Very interresting radar data.
The military plot shows one signal point with about 1000 ft less altitude but maintaines climb afterwards on the same line as before, this looks like a momantary malfunction of the altitude encoder (caused by whatever).
About 10 seconds later the civil radar plot ends for about 100 seconds and then starts again. Altitude is still in coincidence with data before the gap, but heading and altitude shows some deviations from the normal values indicating some trouble. Military data does not look unusual at that time.
At shortly after 7:28 the civil plot ends and the military shows a plane in great trouble with heading reversals.
A altitude gain of about 2500 ft and a ground speed reduction by half together with a heading reversal looks like a half loop upwards. This is followed by total loss of ground speed (which means very steep dive) but only about 5000 ft drop, before military data plot also end (about 3 1/2 minutes later).
Time between first indications of trouble and loss of signal (at the same altitude !) is more than 8 1/2 minutes, and there was no time for an emergency call for any of the 3 crew members ? Verry unusual.
Aviation week online reports, none of the 80 bodies recovered so far show burn marks. Otherweise some sort of fire could explain the temporary signal loss due to loss of electric energy for the transponder, this could also explain the missing emergency call simply because the radio was not avaiable. Very strange alltogether.
Let´s wait for the FDR/CVR and hope data aquisation does not stop at the time the civil signal was lost ...