Is there any plan to add dynamic analysis to prepomax? I normally use FEA for modal analysis and propose structure modification to avoid resonance vibration problem. However, it would be nice if we could predict dynamic stress from measured vibration data as well in some cases which natural frequency could not be shifted away from excitation sources with sufficient margin. With predicted dynamic stress, it can be compared against S-N curve of material for fatigue failure analysis. Also, with proposed structure modification, dynamic stress and vibration can be predicted to see if any improvement compared to the existing ones.
With current version, can we edit keywords to do above analysis? Any guidance or suggestion would be highly appreciated.
Unfortunately, random vibration analyses (as well as response spectrum ones) are not supported by CalculiX. When it comes to linear dynamics analyses, only transient modal dynamic and harmonic (steady-state) vibration simulations can be performed: CalculiX: Overview of the finite element capabilities of CalculiX Version 2.18
In fact, steady-state dynamic analyses are one of the most requested features for PrePoMax. Currently, you could use keywords but it would be quite complicated. It will be somewhat easier if you apply load in form of a concentrated force acting at a reference point connected to the rest of the model with rigid body constraint.
However, it seems that the results from this type of analysis can’t be displayed in PrePoMax (likely because of their nature - they are presented in the frequency domain:
Yes because it’s a mode-based procedure. A preceding Frequency step with storage enabled has to be included but this can be defined without the need to use keywords.
It would be a nice feature to have. I have been using the steady-state dynamics procedure for some time, and it would be nice to at least see the results in PrePoMax for more detailed post-processing.
The next step will be the support for explicit analysis.
If there are result files that are currently not supported by PrePoMax please let me know. Adding support for result files takes much less work than adding the UI for the preparation of the unsupported procedures.
I have tried to make some progress on this topic. First of all, I could not get the analysis to finish from PrePoMax. Using ccx 2.18 and Pastix solver gave me an error. I changed the solver to Spooles, and it was a little better. The analysis terminates with the bottom monitor message, but actually, the .frd file is about 14 MB in size and contains only a part of the solution.
Then I changed the version to 2.19. Run the analysis from PrePoMax. Using the default solver (Pastix) it now runs till the end. But not always. Sometimes the analysis terminates with an error. Running the same version from the command line also works most of the time but not always. When it does not work, I get the error about POSIX WinThreads.
Then I changed the version to 2.20. Run the analysis from PrePoMax. Using the default solver (Pastix) it now runs till the end. But only sometimes. Most times, the analysis terminates with an error. Running the same version from the command line also works sometimes. When it does not work, I get the error about POSIX WinThreads.
Can someone confirm this?
After successful analysis, the results could not be read due to NAN values in the imaginary fields. I have now changed the .frd reader to account for that. The results can now be read but are recognized and shown as ordinary static results.
I also observe such inconsistent behavior when running this input file using the three mentioned versions of CalculiX. The analysis either terminates at some stage (usually the same as in your case - “Composing the steady state response from the eigenmodes” or during factorization) with no error messages (I don’t get that pop-up error regarding POSIX WinThreads) or is completed successfully. However, the difference is that I run it in parallel when submitting from the command line (I set the OMP_NUM_THREADS variable using cmd previously).
When submitting from PrePoMax, I sometimes get the “Job finished” message but often the output is the same as in your case ("Perturbation parameter is active. Process elapsed time: …). Again with parallelization.
What’s interesting, I tried with the beamdy8.inp file from ccx test examples and it runs successfully in most cases but sometimes fails during factorization (in all the 3 versions of CalculiX).
I’m trying to report it on the CalculiX forum but each time I click “Create Topic”, I get error 404. I will keep trying, maybe it’s some temporary issue with that forum.
