Stress analysis on mesh - KARAMBA

Replies to This Discussion

Permalink Reply by Karamba3D on October 27, 2011 at 2:41pm

Hi alek,

increase the size of your cross sections in order to get a better view. Use the ModifyBeam-component for that.

Although the elements of your dome are mainly stressed by normal forces, bending moments arise in the vicinity of loads or supports. They cause stress peaks. Due to the fact that the color range runs from the minimum to the maximum stress most of the structure may turn out white due to that. Use the slider 'Render color margin' of the ModelView-component  under the Render-section to limit the color range to a certain percentage of min/max-stress. Then colors should start to appear on the rest of the structure. Use the default color range (red=compression, blue=tension) at first - so do not feed anything into the color-plug of ModelView.

I hope that helps.

Clemens

Permalink Reply by alek on October 27, 2011 at 4:03pm

Hi Clemens!!

 

Tks for the reply!

 

Im afraid I need a little more insight, what I really need is an stress analysis on the surface of the dome, not on the lattice frame... kind of what happens in the image found attached to this reply.Do you think it is possible?

 

Tks again!!

Permalink Reply by Karamba3D on October 28, 2011 at 1:37am

If you want to calculate the exact stresses on your shell you need to use shell elements.

Karamba currently has only beam and truss elements. Using these to generate a triangular mesh based on an arbitrary shell geometry can give you a qualitative picture at best.

Clemens

Permalink Reply by Chris Ingram on November 8, 2011 at 4:54am

Hello,

I am also trying to perform stress analysis on a surface also. What do you mean when you say shell elements? There are no obvious shell elements in the Karamba component menus...?

Permalink Reply by Karamba3D on November 8, 2011 at 5:39am

shell elements can take in plane loads - like a shear wall - and out of plane loads  - like a plate.

Currently Karamba has no shell elements, only beams (with bending stiffness) and trusses (without bending stiffness).

In case you want to use Karamba for analyzing a surface structure you need to dissolve it into a triangular mesh of beams (like a grid shell). Doing so will lead to overlapping cross sections and the beam model will locally behave differently from a real shell. On a global level, e.g. for calculating the maximum deflection it should be possible to get a good approximation of a continuous shell though.

 

Permalink Reply by Chris Ingram on November 8, 2011 at 5:50am

Thankyou for your quick response,

Does this mean that i could simulate and analyse bending of a sheet surface (such as plywood) in Karamba if I was to set it up and think of it as a very thin wide beam, as a solid object?

...although I guess the shear properties and twisting characteristics may be harder to simulate than the deflection of a beam. Would it be within the scope of the current version of karamba to analyse the bending/twisting and shear stresses on a thin, sheet like solid object? 

Permalink Reply by Karamba3D on November 8, 2011 at 7:39am

It depends on what you want to get out of your analysis. The basic kinematic assumption of beams is, that points that are on a common plane at right angle to the beam axis before deformation are also on a common plane after deformation. This means that beams can only give linear strain distributions over their height and width. The stress distribution in the vicinity of a load is therefore out of the scope of beam analysis.

In order that an object can be viewed as a beam it has to have a certain minimum extension: as a rule of thumb the distance between points of zero moments (e.g. span of a simply supported beam) has to be at least double the height of the cross section.

Karamba would simulate the twisting and shear behavior of a thin sheet like solid object correctly - globally. However you won't get bending moments e.g. at right angle to the longitudinal axis as you would if you did a calculation with plate elements.

The stresses/strains/utilization displayed in Karamba are only the ones due to normal force and bending.

 

 

Permalink Reply by Chi-X on May 19, 2014 at 5:14am

Hi Clements,

I am looking at the shell analysis on Karamba. All works on surface type mesh, see attached picture below

However i would like to try to do the same study on a 3D mesh, such as for example the perforated box shown in the picture below...

I have set gravity as main weight and select the base of the mesh as anchor points. However my results seems not right...

Any suggestion on this matter would be super helpful!

I have attached for u also the GH file extension....

Thanks lot 

Chiara

Permalink Reply by Chi-X on May 19, 2014 at 5:35am

GH file extension

Attachments:

• box_structural analysis 02.gh, 21 KB

Permalink Reply by Chi-X on May 19, 2014 at 5:37am

rhino file

Attachments:

• structural analysis box.3dm, 654 KB

Permalink Reply by Karamba3D on May 19, 2014 at 10:05am

Dear Chi-X,

try to reduce the size (in terms of number of elements) and complexity of your definition while applying the divide an conquer strategy to find the problem.

Best,

Clemens

Permalink Reply by Chi-X on May 19, 2014 at 10:52am

Hi Clements,

thank you i will try to amend the model and see what happens!

One question, sorry for this one more but I am new to Karamba :)

When you refer to ' the divide and conquer strategy' what do you refer to?

Thank you very much for your quick response and help!

Chiara

• ‹ Previous
• 1
• 2
• Next ›
• Page