I am using the shell section block in Karamba3D to visualise the stress on the mid section of a simple plate. I noticed that the block returns the correct numbers, but the plot shows the same numbers multiple by a factor of 100. Am I doing something wrong in the way I use the block? Thanks a lot for helping!
Screenshot and code attached.
thanks for your post!
This is a bug in karamba3D 188.8.131.52.0 .
I will try to fix that ASAP.
I could fix the bug.
You can download the improved version from here.
Thank you very much for your bug report!
thank you so much for fixing the bug! I wasn’t sure if it was a bug or if I was messing around with the units. Thank you very much for you quick reply and update!
sorry do you know how to install the latest WIP version on Mac? On the Rhino manager I can only the stable version, and not the most updated one.
Also, I am interested in retrieving the local stresses on shell elements (not the principal ones) as I am working with an orthotropic material and I would like to implement a different failure criteria than Tsai Wu. I can see that there is no direct component that plots these quantities (only the principal stress). However, I would assume that local stress must be calculated at some point, and then the principal stress is computed. Hence I was wondering:
- If I add a Python module and pass the model to it, can I access the local stress variables using the API scripting guide?
- Regarding the installation of this last, I was following this guide 2.7.1: Results Retrieval on Shells - Scripting Guide 2.2.0. However, I think the installation on Mac is different and I can’t locate the Karamba.gha file. Is there any guideline for API scripting for Mac?
Thank you very much for helping.
i dont know but it might be useful for you
you can combine Shell view and Shell forces with Ladybug tool which is called Spatial Heatmap to visualise mesh legends and its corresponding local forces on cross section.
Hi @onurgoztepe thanks a lot, that was super helpful!
@cp1 maybe I am wrong, but I believe there is a factor of 100 also on the Shell forces component (unless this was solved with the latest WIP release) when dealing with unit conversion. I would expect the forces to be in order or 0-2 kN and not 0-200.
Code of a square plate subjected to in plane forces attached.
thanks for the bug-report and sorry for the inconvenience! I will fix that conversion error at the ‘Shell Forces’-component a.s.a.p.
Regarding Karamba3D on the Mac: it gets installed via YAK. See here for the location: https://discourse.mcneel.com/t/yak-package-manager-installation-folder-mac/122656. The C#-API for the MAC is identical to that under Windows.
I fixed the bug. It will be published with the next service release (see here).
Hi @cp1 , I might be wrong but there might be another conversion error with units. Please see the model below. It is a simple square plate 10 mm x 10 mm x 1mm (thickness) loaded by a 2 kN lateral force. It is made of plywood with G=350 MPa.
If I calculate the shear deformation by hand as: Fh/(GA_s) with F=2kN, h=10 mm, G=350 MPa and A_s=10 mm^2, I obtain 57 mm. I Karamba3d, I obtain 0.6
Hope it helps.
Code and screenshot attached for reference.
Shear_plate.gh (64.8 KB)
thanks for the hint. I will check that.
Thanks a lot Clemens.
It might be a stupid question, but I was also wondering how I can install the latest WIP version on Mac. I can see where the YAK installer install the folders (thanks for the link) and I can see where to download the WIP version on Github. However, I can only see 3 msi files on GitHub (Releases · karamba3d/K3D_NightlyBuilds · GitHub) depending on the version of Rhino (5,6 &7). But these files can’t be opened on Mac, can they?
On the other hand, if I use the package manager, I can only access the latest stable version and not the WIP. Any help is appreciated and thanks for the great work.
for MAC YAK is the only posssibility to install Karamba3D. The msi-installers from Github work only for Windows.
I found the reason for the discrepancy. The term ‘Input Length Unit’ at the Settings-component is misleading. It sets the unit for the input of geometry objects (lines, meshes) only. In order to change the units at the other component inputs one needs to edit the karamba.ini-file (see “CustomUnits”).
Thus in your definition the plate thickness was 1cm instead of the assumed 1mm. When setting the thickness to 0.1cm the result comes close to the value of 2/(35*0.1) = 0.57cm (see here:
Shear_plate_cp.gh (75.1 KB)).
Thanks a lot Clemens, I see now, that makes a lot of sense! Appreciated your time in going through this.