NASTRAN-CATIA

[Pesog]

Hello everybody,

I'm a FEM user who works with FEMAP/PATRAN-NASTRAN. I have just discover that CATIA has a module that allows FEM resolution.

I wonder, which of these two solvers (NASTRAN or CATIA) gives more accurate results? In other words, which is better?

I assume that the advantage of CATIA could be less time consumed than NASTRAN, since the item to analysed/design is yet modelled.

Thank you

Pesog

 

[xerf]

If I were a FEM user, I would not compare a FEM software to a CAD software having some FEM related modules.

In fact, I would not compare any two FEM software without specifying the exact type of problem I want to solve.

In general, for a given problem to be analyzed using a FEM software, the major inacurracies are generated by the wrong understanding of FEM by the FEM users and by modeling mistakes rather than by the intrinsic software accuracy.

 

[cbrn]

Hi,
I perfectly agree with Xerf.
These are two different things.
The Catia FE module, (or, as other examples, the FE tool of Unigraphics NX, and SolidWorks' CosmosExpress which is by far on the lower-end side), is primarily intended for the designer who wants/needs to check his design in the early stages. In order to do that, of course, the designer must also have knowledge of mechanical calculations. However, an extremely "realistic" simulation is not really intended.
On the other hand, we have "specialistic" tools which are meant to be used by specialized people, the "calculation men". The work of these people is focused on simulating the reality as close as POSSIBLE, or more in general analyzing it with a deep knowledge of the simplifying assumptions (you can never simulate reality at 100%, except in some rare and simple cases...). There you will find extensive control about the meshing, the element formulations, the constitutive laws, the BC conditions, the interactions between components, etc etc...

Regards

 

[40818]

Wont say much, but its another pretty tool for people to use without having any knowledge about the basics. If you want a steak you go to a butcher, not a grocer.
I bet if you look in the user manual in Catia about that function, it will probably sum up the whole stress analysis in 2 paragraphs.
And any engineer worth his/her salt will preliminary size a part before designing it so there is no immediate need to go to FE to see if the idea is worth pusuing. I'm all for designers to work prelim calcs, but when they try to start shoving FEM plots under your nose saying that their design is wonderful when in reality is useless is something i can hardly wait for!!