hobbyist with stupid questions

[PerKr]

Being a "beginner" in FEA (I did receive some education in Pro/Mechanica and have been doing small work with Catia's GPS, but I have not been anywhere near a professional FEA code) I have a few questions.

The first question concerns Catia and the restraints (correct term?) available. I need to find out how the restraints can be used and what can not be achieved with what's available. Do you know of any site which provides this information (for free)?

The second question: For simple structures (let's say an aluminium bike frame or a sportscar space frame), are Catia and Pro/Mechanica powerful enough? What are the limitations of Catias built-in analysis capabilities (in a full-blown Catia P3 configuration)?

The third question: what FEA codes would you recommend when the need to take it one step further becomes apparent? Are there any open-source alternatives?

 

[usafeng]

Many of the FEA vendors offer demo versions for free or a nominal charge which would be a good start. These versions are fully functional but have size limits on the number of nodes.

 

[johnhors]

Perkr
R
egarding catia GPS see the recent thread thread727-155017

 

[johnhors]

Perkr

For open source check out the very excellent CalculiX solver.

 

[GregLocock]

There is little that is simple about either of your chosen examples.

I doubt you would get within 30% of the true stiffness of the sports car spaceframe, unless you do some correlation and refine your model.

Cheers

Greg Locock

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

 

[xerf]

The first question concerns Catia and the restraints (correct term?) available. I need to find out how the restraints can be used and what can not be achieved with what's available.

Catia V5 has an excellent documentation, did you check it ?

 

[PerKr]

Johnhors: indeed, calculix seems very interesting. If it will work with a windows configuration and there is a pre- and postprocessor included (it seems to me like this would be the case), this seems like a great alternative (although it doesn't seem to be able to import CAD geometry, but I don't think that's a problem that cannot be overcome). I will have to take a closer look later today. Have you had any experience with this software?

 

[johnhors]

Perkr

Although Calculix is developed in the Linux environment it has been very succesfully ported to Windows by Jeff Baylor at

http://www.bconverged.com/calculix/

It does come with a pre and post processor, although I only use the solver, for pre and post I use Roshaz instead. I actually find the solver to be faster than MSC/Nastran and Abaqus but requires more memory resources which restrict model size to about 125000 node models (in 3D solids with 3DOF's per node)

 

[GBor]

Perkr,

Not wanting to sound like a saleman, you can download the Roshaz pre- and post-processor in demo form from my site:

roshaz.borowskiengineering.com

As johnhors suggests, the calculix solver is very impressive, but the GUI is difficult to use in my opinion. Roshaz is powerful for the price and interfaces well with Calculix.

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group