shell 2d or 3d...differences x analysis fem

[stef_design]

Hello, everyone, I have a curiosity to undergo.
is it possible to do a fem analysis (sollecitations and deformations) of a surface (i.e. the only external skin)?
what differences are there compared to the same surface but with volume?

For example: I have a cover. once I have the characteristics of the material, is it the same thing to consider the outer surface of the cover (the shell) and the cover as a 3d model (so the thick shell)?
Thank you.

 

[sail]

Hello, everyone, I have a curiosity to undergo.
is it possible to do a fem analysis (sollecitations and deformations) of a surface (i.e. the only external skin)?
what differences are there compared to the same surface but with volume?

For example: I have a cover. once I have the characteristics of the material, is it the same thing to consider the outer surface of the cover (the shell) and the cover as a 3d model (so the thick shell)?
Thank you.

in general shell elements do not take into account cutting (depending on the solutor, type of element, etc.)

par contre, if you use a solid model you have to put different points in the thickness to model physics, then the analysis will be more complex, more honest in terms of hw, etc.

It is up to the strutturist to understand when the simplification introduced by the shells is fine and when not.

 

[maxstralis]

Hello ... I present myself I am maximum and I take care of mechanical constructions, I have never done a structural analysis and I would like to get on it ., I ask on this post of info premitting the ignorance of the program

What's the fem program, which is a special program? ?

- to do the fem analysis, must make the design in 3d and import it?? thanks to all those who will respond in advance

 

[Onda]

Hello, everyone, I have a curiosity to undergo.
is it possible to do a fem analysis (sollecitations and deformations) of a surface (i.e. the only external skin)?
what differences are there compared to the same surface but with volume?

For example: I have a cover. once I have the characteristics of the material, is it the same thing to consider the outer surface of the cover (the shell) and the cover as a 3d model (so the thick shell)?
Thank you.

normally when you have thin wall structures (with a much smaller size than the other two) model with shell elements. then it starts from a surface (usually the mid surface, but the upper or lower surfaces are often used).
working with solid elements for models of this type involves having to make very thick elements, to respect the aspect ratio between the various sides of solid elements.
Besides, unless you put more elements in the thickness, you risk mislead the pattern. Obviously putting more elements in the thickness of a thin wall structure entails a huge calculation power expense.
In general for thin and extended structures, shells are much more efficient and more accurate.
I don't understand what sail means by saying that shells don't take into account cutting.

normally all shell elements have a mathematical formulation to take into account cutting out of the plane (for that in the plane have no problem). Of course, it is necessary to respect parameters, which are typically related to the fact that a model size should be much lower than the other two.
modeling with the shell too thick structure obviously loses the cutting trend within the structure itself as the cut in the shell is seen as a constant value and you do not have a cutting trend in
direction of thickness. This is valid until the shell is thin compared to its other two dimensions.
basic all thin wall structures are modeled only and exclusively with shell models, and with these therefore they are made airplanes, boats, antennas etc. the composite is typically shaped with shell elements, even when sandwiches.
modeling with shells entails much more work on the part of the operator, since starting from the 3d model it is necessary to simplify it, to extract the working surfaces (face of the body or mid surface), to combine all angles and edges, then to create mesh, to apply constraints, loads and properties' and finally to generate the solution.
compared to the programs that work with automatic mesh solids, there's an abyss.

 

[stef_design]

modeling with shells entails much more work on the part of the operator, since starting from the 3d model it is necessary to simplify it, to extract the working surfaces (face of the body or mid surface), to combine all angles and edges, then to create mesh, to apply constraints, loads and properties' and finally to generate the solution.

Let's put a shell of a chair.
How do I do that? that is the seat, the back and the legs are solids full. for example legs are cylinders...what extraggo

compared to the programs that work with automatic mesh solids, there's an abyss.

What does that mean? that the programs that generate automatic mesh return to the little precise results

Thank you:

 

[Onda]

modeling depends on the material the chair and its shape is made of.
if it were with tubular steel legs, shell elements for everything.
with full wooden legs, I would use beam elements for legs and shell for seat and back.

There is no unique answer, modeling depends on the thickness and morphology of the component.

I do not consider the lower automatic fems of the "manual" fems. I consider them far superior for what they were created, mechanical elements, with isotropic materials.
When we talk about composite materials, thin-wall models and other strange situations, the fems integrated in the cad show limits.

I meant the abyss in the difficulty of modeling.