Search the Community
Showing results for tags 'superelement'.
Found 2 results
In tutorial there is an example to generate superelemnent in linear static analysis. But I need to do modal analysis using superelement. I hope very much if someone in this field can give me a step-by-step manual I can follow. Thanks!
Dear I have an assembly that I have decided to divide into 5 subassembly (A,B,C,D,E) Using Craig-Bampton method, each of this assembly has been condensed (h3ddmig but I have tried with .pch also) using only the visualization of plot elements (plotel3 and 4) In the master model, the residual nodes (aset) of each assembly are connected each other using beam and rod with different dof (in order to simulate a mechanism) and a rigid with the master node grounded and I asked for a certain number of modes (normal modes). NO ERROR in the output file but neither results (h3d) but a serie of OSxxxx_er.rs~ This is the what i have in the output: Lanczos solver warnings for eigenvalue subcase 1. - At some point in the process, an inconsistent inertial count was found. This implies that some of the calculated frequencies may be inaccurate, especially the duplicated frequencies. To avoid this, specify both bounds and keep the bound away from the currently calculated frequencies. *** warning *** the rayleigh quotient x'kx / x'mx for the 1-th eigenvector does not match the computed eigenvalue lambda ( 1): -9.86447590101136D-06 x' k x : 5.21843506993811D-05 x' m x : 1.00000000000000D+00 discrepancy : 6.20488266003924D-05 *** warning *** the rayleigh quotient x'kx / x'mx for the 2-th eigenvector does not match the computed eigenvalue lambda ( 2): 4.00771124962063D-05 x' k x : 7.34242225598191D-05 x' m x : 1.00000000000000D+00 discrepancy : 3.33471100636129D-05 *** warning *** the rayleigh quotient x'kx / x'mx for the 3-th eigenvector does not match the computed eigenvalue lambda ( 3): 5.23449252423092D-05 x' k x : 1.36220146252925D-04 x' m x : 1.00000000000000D+00 discrepancy : 8.38752210106160D-05 eigenvalues and eigenvector residuals ------------------------------------- eigenvalues norm of estimated (radian/sec)**2 residual accuracy --------------- -------- --------- -9.86447590101136D-06 2.4225D-06 2.4225D-06 4.00771124962063D-05 1.2132D-06 1.2132D-06 5.23449252423092D-05 3.9402D-07 3.9402D-07 *** Warning for eigenvalue subcase 1 Number of modes requested = 3 (0 = all) Number of modes found = 3 warning status index = 10 maximum relative discrepancy = 0.83875E-04 (Scratch disk space usage for starting iteration = 1 MB) Any suggestion? Regards