NASTRAN 95 Open Source 2

[rkiser]

Hello everyone

I have downloaded the source code for NASTRAN 95 from GitHub (The version uploaded by Dan (AeroDME).) Although it was targeted for Linux GNU, I was able to tweak it to compile under Microsoft Visual Studio 2017 with Intel FORTRAN Parallel Studio XE 2018, in a Windows 8.1 64 bit environment. It linked and would execute. I then made up a batch file to run it in console mode, inferring as best I could from reading the source code what it wanted as far as input file names and environment variables, etc.

Using my made-up batch file, I got it to run a little bit but it is experiencing errors where it is wanting to write past the end of some arrays. I didn't have a sample problem for it to solve and wouldn't know how or where to place the sample input anyway or what to name it. So, the fact that errors occur is just what you would expect.

I am a fairly experienced software developer and FEA user (COSMOS, ALGOR, NISA, LS-DYNA, and ANSYS) but with zero experience with NASTRAN, which seems to have a very extensive and elaborate file structure.

I was wondering if anyone had any success in getting it (Open Source NASTRAN 95)to run in a Windows environment. Also, I was wondering if anyone had any sort of simple tutorial on how to do a basic analysis (like a simple cantilever beam with a transverse end load). I have downloaded several NASTRAN PDFs with a lot of information on: Rigid Formats, DMAP formats, etc. but it was difficult (for me at least) to see how it all fits together. (Told me how to build a watch, when I just want to know what time it is - kind of thing.)

I was hoping to get some sort of a simple "Hello World"-like example running without errors. Then I could branch out and experiment to work my way up to less trivial problems.

Would anyone have anything to offer on this?

Thanks in advance for your help,

Bob Kiser

 

[FFB12]

I think you'd be OK with this

http://www.scc.kit.edu/scc/sw/msc/NasDirect/Nas_AT_LinearStatic.html

 

[rkiser]

Hello Mr. Fernandes

Thanks for this link. I did indeed find a test case for a simple cantilever beam. So now I have something to start with. The tutorial instructions are for MSC/NASTRAN and I'll now have to see how that is different from the procedure for NASTRAN 95. Hopefully, they will be similar enough that I can figure out how to put the right data into the right slots.

Best,

Bob Kiser

 

[dmapguru]

The poster in this link might help:

http://www.eng-tips.com/viewthread.cfm?qid=391194

and also try contacting Harry Schaeffer who has crossed this bridge already. This link will help to find him:

http://caelinux.com/CMS/index.php?option=com_kunena&func=view&catid=2&id=8257&Itemid=300025

 

[IDS]

pyNastran might be worth looking at:

Link

I have downloaded from there and also from the Nastran95 site. It's well down my "to-do" list at the moment, but if anyone has a look, and gets it running, I'd be interested.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[goeasyon]

Did you try to build with Mingw64 and gFortran?

https://welsim.com

 

[rkiser]

Hello goeasyon

No, I didn't. Used Visual Studio as noted above. It was not a heroic effort to get it to compile - fairly straightforward.

Rather, the problem I am having, as far as I can tell, is in getting the input file structure set up correctly. I don't think I am telling it (the code) the right info in the right way. I have been searching for some time now for some sort of tutorial or go-by on how to do this, but so far, no joy.

Best,

Bob Kiser

 

[goeasyon]

Hello Rkiser,

I am sorry for your experience in building/running NASTRAN, working on a legacy library is always frustrating.

If it is about the input script, NASTRAN package seems already give input examples at [URL unfurl="true"]https://github.com/nasa/NASTRAN-95/tree/master/inp[/url]. Have you checked this out?

https://welsim.com