Keyword type: step
This procedure is used to determine eigenfrequencies and the corresponding eigenmodes of a structure. The frequency range of interest can be specified by entering its lower and upper value. However, internally only as many frequencies are calculated as requested in the first field beneath the *FREQUENCY keyword card. Accordingly, if the highest calculated frequency is smaller than the upper value of the requested frequency range, there is no guarantee that all eigenfrequencies within the requested range were calculated. If the PERTURBATION parameter is used in the *STEP card, the load active in the last *STATIC step, if any, will be taken as preload. Otherwise, no preload will be active.
SOLVER is the only parameter. It specifies which solver is used to perform a decomposition of the linear equation system. This decomposition is done only once. It is repeatedly used in the iterative procedure determining the eigenvalues. The following solvers can be selected:
Default is the SGI solver. If this solver is not installed, default is SPOOLES. If neither the SGI solver nor SPOOLES are installed, default is TAUCS. Finally, if neither the SGI solver, nor SPOOLES nor TAUCS are installed, no eigenvalue analysis can be performed.
The SGI solver is the fastest, but is is proprietary: if you own SGI
hardware you might have gotten the scientific software package as well, which
contains the SGI sparse system solver. SPOOLES is also very fast, but has no
out-of-core capability: the size of systems you can solve is limited by your
RAM memory. With 2GB of RAM you can solve up to 250,000 equations. TAUCS is
also good, but my experience is limited to the decomposition, which
only applies to positive definite systems. It has an out-of-core capability
and also offers a
decomposition, however, I was not able to run either of
them so far.
If the MATRIXSTORAGE option is used, the stiffness and mass matrices are stored in files jobname.sti and jobname.mas, respectively. These are ASCII files containing the nonzero entries (occasionally, they can be zero; however, none of the entries which are not listed are nonzero). Each line consists of two integers and one real: the row number, the column number and the corresponding value. The entries are listed column per column. Matrix storage only makes sense if no *BOUNDARY, *MPC and *EQUATION occur in the input deck, since these keyword cards reduce the system of equations. Furthermore, the original equation numbering must be kept. This can be obtained by specifying -DNORENUMBER on the CFLAGS in the Makefile.
For the iterative eigenvalue procedure the ARPACK solver [18] is used. The eigenfrequencies are always stored in file jobname.dat.
First line:
Example: *FREQUENCY 10
requests the calculation of the 10 lowest eigenfrequencies and corresponding eigenmodes.
Example files: beam8f, beamf.