Hi @cmcc,
Sorry this got stuck in moderation. I removed the link to the google drive since the file is no longer there and the powerpoint shows your the problem pretty well.
This is an interesting problem and I'm not very experienced with FSI. So going purely off what I see when I open those LS-Dyna examples, it looks like the impacting bar is not node to node connected with the fluid, whereas your yellow solid wall looks like it is sharing nodes with the fluids. This is probably what's causing your mesh to stretch and negative volume so maybe re-think how to set up the problem so that your impacting rigid wall is not nodally connected like those dyna examples.
Again, this might be dumb advice because I don't know much about this particular type of simulation, but just going off what I can find.
I am using Firefox and I don't see the PowerPoint slide deck (I see a garbage in a browser).