Any known issues with HS or HFEA and Windows 7?

[dlj2_msfc]

I was recently updated to Windows 7 and since then I have not been able to run HyperFEA.  I have verified the nastran_submit in the HFEA folder is pointing to the right location, but it still gives an error that is associated with not running the next iteration of nastran.

HS v5.8.11
HFEA v 1.1.6
FEMAP w/NX NASTRAN v 10.1.1

[Phil]

I believe the problem is that you are running:

FEMAP w/NX NASTRAN v 10.1.1

This version of Nastran does not allow you to run nastran from the command line.  You must go through the FEMAP analysis.

One thing that you can try is open the _i02.F06 file for you project and look for an error message.  It will probably say something like:

     THE BULK DATA SECTION OF THE INPUT DATA HAS BEEN CHANGED OR
     WAS CREATED ON A DIFFERENT COMPUTER.

In order to run HyperFEA, you will need access to a full command line version of either MSC, NX or NEi Nastran.

Please let me know if this is the problem.

[dlj2_msfc]

I have been running this setup for a while now with HFEA.  I call the nastran.exe directly.  I think I have found the issue.  When I pre-process the FEM, the _i00.pm1 file changes two of the material id's without changing the material id in the properties.  The _i02.f06 does give a message, but it is this:

"USER FATAL MESSAGE 311 (IFPDRV)
     NON-UNIQUE FIELD 2 ON BULK DATA ENTRY MAT1         220           SORTED ENTRY COUNT =      18117
     User information:
     Sorted Bulk Data entry type indicated must have a unique integer
     in field 2. An error in the input has been detected. Two or more
     grid points, elements,  or properties have the same id.  (For example,
     you may not use two GRID  entries with the GID=2 in the same run.)"

I am working on resetting the materials in FEMAP, and see if that fixes the material re-designation during pre-processing.

[Phil]

I'm glad you found the issue. 

In house we have laid out some pretty detailed plans for updating the FEM preprocessing that should eliminate these sort of conflicts in the future.  Basically what we should have is a way to let the user analyze any single component. group or components, or full model and immediately be able to submit the FEA to see the changes.  This is in contrast to the current method where you must analyze the entire model in order to submit the FEA.

I anticipate development of this capability in the spring of 2011.