Air France 447 - On topic only!

[Evan]

Switching gears a bit, I have question about the last comm sequence of AF447.

01:35:43 “ AFR 447 -AIR FRANCE FOUR FOUR SEVEN, thank you.
01:35:46 “ ACC-AO -Welcome, maintaing flight level three five zero,
say your estimate TASIL?
01:35:53 “ ACC-AO -Say your estimate TASIL?
01:35:59 “ ACC-AO -AIR FRANCE FOUR FOUR SEVEN estimate TASIL?
01:36:14 “ ACC-AO -AIR FRANCE FOUR FOUR SEVEN say your estimate
TASIL?


In that very last transmission to AF447 from the ATLANTICO controller asking them for an estimate to TASIL, are they instructing them to maintain FL350 until TASIL? AF447 never reads back, so we never know if they received the transmission, but we know they received another transmission just three seconds earlier.

When we covered this way back in the thread, the consensus was that AF447 dropped out of VHF range at that moment and therefore the ATLANTICO controller simply repeated the request several times and then gave up. This was considered acceptable because the instruction was simply to maintain current flight level and there was no vital information in the transmission.

Now that I have read the flight plan, I realize that it included an FL change at SALPU to FL370. Therefore, if ATLANTICO was instructing AF447 to maintain FL350 until contacting DAKAR, he was effectively instructing a FL deviation from the flight plan. Since there was never any readback, how could ATC know what flight level they were at after SALPU?

Is it possible to SELCAL them with HF or SATCOM? Is it acceptable to have a flight on a mid-Atlantic lane at an indeterminate flight level? Are flights required to contact ATC if they deviate from their flight plan altitude? Or does the lack of traffic at that point allow them to rely on TCAS alone?

[Gabriel]

I will present this "exhibit" that I prepared with no other comment by now but that something looks odd.

(Sorry for the crappy quality. I prepared a pdf that barely weights 1/5 of this jpg and has much better quality, but the forum would only accpet 19 KB of pdf while several MB of jpg. If someone wants the pdf please PM your e-mail)

[Myndee]

Originally Posted by Evan

Switching gears a bit, I have question about the last comm sequence of AF447.

01:35:43 “ AFR 447 -AIR FRANCE FOUR FOUR SEVEN, thank you.
01:35:46 “ ACC-AO -Welcome, maintaing flight level three five zero,
say your estimate TASIL?
01:35:53 “ ACC-AO -Say your estimate TASIL?
01:35:59 “ ACC-AO -AIR FRANCE FOUR FOUR SEVEN estimate TASIL?
01:36:14 “ ACC-AO -AIR FRANCE FOUR FOUR SEVEN say your estimate
TASIL?

I am so glad this thread is back on topic. I really hope they get the searches going again soon so we will have some new things to discuss.

Are flights required to contact ATC if they deviate from their flight plan altitude? Or does the lack of traffic at that point allow them to rely on TCAS alone?

I am certainly not a pilot, but I thought you had to have clearance for the altitude unless they gave you a block? I also thought that it is ultimately the ATC reponsibility to keep adequate space between planes, and TCAS is there as a protection measure. (Please correct me if I am wrong.)

These are definitely some interesting points that you brought up.

[Evan]

Originally Posted by Gabriel

I will present this "exhibit" that I prepared with no other comment by now but that something looks odd.

(Sorry for the crappy quality. I prepared a pdf that barely weights 1/5 of this jpg and has much better quality, but the forum would only accpet 19 KB of pdf while several MB of jpg. If someone wants the pdf please PM your e-mail)

Nicely done. It illustrates how little of the MTOW is available for payload, how mostly you're flying fuel and airplane around. It also shows how the route reserve in any case is dwarfed by the fuel to alternate and the final reserve, and is nowhere near 10% of the route fuel. I'm seeing that, as this is not a long flight over open stretches of water (like the LAX-Sydney route for instance), the route reserve is not provided as a safeguard against fuel starvation, but is basically just there to prevent an undesirable technical stopover, which is always available on the latter part of this route. In all cases the route reserve isn't providing much (or in the case of the RIF, any) time for diversions on top of other burn-rate factors. But the additional extra fuel seems to make a short diversion possible (about 1/2 as much as what the fuel to alternate provides for).

One thing I would like to see is a chart comparing the route reserve (and extra fuel) to the cargo load (after deducting pax and luggage). These seem to be the two variables that trade off each other. I also think 25 kg/passenger luggage might be a low estimate on a transcontinental flight.

[MCM]

G'day Evan.

The Flightplan is just that... a plan.

It is based on forecast winds, as well as forecast weight at a given waypoint.

So, the indication that the aircraft was planned at FL370 at SALPU is like a "heads up" to ATC that the crew will likely be looking for a climb at that point - but it is not something that the pilots directly follow. The actual time they request that climb will depend on actual winds, actual weight etc... and is far more efficient.

Also, in this day and age, being stuck at an altitude that is not your optimum (say not being able to climb to FL370 when they want to, but an hour later) usually has a negligable effect on fuel due to the way we fly on a cost index.

Its important to remember that the aircraft would maintain its LAST CLEARED LEVEL. You do not climb based on your flightplan, only based on an ATC clearance to do so.

They would not be at an "indeterminate" level, because they would be maintaining their cleared level.

They could be SELCAL'd if they had tuned the correct HF frequency. I don't know the HF/VHF mix in that part of the world.

[mfeldt]

So the conclusion is they DID maintain FL350 - either because the understood what the controler said but for some reason did not acknowledge, or because they didn't hear him and thus had no clearance.

[MCM]

They maintain FL350 because that is their existing clearance. The new controller is just confirming that existing clearance... if the controller said nothing, they would maintain 350.

[Evan]

Originally Posted by MCM

They maintain FL350 because that is their existing clearance. The new controller is just confirming that existing clearance... if the controller said nothing, they would maintain 350.

The flight plan (and the route fuel) called for a flight level change at SALPU. If they planned this FL change, how do they go about getting clearance for it? Do they request it in advance before they lose VHF with ATLANTICO or do they have to try to contact ATLANTICO on HF once they reach SALPU? (I always thought the flight plan was a pre-approval, and only deviations had to be cleared). Because they didn't request FL370 at any point, can we then assume they opted to stay at FL350 due to turbulence?

[Peter Kesternich]

Originally Posted by Evan

The flight plan (and the route fuel) called for a flight level change at SALPU. If they planned this FL change, how do they go about getting clearance for it? Do they request it in advance before they lose VHF with ATLANTICO or do they have to try to contact ATLANTICO on HF once they reach SALPU? (I always thought the flight plan was a pre-approval, and only deviations had to be cleared). Because they didn't request FL370 at any point, can we then assume they opted to stay at FL350 due to turbulence?

To change the FL they need to be cleared by ATC - and by the control sector they are in, and it doesn't matter whether they use HF, VHF or satellite communications to obtain the clearance.

If the guys on AF447 didn't request a clearance from FL350 to FL370 at any time during the flight, we only know that they didn't request clearance and we can assume nothing, because there are so many reasons why they didn't request a clearance. We can speculate that turbulence may have been a reason for not requesting a climb to FL370, but then taking the aircraft higher can also be a course of action to get OUT of the turbulence and into calmer air. I think there is no direct relation between staying at FL350 and turbulence.

[Evan]

Originally Posted by Peter Kesternich

I think there is no direct relation between staying at FL350 and turbulence.

What other reason would cause them to maintain FL350 at a higher burn rate?

[Peter Kesternich]

Originally Posted by Evan

What other reason would cause them to maintain FL350 at a higher burn rate?

Maybe fuel burn had been lighter than expected for whatever reason and the plane wasn't light enough yet to climb economically to FL370.

[Gabriel]

- Maybe because they were not cleared to do so.
- Ok, but if they wanted higher and were not "proactively" cleared, why didn't they request it to ATC?
- How do you know they didn't try to? Maybe they tried but couldn't make contact (after all, why wouldn't they answer the Atlantico center).
- Or maybe they were already with big troubles and too much workload by when they got to SALPU (when dealing with an unreliable airspeed situation and a lot of systems going off-line as a result is not the best time to request a more economic flight level). Or maybe they never reached SALPU (I don't know where is SALPU relative to the point of last radio contact and to the zone of loss of control and crash)

[mfeldt]

Hmmm....

the flight plan reads:

UN873 INTOL/M082F350 UN873 SALPU/M082F370

What does that actually mean? is FL270 planned to be achieved by the time they reach SALPU? Or would they request permission to climb once they are at SALPU?

Communications go like this:

01:35:15 “ AFR 447 -AIR FRANCE FOUR FOUR SEVEN, by checking
INTOL zero one three three, level three five zero,
SALPU zero one four eight, next ORARO zero
two zero zero, selcall check Charlie Papa Hotel
Quebec.
01:35:38 “ ACC-AO Acionamento do código SELCALL
01:35:43 “ AFR 447 -AIR FRANCE FOUR FOUR SEVEN, thank you.
01:35:46 “ ACC-AO -Welcome, maintaing flight level three five zero,
say your estimate TASIL?
01:35:53 “ ACC-AO -Say your estimate TASIL?

So apparently while going by INTOL, they estimated to arrive at SALPU about 13 minutes later. Here at INTOL, they were ordered to maintain FL350. Should they have requested 370 in this conversation?

While not answering the request to estimate TASIL, ORARO had already been given and it's not far from there.

Way, way back in the old thread I posted a graph of a sloppy reconstruction of the average ground speed, which started to drop dramatically after 01:53 at EPODE (Is that now SALPU??). resulting in an average speed of only 300KTS between EPODE and ORARO. This is strange, since at 01:33z AF447 already estimated the arrival time at ORARO correctly to be 02:00z. Did they plan to slow down? Or did they reduce speed due to entering turbulence? Somewhere I read that the manual of the A330-200 gives the turbulence penetration speed as 260KTS. Or is the reduced ground speed maybe a sign of the aircraft climbing?

I'm confused...

[Peter Kesternich]

Originally Posted by mfeldt

What does that actually mean? is FL270 planned to be achieved by the time they reach SALPU? Or would they request permission to climb once they are at SALPU?

Normally, pilots request clearance before a waypoint so as to cross it at the altitude given in a flightplan (otherwise a flightplan and the format used for communicating speed and altitude wouldn't make much sense).

Originally Posted by mfeldt

So apparently while going by INTOL, they estimated to arrive at SALPU about 13 minutes later. Here at INTOL, they were ordered to maintain FL350. Should they have requested 370 in this conversation?

Like MCM already stated, a flight plan is just a plan. What really happens on the flight as to routeing, speed and altitude is a different matter and of course any or all of these parameters can be changed by the flight crew or ATC. So the question is not what the crew of AF447 "SHOULD" have done about the altitude. They simply didn't do it and the only question possibly relevant to the accident is "why" they didn't do it.

Originally Posted by mfeldt

Somewhere I read that the manual of the A330-200 gives the turbulence penetration speed as 260KTS. Or is the reduced ground speed maybe a sign of the aircraft climbing?

Ground speed is a messy thing to deal with when talking about flight because it has so little relevance to the operation of an aircraft. If the manual says that the turbulence penetration speed for the A330-200 is 260kts, then of course it is stating AIRspeed not ground speed. A reduction in ground speed can have several reasons but at this point it is difficult to say what caused a drop in ground speed.
As your post referred to a "sloppy (!)" reconstruction of average ground speed, I suggest we drop the topic of ground speed until there is any hard evidence about it and until we have the airspeed data to correlate with the ground speed data. Anything else would be wildly speculative.

Originally Posted by mfeldt

I'm confused...

No need to be, mfeldt

[Evan]

Originally Posted by mfeldt

I'm confused...

Maybe you should go back and review your calculations. AF447 was on the UN873 airway, where EPODE is halfway between INTOL and TASIL. SALPU is less than a third of the distance to TASIL, so they are not the same waypoint. SALPU is approx 215nm from the first ACARS transmission.

One thing that I considered from the very beginning was the possibility of a collision with something that wasn't supposed to be there, something covert or illegal flying on a stealthy course parallel to the UL695/375 airway. I thought this because it seemed strange that AF447 disappeared in such close proximity to this major intersection of crossing airways.

UN873 intersects the UL695/375 airway perpendicularly at ORARO, and the first ACARS occurs just beyond that intersection. When I overlay the weather images, it appears that such a collision would have occurred in very poor visibility within the cloud formation. As we have learned from the Gol 737NG collision in Brazil, if the other a/c does not have its TCAS active, TCAS can fail to warn against encroaching traffic. If the illegal aircraft had been monitoring flight plans for that day, they would have expected AF447 at FL370, so maybe they were at FL350. AF447, I'm told, would have deviated from the flight plan and maintained FL350. I speculate that a non-catastrophic collision might have severely damaged the front of the A330, including the pitots, resulting in the ACARS messages that were received. The second plane might have flown a bit further and then disappeared in the ocean (although I would expect something to be recovered). I'm sure there is a reason why this theory can be refuted (other than the remote odds of course), and I'm counting on the forum to let me know what that is.

[Peter Kesternich]

Originally Posted by Evan

One thing that I considered from the very beginning was the possibility of a collision with something that wasn't supposed to be there, something covert or illegal flying on a stealthy course parallel to the UL695/375 airway.

That's an interesting thought, and since we have now started speculating I would like to throw in another idea: What if it was a bomb? Not the political, terrorist kind of bomb that someone would claim responsibility for but more like the life-insurance fraud or personal act of revenge kind of bomb. It could have been small enough to cripple the aircraft but not blast it right out of the sky which could explain the progressively worsening automated reports from the aircraft. Any thoughts or comments about that?

[Evan]

Originally Posted by Peter Kesternich

That's an interesting thought, and since we have now started speculating I would like to throw in another idea: What if it was a bomb? Not the political, terrorist kind of bomb that someone would claim responsibility for but more like the life-insurance fraud or personal act of revenge kind of bomb. It could have been small enough to cripple the aircraft but not blast it right out of the sky which could explain the progressively worsening automated reports from the aircraft. Any thoughts or comments about that?

I don't know if it can be ruled out, but the aircraft did nor decompress, nor was there any ACARS indication of overpressure at the beginning of the sequence. There was also no MAYDAY, as one would expect after a non-catastrophic explosion. If a timed explosive were planted on the aircraft outside the pressure hull, I suppose that could be a possibility. But none of the recovered wreckage indicated fire or explosion. Most importantly though, how do you directly explain the ACARS sequence, which begins with an air data disagree? The sequence you would expect from an explosion would have a different 'fingerprint'.

The problem with a bomb theory is that it is pure speculation. Nothing points to it. I'm trying to speculate from clues in the evidence we have.

[Peter Kesternich]

Originally Posted by Evan

The problem with a bomb theory is that it is pure speculation. Nothing points to it. I'm trying to speculate from clues in the evidence we have.

Okay, so where are the clues in the evidence for your theory?

[EconomyClass]

Flying through the ITCZ usually presents no great problem if one follows the usual practice of avoiding thunderstorms. He usually can find a safe corridor between storms

http://www.aviationweather.ws/092_Circulation.php

In short, you fly like a halfback runs, to "daylight". You can't set a flight path and grimly stick to it.

[Evan]

Originally Posted by Peter Kesternich

Okay, so where are the clues in the evidence for your theory?

Close proximity to the UL695/375 airway, deviation from planned flight level, near zero visibility. Light collision might have sheared off the pitots. That's all I've got for this one.