Introduction

In order to learn more about composites and FEA in CAD and CAE environment, I'll be following a course created by Nader G. Zamani, professor of mechanical engineering at University of Windsor, Canada.

I would like to thank him for the time and effort spent to produce all his content.

Exercises

In the following sections, only notes and results about the tutorials are going to be shown. For further information, please follow the tutorials linked on each exercise.

A key option of the Imported Composite Property dialog box is the symmetry. If symmetry is selected, the applied loads are going to be executed in the middle plane of the carbon sheet. If the symmetrical tick box is not selected, the loads are going to be applied as if all the material was above or below the action plane, producing a curvature on the sheet. All these results are with the displacement function after computing the case.

After that, we can take a look at the different stress tensor components. The interesting visualizations are obtained by selecting the global and local axis systems or rosettes and then selecting the component.

[Static case solution>(Generate Image)>Stress full tensor component (nodal values)>Ok and click again>More>Axis System ...>Type: User>Name: select axis system>Ok>Component:C11/C22/C12]

If the global axis system is selected, the C11, C22 and C12 correspond to the global X, Y and XY components.

If the local axis system (aligned with the fiber) is selected, the C11, C22 and C12 correspond to the component on the fiber direction, the component on the transverse fiber direction and the shear component, which, all in all, are the principal orthotropy axis.

All three failure criteria is met since all the values are below 1.

This time, the laminate is symmetric, therefore, the composite terminology matrix B is null, in-plane deformation and bending are decoupled and there should not be any out-of-plane deformation.

Exactly the same deformation is obtained, however, the stress representation is not showing up as in the previous exercise when the global or user reference is selected. We might pass on that detail. The results are indeed the same as with the thicker single ply laminate.

In conclusion, if several plies of the exact same fiber lamina are going to be used, a single ply with the total thickness is going to give the same result and it will be less complex to configure.

There is something to take into account when performing the structural analysis. This exercise is just one half of the model. This approach has been done due to the symmetry of the problem. There is no need to compute the whole part if this is done correctly, therefore, reducing computational costs. The key is to constrain the part appropriately, in this case, clamping the back edge and making restrains to the left edge (mid line of the part), so that it doesn't have translation on hte Y axis and no rotations on the X and Z axis (the mid line must be on the XZ plane and the immediate surface to that plane must maintain the same angle on both sides).

The analysis is computed and the expected results are achieved.

After that, the mesh is defined (one side at a time). Once done, we take a closer look around the cylinder surface to check if all the cell directions. That is achieved by computing only the mesh, then generating an image from Properties (tree) and finally selecting Composite angle symbol. There are no discontinuities.

After that, we can compute the whole analysis and see the results. The deformation is as expected and the Von Mises stress match with those from the original reference. Notice that the outer layer has a tad higher values of stress, this is due to the cured thickness of the material is set to be 2.54 mm (0.1 in) which is oddly thick. The thinner the plies, the less diference in stress will be between them.