JD-EE
Other than ACARS messages that indicate the SATCOM ANT was still locked to the satellite at 0214z, and the rejected attempts by AF447 to log on to DAKAR ATC, the only other recorded contact with the A/C was the last squawk received at 0148z when approx 250NM from SBFN secondary radar.
I apologize for not being more precise in my previous post.
My experience with satellite communications goes back the the OSCAR 1 and later series of experimental satellites. However without getting into the nitty gritty detail of antenna design, phased array gain / isotropic dipole and linear / circular polarization, my intention was to state that the A/C was still was in comms with an INMARSAT-3 geostationary satellite at 2014z. We are informed by BEA that for a short period between 0211z and 0212z that the satellite lost contact with AF447. That is the important point, as if the antenna in use was an omni-directional with a nominal gain of 6 dBi the beam width would have been about 70 degrees - quite large for a L Band antenna (nominal 1.6GHz).
One can only imagine what was happening during this loss of signal.
mm43