Participation factors/modal mass is zero with cyclic symmetry

[TimAero]

currently I'm running an Eigen frequency analysis with cyclic symmetry. When running the model, all seems to go according to what I expected. Eigen Frequencies are in line with manufactured products for all cyclic modes. However, for cyclic modes > 1 the participation factors and the modal masses become zero. I have run also the example problem from the manual "Analysis of a rotating fan using substructures and cyclic symmetry", and I see the same result; both participation factors and the modal masses become zero. Can someone explain to me why these become zero?

Thank you in advance, regards Tim

I tried newer versions of the Abaqus solver to see if it was a bug. I read the manual and ran input files from the manual. I tried to see of there are different solvers then Lanczos or Guyan might cause this. No succes. Input file directly from the manual gives same result (but also uses Lanczos and Guyan technique). I tried google and deliberate with colleagues.

 

[FEA way]

That's a good question and I'm glad that you've already done some research. It's clearly not a bug but the way it works. I've noticed in ANSYS documentation that the software calculates the participation factors only for harmonic index/cyclic mode 0 (only UZ and URZ DOFs in the cyclic CSYS) and 1 (UX, UY, URX, URY in the cyclic CSYS). In the first case, they are multiplied by sqrt(N) while in the second case, they are multiplied by sqrt(N/2) to match the participation factors from a full model. Of course, N means the number of sectors. The same seems to be true also in Abaqus. I'm just surprised it isn't mentioned in the documentation.

 

[TimAero]

Hi FEA way, thank you very much! Do you know why it does this? The mode shapes are not the same for different nodal diameters, so I would expect a different modal mass. Can I calculate the modal mass for higher nodal diameters myself from the generated output?

 

[FEA way]

The document about cyclic symmetry in ANSYS ("ANSYS Mechanical APDL Cyclic Symmetry Analysis Guide" only mentions that the modal mass is N/2 (or N if the harmonic index is 0 or N/2). I haven't found any details in the Abaqus documentation.