[John Rogers]

James pretty much hit all the right points spot on. When I went to basic nuclear power school at Bainbridge MD in 1966 the first week of physics training classes were spent explaining why the US Navy's reactors (shipboard, NOT test) could not have that happen. About the only exception was during a refueling when the cover was off and fuel cells being inserted. That time frame is when the famous melt down happened on the Chalk River reactor in Canada when all the water was allowed to drain out of the transfer vessel instead of working extra time to put the fuel in place.

Not sure about civilian reactor control rods but US Navy reactors have enough spring pressure to seat the rods even if the ship is upside down and I worried about that stuff for over 20 years!