Why are frequencies different in NASTRAN/ANSYS/ ... and SDT

Question A simple plate model consisting of quad4 elements is made using PATRAN/NASTRAN. The modes resulting from a NASTRAN run and from importing of the model into SDT then computing eigenvalues differ. Why ?

Answer The element formulations are not the same. There are many shell element formulations and the best is not STANDARD as is usually the case for volume elements (tetra4 is a notable exception). Note also that there are different shell formulations implemented in SDT (use pl(:,3) to select the formulation).

If you want the same frequencies in SDT use the same elements (run NASTRAN with PARAM,POST,-4) and import element matrices using FEMLink

  a=nasread('strip.op2')
 def=upcom(a,'computemode full 6 10 1e3 0');
 [def.data a.def.data(:,1)]

You then have identical values. Element matrix import is also implemented for ANSYS.