. This includes the KC-X contract for the USAF, which is a modified A330. Tankers have to work, or you could lose a whole flock of accompanying fighters.

Taz
Terry Phillips

 

For the conspiracy theorists. Anyone see a common shareholder in the holdings of EADS, Thales (maker of the pitot tube) SA and Air France? Might want to remember these as the spin process starts.
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

With regard to sealing/taping/icing of pitot/static ports:

As some of you will know taping/sealing of ports is a VERY standard maintenance procedure. That being said in the case of the 57 that had clear tape covering the static ports it was much the fault of the tech, as it was the pilot that failed to notice it. In fact it was MORE so the techs fault.

My standard operating practice is to use a contrasting color tape, attach a length of safety orange surveyors ribbon to the sealing tape that is long enough to hang down below the level of the fuselage, AND lift the front of the tape such that in the off-chance that myself and the 6-8 other sets of eye's miss the whole shebang at least the tape has a chance of getting ripped off due to airflow over it. In the 57 incident it seems that there was not very good maintenance practices.

I would tend to concur with the speculation that icing played a role here in this incident, but frankly the Brazilian/French investigative authorities really don't have anything to go on. A few ACARS messages and the maintenance logs? The logbooks are only as reliable and trustworthy as the person writing in them (not to suggest that there was ANY wrongdoing involved, just saying that a log on it's own is not worth much), and ACARS messages will only tell you that something went wrong, not why.

Hopefully some major wreckage is found, and soon the flight recorders.......times ticking away on those ULB's......

 

I think it's quite something that, given all the stuff going on in that aircraft, the ACARS was still able to send messages.

 

I've just heard on the news that there may have been two terrorist suspects on board.

 

Speculation is all my post was. Just my take on what might have happened based upon the messages and what I know of the history of the A330 and Airbus systems. But you are right I only know what has been released. And yes there are undoubtedly facts that have not been released and lawsuits will be filed shortly if not already.

My thought is that if my scenario is anywhere close to what happened, Airbus will do all it can to protect themselves. This accident will challenge Airbus' complete design philosophy. A philosophy that works great almost all of the time. In fact for the Boeing lovers out there, Boeing has adopted some of Airbus' ideas from fly by wire to reduction of memory items and trouble shooting when problems do occur. And as for easy on the pilots, the Airbus is far and away a lighter workload. I have not flown a Boeing but commuted every week for over a year from SEA to FLL for work and rode the jumpseat in 75s 76s and 73s nearly every time and the workload is significantly higher on those planes than in a bus.

But any airplane works great when it works (MU-2 for instance) and they don't call it the ScareBus for nothing. Airbus' do crazy things at times. Most problems occur when you program the box wrong or you do not activate the proper sequence and you expect something that you didn't request (operator error). But I've seen the plane get weird for no reason and it is not fun. One of our planes had an uncommanded roll while going into LGA on day. I was in the training dept. that day and lots of people were very concerned. If it comes out that there is another flight control issue due to computer problems, Airbus' future may be in serious trouble.

 

Automation and self-thinking airplanes do indeed lighten the workload and as you point out, opens the door for unanticipated events that have to be nerve-wracking.

Like my systems and programs here at work, (3D Fly-through being one of them), when they work they are a God-send but when things lock up it is the ultimate PITA! Fortunately when MY systems go down my elevation is only 25 feet , not 38k feet.

I still lean towards PIC Final Authority over flight systems.
I guess there is always a trade off.

 

As I mentioned one time previously I've had personal experience with software failures on flight critical equipment, it wasn't pretty, but thankfully nobody was hurt, and the only damage was a pair of bent skids from a night autorotation. The problem with giving PIC final override is that it has to be intutive, so that whatever he does with the controls or throttles is correct and natural. Sometimes it isn't and if the PIC overrides he can end up in a smoking hole if he doesn't do the right thing. In our case the pilot got handed control of the engine via a manual throttle, but he didn't recognize it and didn't reduce power and oversped the rotor when he leveled off. Coulda, shoulda and woulda been a non-event, if he had taken control of the engine, either kept climbing, or reduced power, hit the computer reset button he would have flown on with no problems. As it was, he was faced with a night autorotaion on NVG's, which he was lucky to not do more damage.

I have a concern in that the Airbus has a weak tail mounting arrangement that they were bandaiding with software (limiting control movement above certain speeds). Problem is, if you get a bogus speed signal (and you WILL at some point get a bogus signal), does the airplane fall out of the sky? As I said before, the system has to be sufficiently fault tolerant so that even two or three sensor or failures don't let that happen. There are other things you can do to add levels of additional redundancy like limiting rudder deflection at altitude, but still, you are covering a fundamental structural weakness with software and that bothers me.

The previous example of an Airbus losing its tail was a result of wake turbulence. In this case we have an airplane that encountered severe turbulence in TRW, and we have a tail that doesn't look like it was on the airplane when it crashed, not sure if it was the root cause of the disaster, or if the tail came off after the plane broke up, but we will likely find out if they can get the FDR out of the water at some point. It may not have needed a bogus speed signal to fail, maybe just severe turbulence....

I'm biased, but I havent't seen any control surfaces falling off of any Boeing airplanes lately...