Importing beam cross section

[sylwcios]

Hi Guys,

Can you give me some guidance of how to import my own cross section to build a beam model.

Thanks

 

[rob768]

The following works:

You need a drawing (in dxf or similar format) of your section, or create it within femap. Import it and create a boundary of the section. Then, create property, choose beam type. In the beam menu, choose "shape" and the last option is "general section". It then prompts you to choose a surface which should be your section. There's your beam with its properties.

greetings
rob

 

[rb1957]

or you could generate the section properties from the CAD program you're using to plot the section.

or create a surface in Femap, from the Xsection, and link to this as above (beam,shape,general,...)

another day in paradise, or is paradise one day closer ?

 

[sylwcios]

Thank you very much I imported my cross section than I used a boundary surface option and it works fine.

 

[Worldtraveller]

I want to bump this to ask if what I need is possible within FEMAP, or if I need to figure out how to write code to do it in the API.

I have figured out the above, but I have a tapering section (spar) that I would like to model, of possible, with tapered beam elements. Is there a way to put in a few sections and have FEMAP do a linear interpolation between those shapes?

 

[trainguy]

I'm not certain, but:

You could import solid geometry into FEMAP, as a step or parasolid file. Then, Geometry/Solid/Slice at the various cross-section locations. Then, generate a beam property for each portion and select the beginning and end surfaces using rob768's method.

The key tip here is the importing of a tapered solid, the ability to "slice" at various locations, and select those surfaces as needed.

tg

 

[Worldtraveller]

Thanks. That's more or less the process as I envisioned it too. That would be very time consuming and tedious. (Although I could at least write a code to extract the cross sections at set lengths from SW.

I guess it's either do all the work in SW, or do even more work in Femap.

 

[trainguy]

Doesn't it make more sense to use plate/shell elements?

tg

 

[Worldtraveller]

"Make more sense"....

Depends. Usually beam elements will give more accurate results than plate elements when modeling a beam, and those are usually more accurate than using solids.

As a general rule, the lower dimension element, the better.

 

[BlasMolero]

Dear Trainguy,
CBEAM/CBAR elements runs OK with solid cross sections, but with hollow sections (and of course with open & thin walled cross sections!!) the use of Shell elements is a must, you don't want to have surprises!!. You MUST create a 1-D beam model, of course, and use results as a reference, but Shell models give fully accurate results, not doubt at all. You can run linear & nonlinear analysis, as well as modal & dynamic frequency response analysis with a reasonable model size and not so long solution times, Shell CQUAD4 elements are the best!!.
Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48004 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/