Same settings but Different results in two versions of Karamba

Grasshopper Karamba3D
,
1

Hi there!
I’m currently working with a script in Karamba 3.1.5 (build 0121.0), Rhino 7, but I’m getting highly inaccurate results for both beam and shell stress/strength calculations. Interestingly, when I run the same script in Karamba 1.3.3, the results appear normal and as expected.
I’ve attached screenshots and files showing the discrepancies between the two versions.
Could anyone help me understand what might be causing this issue?

K3D_1.3.3.gh (75.6 KB)
K3D_3.1.5.gh (91.2 KB)

image
image1079×534 170 KB

3

Hi Hanbing,

Thanks for sharing your files and screenshots. Changes in solver settings or default parameters between Karamba 1.3.3 and 3.1.5 may be causing the differences in results. I’d suggest double-checking that all input parameters, like loads and materials, are the same in both versions. Also, look into any mesh or solver settings updates in 3.1.5 that might affect accuracy. If the issue persists, try testing a simpler model to identify where the discrepancies begin.

4

I’ve tried several things:

  • A simple model in both versions is tested, and the stress/strength results are the same. So it may be a problem of inputs.
    image
    image671×645 48.3 KB
  • Check loads. If multiply the loads by 0.1, the stress/strength of Karamba 3.1.5 will be the same as that of version 1.3.3. But the input of loads should be the same.
  • Check units. It is not the problem.
  • Check the “Create Linear Element”. There is a difference; the BklLenY/Z/Lt are not -1 in version 3.1.5, even though the input of these three parameters is default. I don’t know if it matters.
    image
    image1607×447 127 KB

    image
    image1786×667 166 KB
5

Hi, version 1.3 does not regard the tensile strength of the concrete, so it assumes that the tensile strength is the same as its compressive strength. In version 3, the material concrete has a tensile strength 10% of its compressive strength (raw concrete material without steel reinforcement). This is the reason why the stress strength ratio shows 10x worse results.

6

During the design, what shall I do if I want to make the stress strength ratio properly (<100%)? One is to add more columns and rigid supports; two is to reduce the load; anything else may make a big difference for the stress strength ratio?

7

Is it possible to set up C20 concrete material with reinforcement in karamba?

8

You can simply create a custom material for concrete that considers the tensile strength of the reinforcement - ie either equal to or slightly less than the compressive strength

10

Many thanks. It’s solved.