Analysis I-beam Calculix CCX

[Cristobal Lopez]

Hi all,

I am currently working on a project about FEA with the open source program Calculix.

I am doing the analysis of an I-beam with one fix extreme and an applied load at the other free extreme of the beam with shell elements.

As boundary conditions I only stablish the fix extreme and the load at the top center point of the free extreme of the beam.

The distribution of the shear stress along the web of the beam looks as it should, a parabolic function with the values I algo got from the analytical analysis. The problem comes out when I do the analysis of the shear stress at the flanges of the beam. Doing the analysis based on the shear flow along the beam, the shear stress should be 0 at the extremes of the flanges and increase linearly until the center of the flange, but I only have that result when analyzing the beam at the extreme where the force is applied. In most part of the beam (all but the extremes) the shear stress along the whole flanges is almost 0 with any kind of increase of the value, as shown in the picture.

I think this problem may occur because the analysis does not carry propertly the stress along the length of the beam, but I haven't been able to solve the problem.

In order to make the flanges act as they would in real life, I have tried setting a wall at the free extreme of the beam closing it or several walls along the length of the beam to try to make it carry the stress, but it didn't work out.
Also I have tried doing coupling at the free extreme but I don't really know how to use coupling in Calculix and it hasn't worked out.


I was wondering if any of you who has enough experience with Calculix or who has done this case could help me solve this problem and have the proper shear stress behaviour at the flanges along the whole beam.

Thank you very much in advance,
Cristobal.

 

[whitwas]

Some ideas:

[ul]
[li]The picture shows the surface shear stress which is supposed to be zero, isn't it? What about inside?[/li]
[li]Are you sure it's wrong? The picture is consistent with what you predict "should be 0 at the extremes of the flanges and increase linearly until the center of the flange" because the color scale is too coarse to see any variation in the low stress regions.[/li]
[li]Are the elements connected properly? Check the deformed view to see if gaps open up or the flanges are rotating freely about their connection to the web. This can sometimes be an issue in CCX if you're using shell elements.[/li]
[li]Have you checked for mesh convergence? Whenever I see people ask "why are my results wrong?", this is by far the most common reason, even when it's qualitatively wrong.[/li]
[/ul]

 

[Cristobal Lopez]

I already found the problem!

I was looking for the shear stress with the SYZ for the web and for the flange, but to see the shear stress at the flange I have to look at the SXY stress, because its the shear stress in the horizontal plane...

Anyway thank you very much for replying whitwas! I also checked the things you told me and I think I have all of that correct.