Dear experts,
I´m trying to include angular profiles in my Karamba 3D model. In this old 2017 post a procedure is indicated… However, I wonder if there is a more elegant and simple way to do it?
Thanks a lot!
Dear @Vigardo,
there are still no angular cross sections available in Karamba3D 1.3.3.
For angles shear center and centroid do not coincide and the principal directions of the moments of inertia are inclined with respect sides of the angle.
You could e.g. specify a dummy-I-profile and assign to it the properties of an angle via a ‘Modify Cross Section’-component.
–Clemens
Hi Clemens, thanks for your response!
Then, as long as the truss elements of my lattice structure are not loaded in shear or bending the deformations and utilization results should be ok but buckling and bending should be wrong, am I right?
If I understood you well, in any truss made of angular profiles it is as simple as just providing the appropriate cross-section Area (A) as well as the Iyy and Izz moments of inertia (e.g. computed using this online calculator) to the Modify Cross Section component, right?
One question more about element orientation. Should I first use the Z-ori input of the Line-to-beam component to align the local z-axis of the element (i.e. the u-axis in the figure below) and then introduce Iuu and Ivv into Izz and Iyy inputs of ModifyCrossSec? Please, correct me if I´m wrong (I´m not an expert 
yet)
The last question, would I use a dummy-Trapezoidal-profile if both sides of my angles are equal?
Hi Vigardo,
if there is only normal force in the member you will get correct results for the angle; also for buckling.
Besides A, Iyy and Izz you need to provide values for the resisting moments Wx, Wy.
The cross section properties in Karamba3D refer to the principal axes. These are rotated in case of angles. Orientate the principal axes of the angles using Z-ori.
–Clemens
Hi Clemens, thanks for your kind answer!
Just a few minor details more:
-
Would I employ a dummy-Trapezoidal-profile instead of a dummy-I-profile?
-
Which Cross-Sec outputs correspond to Wy and Wz?
-
Sorry for my ignorance, but are the resisting moments Wy and Wz really needed if only normal forces are present? (No eccentricities as well, I forgot to say)
Thanks a lot for your help!
Hi @vigardo,
ad 1.) you can use either of them.
ad 2.) Wely+, Wely-, Welz+, Welz- for elastic cross section design, Wply, Wplz for plastic cross seciton design
ad 3.) In that case the cross section area should be sufficient.
–Clemens
1 Like
I just wanted to share here this table with the data required to generate custom cross-section profiles (e.g. angles of equal or different legs, as well as, I, H, tubes, and others).
It is in Spanish, perhaps somebody knows one in English and wants to share it here .
Clemens (@karamba3d) - following up on this discussion, and using the Karamba section property table, can you point me to a definition of all the properties included? Most are clear, but it would be useful to know what is collected in the columns marked:
- zs (cm)
- i_y (cm)
- i_z (cm)
- Product (-)
Hi @david10,
- ‘zs’ is the vertical distance of the center of gravity from the topmost point of a cross section,
- ‘i_y’ = Sqrt(Iyy/A) is the radius of gyration about the y-axis
- ‘i_z’ = Sqrt(Izz/A) is the radius of gyration about the z-axis
- ‘Product’ is an integer number which indicates the production process of the cross section. From this the imperfection coefficients for buckling according to EC3 derive: 0: cold formed, 1: hot finished, 2: welded, 3: rolled, otherwise: undefined.
– Clemens
Many thanks!
Hi! I´ve just realized that there is a new “Cross Section Properties” component in Karamba 2.2.0 that would serve to create angular cross-sections. Am I right?
Can this new component be employed to generate fully valid angular profile properties for Analyse I and II? I mean, is it valid even if normal, bending, shear and torsional loads are present?
If so, I don´t know where to plug-in “alpha” output, and how to obtain Ay, Az, Cw, Wt and lg for the Modify CroSec component. Would you help me?
TestCroSecs1.gh (55.5 KB)
How well would this cross-sections perform in EC3 related components (e.g. OptiCroSec or Utilization)?
Thanks in advance for your help!
Hi @Vigardo,
the new ‘Cross Section Properties’-component is for calculating cross section properties of arbitrary thick-walled cross sections.
At the moment (Karamba3D version 2.2.0) calculation and design is still limited to symmetric cross sections. Anglular sections are on the ToDo-list.
–Clemens
Hello. For more general loading conditions of the C section, would a shell be a more reliable option?
Hello @valbe363,
in Karamba3D 2.2.0 C-sections are not included as beam cross sections. They need special care with respect to cross section design and displacement calculation since the center of gravity and the shear center do not coincide.
The ‘Cross Section Properties’-component does not output Ipp, Cw and Wt. These need to be provided by the user.
You can do a shell calculation for result comparisons. For large structures this approach is however not feasible since one needs a large amount of elements and results retrieval is more complicated as compared to beams.
– Clemens
Ok, thank you for reply. I will check what can I do with shells. I am not worried about being complicated, as much as the results may work : )
Hi Clemens,
to find Wt for thin-walled cross sections. Is this the formula
Wt = Ipp/tmax
(with tmax: thickest flange thickness of the cross-section)?
Thank you,
Yes, it is. Do not forget the web thickness.
– Clemens
1 Like
Hi Clemens,
Sorry to come back at this, but I am still figuring out what the best way would be to introduce a library of cold-formed C-profiles into Karamba. If I try it with a .csv-file, could you give me a base file to start with, specifically for this non-closed, thin-walled members?
If I start with a regular profile, I miss for example the Ipp-value to fill in.
Or are there other ways to do it?
Many thanks
Hi Pieter,
please find attached a GH definition that produces a cross section table template:
CrossSectionTableTemplateCreation.gh (10.6 KB). I hope you find it helpful.
Karamba3D 3.0 takes into account the St. Venant part of torsional resistance (Ipp) only. Warping resistance is not yet included.
Be aware that K3D currently assumes that the principal axis of the moment of inertia of the cross section are aligned with the beams local coordinate system.
Best,
Clemens