HarryMann;
If I may offer some thoughts on why there wasn't much acceptance of a "tail-heavy due to the trim tank" theory and why, in comparison with other theories, it is far less plausible to theorize loss-of-control due to a trim-tank/CofG issue.
The trim tank will carry approximately 4800kg when full. Aft fuel transfer begins at FL255 (if the trim tank wasn't full at takeoff), and forward transfer occurs during flight as fuel is burnt. The fuel load for AF447 was 70.4k kg. I can't recall directly from my own experience with similar fuel loads but am almost certain that the trim tank would have been full at takeoff - I don't have the distribution tables for an A330 with center tanks.
First, the CofG for takeoff was well within limits and the calculated CofG 3:41 into the flight was approximately 37%. My numbers show the aft limit is 39%. The QRH Trim Tank Fuel Unusable Abnormal Procedure states that if the Trim Tank fuel is unusable the maximum flight time, depending upon fuel distribution, is 4hours at which time the aft CofG would be reached.
Since the flight was only 3:41hrs long (BEA Report), the trim tank would have had about 4500kg of fuel which begins transferring forward at cruise altitude. The crew would have to have known about any trim tank problem resulting in unusable fuel at least upon reaching cruise altitude if not slightly after. The flight would require this fuel for the arrival at Paris. Unusable fuel, especially given the weather enroute (diversions), would present a significant operational decision to turn back or continue which would have had to have been made in these first 3hrs.
There is no evidence either by communication or in any ACARS messages received that there were trim tank issues.
Second, the QRH drill then states that "For Landing", *if CG > 39% - Approach Speed Vls+10kt with the landing distance procedure to be applied, (landing distance increase by 1.25).
So a CG > 39% is catered for by a 10kt increase in Vls, meaning that the aircraft can be flown without issue. There are no cautions, no imbalance limits stated between trim tank and wing tanks and no warnings associated with this condition to which the attention of an A330 pilot is drawn in the AOM.
That doesn't mean that handling characteristics in abnormal flight such as approach to, or full stall would not be affected. What it means, as I interpret this as an experienced A330 pilot is, (and what I would expect from my aircraft is), a fully controllable airplane with no squirrely habits or handling characteristics in all flight regimes with a CG > 39%. That is, I expect as a pilot of the airplane, what the certification of the aircraft means - it is controllable even with maximum fuel in the trim tank at the expected landing weight, on approach.
That is why I would consider the CG and Trim Tank theory highly implausible at least as an initiating cause when in heavy turbulence etc.
All bets are off in any severe loss of control/jet upset.