Grasshopper

algorithmic modeling for Rhino

Dear all,

I have been working on a model containing a series of beams, forming a so-called meander pattern. In reality this panel is made by cutting a specific pattern out of a rigid material (like wood), which increases its flexibility. 

To model this behavior, I drew a simple form of the pattern and used the component Line To Beam to convert the lines of the pattern into solid Beam Segments. When I use the Analyze(Thl) component, the behavior seems very reasonable, as bending and torsion of the individual elements does occur. But as the panel is meant for larger deformations, I tried the Large Deformation Analysis (LaDeform) component, which should give more accurate results.

At first it seems all right, as the deformation of the panel as a whole does behave more accurately. However, the bending and torsion in the individual elements is no longer present. All lines in the deformed model are straight, and the beams as well. There is no bending or torsion visible within these parts. It is also not possible in BeamView to see the axial stresses, utilization or displacement.

I attached the models and a screenshot of the baked panel using the LaDeform component, as the final result without torsion and bending is better visible in here. 

Can someone please help me to fix this issue? Did I miss something?

Best, 

Andreas

Views: 697

Attachments:

Replies to This Discussion

Hello Andreas,

the 'LaDeform'-component is for form-finding only and returns no cross section forces. See the manual for details.

In Karamba3d 1.3.0 there is the component 'Analysis Nonlinear WIP' which can deal with large displacements and returns cross section forces also (see attached example).

Best,

Clemens

Attachments:
Hi Clemens, thank you! This solution is indeed the one I was looking for.

But I got a new question about this setup. If I now increase the load, the curvature of the panel does not seem to behave as expected. You might expect that with a higher load or lower thickness, the deformation would get larger as well, but somehow this is not showing correctly.
Has this to do with the fact that the Analysis Nonlinear WIP is still work in progress?

And lastly I would like to share an observation: the axial stresses being displayed in the model are the highest at the segments subjected to bending. However, in the physical model I made, the highest stresses seem to occur at the segments subjected to torsion and not the segments to bending. Because precisely through the torsion, the panel obtains its flexibility.

Best,
Andreas

Hi Andreas,

the 'Info'-output of the 'AnalyzeNonLin'-component contains information regarding the solution process and whether a state of equilibrium for a certain load-increment was found. The output 'Lambdas' provides all load-factors for which a converged solution could be achieved.

Maybe you have to increase the maximum number of iterations per load-step or choose a larger number of load-steps to get to higher load-levels.

The component is still work in progress. The convergence characteristics of the procedure will be improved in future releases.

The stress display for beams shows only normal stresses and no shear stresses due to shear forces or torsion moments. These are however covered in the design procedure according to EC3 or when you use the 'Utilization of Elements'-component.

Best,

Clemens

RSS

About

Translate

Search

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service