hello,
As a computational engineer, I have worked on simulation of dynamic meshing problems associated to aerodynamic flutter and the "Wing Rock" instability through free to roll and forced oscilation techniques. I usually work on Ansys (Fluent+mechanical+system coupling). As per my Experience, the generation of negative cell volume is more of a re-meshing setting specific problem and less of a time step. Although time step play a role, as the motion to capture is small for smaller time step, a proper smoothing and explicit setting is a must for such problems. I can help you more, if you can share some more details of your problem.