Flipped cross sections - Inconsistency between beam view & cross sections input/output & beam results/utilizations

Grasshopper Karamba3D
1

Hi there,

We have been working with asymmetric (around local y-axis) cross sections in Karamba 1.3.2 and found some strange behaviour, specifically I-sections modified to become a T (the behaviour I am describing here is consistent for any asymmetric I-sections though).

Below you can see I am setting a simple cross section. The input has the T upside down (so web is top), whereas in the output of the cross section component, the T is flipped! In beam view, however, the cross section corresponds to my original input.

image
image1775×559 173 KB

Now looking at the results I see some similar inconsistencies. The beam view results and legend seem to correspond to the input that I have given (and are correct). The beam results, however, (min max stress and moments) are not.

image
image1703×550 133 KB

image
image908×335 21.3 KB

I have attached a simple script replicating the problem as well.
It seems to occur in Karamba 1.3.2 and 1.3.1. Is this a known problem? Has this been solved in 1.3.3?

cross section flip.gh (32.6 KB)

Would be great if you could help out!

  • Stijn
2

Hello @stijn.joosten,
thank you for your bug report! In the current version of Karamba3D (1.3.3 201012) the problem is mostly solved. The flanges appear on the correct sides and the minimum and maximum stresses are as expected. The only thing I noticed is, that the textual output of the cross section still wrong. This will be remedied in the next release.
– Clemens

3

Hi Clemens,

Thanks for your answer. Good to hear that it’s on the radar and has been solved (mostly) in 1.3.3. I have alerted my colleagues and advised to update.

Thanks!

Stijn