I agree that simulator training is a very valuable way to experiment with handling TR problems but would add the proviso that the simulation may be innacurate.
Having flown the FSI 212 sim at DFW several times it is easy to come away with the impression that a TR drive failure can be controlled fairly well. In an extreme example which was basically a bit of 'wazzing' I took off from the simulated aircraft carrier, had a loss of TR drive at about 65 KTS, flew the thing horribly cross controlled to about 1000' abeam the ship then autorotated back down to the deck and survived. I don't believe that to be a true representation of the a/c's behaviour. The SAS sim at Stockholm, on the other hand, is a newer generation. TR drive failures can be truly aerobatic events, generally resulting in the sim freezing with the 'world' inverted, to protect itself.
Having spoken to pilots that have had TR drive failures in two separate AS332 accidents, one happened in a relatively high power setting in level flight and even entering autorotation did not stop the rolling/pitching/yawing; the PNF had to retard the speed select levers before things settled down. The handling pilot compared the manouver to a lomcevak! (negative flick roll going vertical, in an aeroplane). They executed a text book ditching and were all saved. The other one happened in a descent at 70 kts after the pilot had diagnosed the probability of impending TR drive failure and was a far less dramatic event, again resulting in a perfect ditching with no casualties.
Yet another accident with a 330J in the cruise resulted in the a/c tumbling all the way down to the sea.
So, being aware of the variables and trying it out in a sim is great, but come the day there is a very good chance things may not go to plan.
I suppose that is one thing in favour of fenestron a/c; they should fly happily in a straight line at about 100 kts with no drive.
------------------
Another day in paradise