Degree of Freedom Ordering of [M] and [K]
Degree of Freedom Ordering of [M] and [K]
(OP)
I am able to extract the mass and stiffness matrices using the following codes in APDL:
*SMAT, MatK, D, import, full, file.full, stiff
*SMAT, MatM, D, import, full, 2d_beam.full, mass
*export, MatK, mmf, matkMMF.txt
*export, MatM, mmf, matMMMF.txt
I then read them in MATLAB as two full matrices.
The modal analysis results in MATLAB are consistent with those from ANSYS. This proved the correctness of the two matrices.
However, I am wondering how the degrees of freedom are indexed in these two matrices by ANSYS.
From this webpage https://www.sharcnet.ca/Software/Ansys/17.0/en-us/..., I tried to read the Nod2Bcs vector and map with the node numbering in APDL. They don't match.
I also tried this one: http://www.ansystips.com/2017/10/export-stiffness-.... The DOF's do not match as well.
Does anyone have experience in the DOF ordering of these matrices?
*SMAT, MatK, D, import, full, file.full, stiff
*SMAT, MatM, D, import, full, 2d_beam.full, mass
*export, MatK, mmf, matkMMF.txt
*export, MatM, mmf, matMMMF.txt
I then read them in MATLAB as two full matrices.
The modal analysis results in MATLAB are consistent with those from ANSYS. This proved the correctness of the two matrices.
However, I am wondering how the degrees of freedom are indexed in these two matrices by ANSYS.
From this webpage https://www.sharcnet.ca/Software/Ansys/17.0/en-us/..., I tried to read the Nod2Bcs vector and map with the node numbering in APDL. They don't match.
I also tried this one: http://www.ansystips.com/2017/10/export-stiffness-.... The DOF's do not match as well.
Does anyone have experience in the DOF ordering of these matrices?





RE: Degree of Freedom Ordering of [M] and [K]
How did you determine the DOFs does not match?
Did you use both MapForward & Nod2Bcs?
Kind regards,
Jason
RE: Degree of Freedom Ordering of [M] and [K]
I've figured out the DOF ordering of [M] and [K] using MapForward and Nod2Bcs. Thanks for your tips.
I'm looking for a method to transform them back to the user ordering in MATLAB for other analyses. Do you have any thought on it?
Raymond
RE: Degree of Freedom Ordering of [M] and [K]
Matrix from the .FULL file is in BCS ordering. The index array from MapForward and Nod2Bcs maps in both directions of User Ordering & BCS Ordering.
Perhaps look into SPMWRITE to see if it fit your needs better in Matlab. The matrix is in state-space and allows for simpler DOF identification. From the help:
Kind regards,
Jason