test beams to t

[nerork]

hello to all and thank you in advance for support.

I have checked "paper and pen" a beam to t the limit states, so I have identified the loads for the most charged section and I have verified the latter regarding the admissible voltage. I'd like to go to the fem now (catea v5). I do static fem analysis and verify the loads in the most loaded section and take note (von misses stress) that what happened on the sheet coincides with the fem results. Now, however, note the loads, I would like to know how to carry out the verification of the section, that is, how/where do I "say"/"I tell him" that sigma_equivalent<=sigma_admissible for the chosen section?
Thanks again

 

[nerork]

This time I ask myself questions and I give answers,
from the theory it seems that the tensions calculated with the von mises method are those to be compared with the admissible tensions (see attached image. )

to the next

 

[marcofa]

So I understand that the cat's results are directly applicable without further coeffs looking at the colors of the sections related to the scale. correct?
Mar

 

[nerork]

It would seem to be, even if from what was obtained on paper the sigma von mises is about 25% greater than the equivalent calculated for the verification to the admissible tensions. therefore the von mises is more conservative. You should make some more paper/cathy comparisons. I don't have time right now, if you take care of her, you let me know, bye.

 

[marcofa]

one thing I noticed is that by honing mesh compared to those that catia considers by default often, but not always, you have a remarkable worsening of the results, greater than one percent predictable around 10%.
is it possible such a big waste?
I have not yet realized what affects the result and how much influence the mesh size affects the result.
often, in case of very complicated geometries it is difficult to compare paper/cathia, as the manual count is not always simple or correct.
Mar

 

[Fulvio Romano]

one thing I noticed is that by honing mesh compared to those that catia considers by default often, but not always, you have a remarkable worsening of the results, greater than one percent predictable around 10%.
is it possible such a big waste?

the discard can be big, but infitting mesh, the result should approach reality, not vice versa.
a mesh little dense, or however wrong, can increase the "mathematic hardness" of the structure. in practice the effect on tensions is that of a more rigid material.

 

[marcofa]

What is the optimal size of the piece?
What do you mean wrong? catia assigns size and sagitta by default, normally with ratio 1/8 - 1/10, and parabolic.
can you make a small summary of the most correct selection criteria?
Mar

 

[Fulvio Romano]

What is the optimal size of the piece?
What do you mean wrong? catia assigns size and sagitta by default, normally with ratio 1/8 - 1/10, and parabolic.
can you make a small summary of the most correct selection criteria?
Mar

I hope you are not one of those who claim to graduate in engineering studying from wikipedia...discussed, but lately there are several on the forum.

the optimal size of the mesh depends on geometry, the presence of edges, what you are looking for. If I could make a sun, no one would pay me to solve this kind of problem. It would suffice a cra@ata version of catia, my "sunto", and to walk the engineers.