Problem in Compution of Transonic Unsteady Aerodyn
 

I am computing the drag, lift& moment of a fluttering wing using Dynamic mesh module in Fluent 6.1.22. The initial velocity is given to the wing to start the first time step, then UDF is employed to calculate the pressure on the wing to get the current velocity. The time step is as small as one thousand percent of the fluttering period. Unfortunately, I still got non-positive volume afer about 40 time steps. Has anyone experienced such thing or does anyone have any suggestion? Thanks!