Error when reading Samcef DAT file

[A. SANZ]

Dears,

I would like to thank you for sharing these libraries.

I performed contact simulations using NX 11.0 pre-post processor and Samcef 2015 V.17.1 solver, using finite elements HEXA (8) for the meshing. I have all the results and now I would like to visualize the results using FeResPost software.

I’m trying to do it with the following excel flie C:\...\TESTSAT\COMEX\EX09\PostDesFacRandom.xls. After indicating the path to my DAT file (attached Bison_G3_80.dat), when I execute the ReadModel option (FemDbIds tab) the “Error when reading Samcef DAT file”  message appears, twice.

Regarding the LOG file generated (attached FeResPost_postDesFacRandom.log) by the excel error message appears referring to the line 96854 of the   Bison_G3_80.dat file. In that line of the Bison_G3_80.dat  file the following is written: GROUP 1 MAILLE NOM "3d_mesh(1)_3"  , I believe that this line is related with rigid link connectors, but not sure about it.

Even if I try to erase the lines defining the rigid links (being afraid of modifying my model)  the message “Error when reading Samcef DAT file” keep appearing and I do not know how to solve it. Finally, I’m not able to view my results.

When doing the scan of DES and FAC files, no error messages are generated (attached Bison_G3_80_as.des and Bison_G3_80_as.fac). However, when I try to extract the results (ResExtract tab) the following message appears: “Error when modifying Result’s reference coordinate system”.

I installed the latest version of FeResPost, 4.5.6, win64  in a computer running Windows 7 professional 64 bit.

Thanks in advance if you can give me some hints to solve the problem,

Regards,

Aliona Sanz

[FeResPost]

Hello Aliona,

Thank you for your interest in FeResPost.

I had a quick check on your problem. If you replace the keyword "GROUP" by "GROUPE", it fixes the problem. But there are other problems, and I will have to investigate them one by one. This might take some time...

but I also notice that you have been using directly the model exported by NX program. Instead, the recommendation is to save the model from Samcef bacon pre-processor with ".sauve banque ..." command. This might fix a lot of problems. I have no access to Samcef program or manuals right now, but if I remember well the sequence of commands looks like:

samcef ba Bison_G3_80 n 1

input

.sauve banque "banque.dat"

.sto 2

(You will have to read Samcef manuals for more information.)

Regards,

Renaud

[A. SANZ]

Thank you for your response.

I have tried changing "GROUP" by "GROUPE" as you said and some errors persist. At present, I do not have acces to NX software, but as soon as I get the chance I will try to save the results the way you propose.

Thank you again,

Aliona S.

[FeResPost]

Hello Aliona,

As I said, there are several problems in the reading of "dat" files directly exproted from NX. I will work on it, but this will take some time.

You have better chance of success by reading the dat file exported from Bacon pre-processor. Note that this is also a "dat" file, but hopefully formatted in a more acceptable way for reading by FeResPost. "Dat" files contain the finite element model definition. For the results you will use the "des" and "fac" files you have already produced.

I re-exported you model with Bacon ".sauve banque ..." command (version V 17.1-1  of Samcef), and it seems to work better. (See attached file.) I presume, it should work with more recent versions of Samcef as well.

Regards,

Renaud