Assembly Project Feature Issues

[James.rogers@nasa.gov]

I have been using the "Assembly Projects" feature to perform some analysis in a timely fashion. However, after I reimport the results to the original hdb file and use "Update Nastran" to generate a new bdf/f06, when I select these new files for my project's run decks I run into an issue. When using these new files with the "Assembly Projects" feature, I get the same results as  my first attempt. In this sense, I am unable to progress past initial iteration when using this feature. Am I missing something trivial?
Thanks!

[Stephen]

Hi James,

When you choose the "Update Nastran" option, it will generate new BDFs with the sized properties as you have seen. However, it does not automatically execute Nastran to generate new results files (OP2); it just copies the latest ones (which are the original, if you have not iterated with FEA). So if you simply size and then update the BDF, but use the original OP2 files, re-sizing will generate the same results.

You can either manually resubmit these updated rundecks to Nastran, or use the HyperFEA tool to automate this process.

Thanks,
Stephen

[James.rogers@nasa.gov]

To be clear, after I generate the new bdf I have been solving with nastran to generate the new f06. I then select these files (the new bdf and new f06) as the new run decks. I am then unable to progress past the first iteration.

The bit about the op2 file interests me though. Do I need to have the updated op2/xbd file in the same destination as the run decks as well to progress? Currently I have been isolating the bdf and f06 files within a folder for bookkeeping purposes.

[Stephen]

James,

Thanks for clarifying. The location of the F06 / OP2 / BDF / XDB files do not matter, as long as you are pointing HyperSizer to their location correctly when using the "select run decks" form.

After you've changed which rundecks you are pointed at, it will prompt you whether or not you want to re-import the model. I am assuming you are (correctly) choosing "no"? If so, are you clicking the "Re-Import Element Loads" button on the setup form afterwards? This may be required if you point to a new results file (or replace the old one) because we cache the element forces in a more efficient temporary file and this needs to be replaced when the results file changes. I am not sure if the software "knows" to re-import forces when you point to a new results file.

Thanks,
Stephen

[James.rogers@nasa.gov]

I am actually choosing "yes" when it asks to re-import the model. I was under the assumption that this was necessary. I can try the "no" option, re-import the element loads, and see if the issue persists.

[James.rogers@nasa.gov]

Choosing the "No" option does not seem to alter the results. Progress past the first iteration continues to be an issue.

[Stephen]

James,

I am reaching out via email for more information.

Thanks,
Stephen

[Stephen]

For reference, the issue being encountered here was because when creating new part databases with the updated FEM results imported, the user was choosing identical names for each project.

When importing a project, HyperSizer creates temporary files based on project name that contain (among other things) a version of the FEA forces. If another project is made with the same name, the software will think that the forces have already been imported because it will find the previous version's files.

The solution to prevent this from reoccurring is to either delete these temporary files before creating new databases (located in C:\HyperSizer Data\Temp or D:\HyperSizer Data\Temp), or give the next iteration's projects different names from the first.