Getting NaN values for material laminates with cores.

[Mohamed Hetta]

Hey guys,

Trying to create a laminate made out of 2-Orth. plies and an iso core results in NaN value for the properties. Currently using 3.46.0, I think this is a bug as it wasn't the case in earlier version.

[Rob McDonald]

3.46.0 introduced a new variant of lamination theory.  The new theory can calculate the 3d through-laminate material properties.

However, this means that you need to supply information for all directions of material properties (or use orthotropic with transverse isotropy to figure them out).

I suspect you have some material properties set to zero.  Before, they were ignored, but now they're probably resulting in a divide by zero somewhere.

If that isn't the case, please post an elementary example that I can use to debug.

Rob

[Mohamed Hetta]

It was my mistake actually. I was using an incorrect value for Poisson's ratio = 0.5 which led to a division by zero and hence the NaNs, so my mistake.

I did encounter a GUI bug though. Whenever I'd create any custom laminate, then I would assign that laminate to a shell property, all of the sudden you can't navigate with the mouse cursor anymore nor can you rename properties or anything, only resolves when I delete the shell property containing the laminate by navigating with the arrows, and sometimes it just crashes as well due to the same reason.

Just wanted to point that out.

Thanks, Rob

Mohamed