Differences between Nastran and Femap

[Vincentlaval]

Hello
I am working on MSC.VisualNastran for Windows 2001 Femap 8.0, and I'm trying to recover some results obtained with Nastran.
When I run an analysis with MSC.VN4W with exactly the same model than with Nastran, I obtain significant differences on all my outputs.
Are they some basics operations to do when sarting an analysis with Femap?
Thanks

 

[SWComposites]

What version of Nastran are you trying to match with VN4W? Did you import the original model into VN4W? WHat outputs are you comparing, and what are the differences?

What is the configuration of the model are you analyzing? What are the loads and boundary conditions? What elements are used? What material types?

I would look at the following (in order of likely problem):
coordinate systems
materal properties
material property orientation
output coordinate systems
load directions
element types

 

[GregLocock]

I'd do a visual comparison of the two model files, using WinMerge or gdiff.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[dmacx]

Specifically, look at the .dat files and compare line by line using a file comparison utility.

 

[rb1957]

doesn't "with exactly the same model" at least impky that the same data set was used, with two different versions of NASTRAN ? then the difference isn't in the dataset but in the NASTRAN code ...

 

[ThomasH]

I don't understand what you mean by "exactly the same model"?

First you run N4W and get results, right?

Then you you say that you run Nastran, does that mean that you use the same data file and run it on a stand-alone Nastran or ?

When you run N4W you run the Nastran solver "inside" Femap so in both cases you "run Nastran". There might be different version but in my experience the results are the same.

Or are you comparing different versions in terms of MSC/NE/NX?

Regards

THomas

 

[Vincentlaval]

Hi all

Thanks for all your answers,

Some precisions about my question:

-The results that I use as my reference are from an old run in Nastran version 2001.0.1

-The only runs that I can start are under Femap 8.0b, I don't use the pre and post processors Nastran

-So my analysis are under Femap and I would like to find the same results than in Nastran

-I run an analysis with the same input .dat file than the analysis under Nastran was

-My .dat model contains:
CORD2C and CORD2R coordinates systems
CQUAD4,CTRIA,CBAR and CROD elements
Nodes
PCOMP,PSHELL,PBAR and PROD properties
Mat1 and Mat8 materials
SPCD,SPC, and SPC1 constraints
PLOAD4
FORCES

-I can observe significant differences in all my outputs of my .f06(forces, strains, displacements, forces of SPC,...)

-The only diffrences I obtain in my outputs .dat files are:
-the coordinate systems has been modified under Femap
-some unsignificants differences on the grid point definition (about 1E-15)
-the params are not the same:
For Femap
PARAM AUTOSPC YES
PARAM GRDPNT 0
PARAM K6ROT 10.
PARAM LSTRN 1
PARAM MAXRATIO1.5E+8
PARAM NOCOMPS 0
PARAM OGEOM NO
PARAM POST -1
PARAM SNORM 20.0
For Nastran
PARAM AUTOSPC YES
PARAM EPZERO 1.0E-6
PARAM GRDPNT 0
PARAM K6ROT 10.0
PARAM MAXRATIO1.5E8
PARAM NOCOMPS 0
PARAM POST -2
PARAM PRTMAXIMYES
PARAM SNORM 20.0

Maybe I forgot something to put in Femap Analysis Window when runing an analysis?

I am new with Femap, so I hope that my explanations are clear.

Thanks very much

Vincent