Calculate_limit_axial_load in sections with origin outside the center of gravity

[MestreCarlos]

Hi,
Calculate_limit_axial_load does not take into account that the load origin may be outside the center of gravity, which would produce curvature and eccentric moments. Do you think it would be worth updating the function to include this functionality? I believe there is currently nothing that calculates the maximum axial load that a section can support at the load origin.

[talledodiego]

Hi @MestreCarlos. Thanks for pointing this out.
Actually the idea of maximum and minimum axial forces refer to the maximum and minimum axial load that the section can ever take or, saying it differently, is related to the case of uniform stress. This is independent from the reference system (therefore from the eccentricity moment due to being outside the center of gravity). This is graphically observed looking at the following interaction diagrams.

The blue domain is for the section when centered in $\left(0, 0\right)$, the orange domain is the same section with an eccentricity in z.
The value of n_min and n_max does not change since it represents the maximum and minimum axial load that comes from uniform stress distribution. I think that the implementation of this function should remain as is right now.

I think what you are referring, correct me if I am wrong, is that if we move horizontally from $\left(0, 0\right)$ in the orange domain we will fail for a smaller N respect the blue domain (see orange dots path in the following plot). This is absolutely true, and would correspond, in a non-translated section to having centered N with My=N*e (see blue dots path in the following plot).

I think that this value can already be retrieved by the interaction diagram and anyway it can come from manipulation of the results objects for interaction domain.

Perhaps we can improve the documentation even adding a usage example to show these concepts and how to get those numbers from result? Maybe we can also consider adding in the future a method in the interaction domain results for getting this value?

[MestreCarlos]

Thanks @talledodiego . That was exactly what I was referring to. In my mind, calculate_limit_axial_load should return the green points in the image for the translated section. But with your comment it’s clear to me now, and also that those values can be obtained from the N–M diagram. That would be sufficient for me.

The only thing is that I would make it a bit more explicit in the function’s docstring, so that there’s no doubt about it in the future.

[talledodiego]

I totally agree with you, and thanks for raising this issue, which allowed us to look further into it.

I suggest the following plan that can be pursued also with separate PRs, if you agree:

• Improve docstrings to make this clear. Possibly in the short term.
• Create a section in the theory docs specifically stating this aspect. Possibility in the short term.
• Eventually consider adding a usage example that shows how to extract the green points. Medium term.
• Eventually consider adding specific methods in the results objects to extract those green points. Medium term.

May this work?

[mortenengen]

Thanks for raising the issue and triggering this interesting discussion, @MestreCarlos. I support the plan that you have indicated, @talledodiego, which is also very much in line with features we are currently working on. Could you please create issues for the various points so that we can keep track of the progress?

[MestreCarlos]

I think the approach sounds good. Thanks to both of you