Hmmm, this seems like it’s going to be harder than I realised! I’m trying model a sparse network of unidirectional tows to form a lattice to reinforce a plate (the arrangement of the lattice to be generated by an optimisation routine). It’s a similar situation to this diagram (from the ABAQUS manual):
So I’ll have lots of varying thicknesses elements, in fact when I have a full lattice almost every element will be a different thickness and number of layers to some of it’s neighbours. At this stage I’m only interested in linear elastic behaviour, so I don’t need to worry about the stresses within layers.
So my plan had been to:
- work out the number of tows and orientations of the plies in each element (each element will have to be in it’s own set)
- Perform the CLT calculation to work out the ABD matrix for each element
- Apply these to shell elements and solve
However it looks like I can’t do this approach in Calculix. I can apply a composite lay up directly to each element in Calculix. I can’t see this working though, what happens for example in the above diagram where ply 3 goes over ply 2? There’s going to be hundreds of areas such as that in my model.