fibre-reinforced plastic

Hello there,

I´m pretty new at Karamba and i realy appreciate and like it.

At the moment I´m trying to generate some fibre-reinforced plastic for a freeform, that´s exposed to gravity. I would love to recieve the firbes arranged in number and direction, that they can cope with the loadcase and I would just have to glue them together with some kind of resin. Like in the linked Example http://www.achimmenges.net/?p=5561 .

I read in the Karmba 1.1.0 manual, that "In case you want to strengthen a structure with linear elements (e.g.fibres) align them with FF-lines to get the most e ective layout."

I tried to do so, but the solution seems a bit missleading. I guess i did a mistake somewhere. What do you guys think?

Thanks,
Dominik

PS. gh file attached

Attachments:

150805_Fibre_Sunpath_ShellCrossSec.gh, 454 KB

Replies to This Discussion

Permalink Reply by Karamba3D on August 6, 2015 at 12:31pm

Hello Dominik,

you could try this:

increase the density of the underlying mesh to make the curves smoother.
be more selective about the starting points of the force flow lines. I think the points at the supports would be a good choice or where concentrated loads act.
If the structures behavior is governed by gravity use the Z-direction as the force flow direction.

Best,

Clemens

Permalink Reply by Dominik Simon on August 12, 2015 at 9:44am

Hello Clemens,

thank you very much for your response. It looks good now.

I have another Question now :) I did a shell crossection optimization on a triangular shell wich is supported on the 3 corner points and the next closest two points of ech corner point. The shell is a bit asymetrical.

Now i get a unusual crossection height for one face. It´s way bigger then the other ones. For me it´s difficult to comprhend, and i suspect it´s somehow a mistake. What do you think?

Furthermore i would like to know if its possible to let the crossection "grow" in only one direction instead of puting out the original form as a middle layer? So that i would be able to put the original shell on the "ground" and during the optimization process nothing would tend to reach underneath that level?

Best regards,

Dominik Simon

Attachments:

Shell_optimize_crossec.gh, 44 KB

Permalink Reply by Karamba3D on August 13, 2015 at 12:26am

The excessive shell thickness at the support of is due to the clamped support condition there which leads to bending moments.

You could try to add short pieces of beams use only one support per edge. The purpose of the beam pieces is to distribute the support reactions so that a stress singularity is avoided.

In Karmba 1.1.0 it is not possible to define eccentric shell cross sections.

Best,

Clemens

Permalink Reply by Rasmus Holst on August 13, 2015 at 1:35am

Fun project :)

- I quess it would be possible to divide the shell into "slices" of shells/meshes and give these different thickness in whatever direction and style wanted?

I can't remember what egde conditions it is possible to set for shells in Karamba?

Permalink Reply by Karamba3D on August 20, 2015 at 12:38am

it is possible to set different thicknesses for the elements of a shell. In Karamba 1.1.0 an isotropic material is assumed - therefore it is not possible to define different thicknesses for different directions. I think what would be needed here is the possibility to define composite shell sections.

For vertexes of shells one can define six displacement constraints. Line supports result from fixing multiple, consecutive points.

Best,

Clemens

Permalink Reply by Dominik Simon on September 30, 2015 at 10:59am

Thanks again, and sorry for the late Response :)

RSS