I am having some problems with a bird strike analysis and would like to see if anyone has any suggestions.  I am using the methodology from "4493 Best Practices for Birdstrike Analysis" on Simulia's support sight, as well as the bird provided there.  This approach uses the CEL method.  When the bird impacts structure, little pieces of the Eulerian mesh often break off, as is expected.  However, they often fly off at such a high speed that they punch holes in downstream structure in an unrealistic manor.  The only thing I have thought to try is to increase the size of the Eulerian mesh (adding more buffer between the bird volume and the edge of the Eulerian mesh), though this will likely add a lot to the computation time.  I am trying this now, but in the mean time, I thought I'd ask to see if anyone else has ran into this.  I have experienced this through numerous Abaqus CEL bird strike analyses.