sheared geometry with GUI

[louisepb]

Hi Yanie,

The warnings will not prevent an abaqus inp file from being created. They are warning that where the yarn has been fitted to the domain boundary the cross section is smaller than the original cross section of the yarn. At some point I may put in a check to see whether the yarn volume fraction is still valid but at the moment you may want to check that the cross sections at these points are viable.

When you create the CShearedVoxelMesh class it should have a type parameter. For a sheared textile this will be "CShearedVoxelMesh". I tried the last two lines of your code with this added and it works fine.

I think that the discontinuities that you can see are probably another result of the bug which I've just corrected. They don't appear when I run with the new version and the yarns are smooth. I will do a new release in the New Year.

Best regards,
Louise

[dyeyanie]

Hi Louise,

Ok, thanks a lot for your help. Seems like I need to wait for the new version then. Perhaps you won't take long for that.

Merry Christmas and Happy New Year to you and your members!

Regards,
yani

[maryam_sodagar]

Hi Yanie,

You didn't specify the weave pattern so I'm not sure if I've created exactly the same model as you.

The 'isolated' mesh that you describe is a function of the way that the dry fibre export separates the elements at either end of the domain. It takes the element centre point and checks if that is inside the domain. If so, then the whole element is assigned to that position. If you look at the opposite end of your model you should see the corresponding elements 'missing' at the other end. This maintains periodicity without splitting elements.

If you let me know the exact weave pattern that you have created then I will see if I can find the problem with the refine.

Best regards,
Louise

Louise,
Me again!
I have the same problem as Yanie had before.
How we can delete "isolated" mesh? What should we do to domain/ yarn geometry/etc.?
I understand what you mean about if centre point is inside the domain then the whole elements would assigned to that position. But how we can have a smooth surface with no isolated mesh in our yarn end surface?
I have not had this problem in Plain woven fabric but I have this problem in Satin fabric.
If I export dry fibre to abaqus and try to simulate a uniaxial load, the outcome would be incorrect?
Thank you so much,

isolated mesh.jpg (111.73 KiB) Viewed 876 times

[louisepb]

Hi Maryam,

The mesh is not actually 'isolated', it is simply created such that there are whole elements at the end of the yarns, which match the profile at the opposite end. If you have periodic boundary conditions then these will match up with no problem. The problem with smoothing the mesh to fit the domain exactly is that there is the potential to have small and/or badly shaped elements where the domain slices through the yarn.

The fact that you didn't have it in the plain fabric is just a function of where the centre points happen to fall.

As long as the profiles of the elements at either end of your yarns match then you should get correct results.

Best regards,
Louise

[maryam_sodagar]

Hi Louise,

Your answer was clear. Thank you so much

Best regards,