Meshing options and cross section of the yarns with domain.

General discussion about TexGen.

Moderators: Martin, Developers

Post Reply
UOMO
Regular
Posts: 45
Joined: Tue Feb 24, 2015 10:02 am

Meshing options and cross section of the yarns with domain.

Post by UOMO »

Hi!
I'm kind of new user of TexGen...
I've been using it to generate geometries (with Matrix domain) and then mesh and export elements and nodes to use any FEA code (home code/commercial) ... However, I have some questions.
OBS: I'm also interested in the MATRIX DOMAIN, so, apparently, exporting the "Dry file textile" is not an option for me...

1st. The 8-noded bricks generated are always "horizontal-oriented" (See CurvedYarnHorizontal.jpg) Is there a way to resemble the curvatures of the yarns by applying rotations on the bricks? (see CurvedBricks.jpg).

2nd. The cross sections of the yarns are "step wise like" (See Cross-Section-Stepped.jpg). Is there a way to make them look more like (See CurvedBricks.jpg)?.

3rd. Does TexGen assign automatically local coordinate (material) system to the yarns? (Abaqus?).

4th. Which you consider that are the advantages of using a "voxel mesh" instead with a normal tetrahedral/structured hexa mesh? Do you know or have seen a study on how much the homogenized properties changes/maximum stress-strain concentrations?

Thank you very much!.
Attachments
Cross-Section-Stepped
Cross-Section-Stepped
Cross-Section-Stepped.jpg (36.38 KiB) Viewed 10263 times
CurvedYarnHorizontal
CurvedYarnHorizontal
CurvedYarnHorizontal.jpg (66.71 KiB) Viewed 10263 times
Curved-Bricks
Curved-Bricks
CurvedBricks.jpg (168.27 KiB) Viewed 10263 times
louisepb
Project Leader
Posts: 998
Joined: Tue Dec 08, 2009 2:27 pm
Location: Nottingham

Re: Meshing options and cross section of the yarns with doma

Post by louisepb »

Hi,

There is currently no way of adapting the voxel mesh (applying rotations etc). You need to select the voxel size carefully so that it picks up all of the detail in your textile (for example small enough to provide continuity in narrow binder yarns). Ideally you should run a sensitivity study with different size meshes to find the optimum voxel size. I have attached a paper which gives a discussion of voxel vs conformal mesh.
The fibre orientations are contained in the .ori file.

If your textile does not contain vertical or near-vertical (ie through thickness) yarns then Export->Volume Mesh might work for you. This generates a conformal tetrahedral mesh of both the yarns and matrix (ie the domain).

Hope that helps,
Louise
Attachments
Modelling of textile composites with fibre strength variability.pdf
(818.2 KiB) Downloaded 288 times
UOMO
Regular
Posts: 45
Joined: Tue Feb 24, 2015 10:02 am

Re: Meshing options and cross section of the yarns with doma

Post by UOMO »

Hi Louise,
I appreciate your response. I'm actually using the voxel mesh, however I'm using the elements and nodes generated for matrix and yarns to use in Matlab and APDL (not abaqus). I'm currently writing a script to recognize the rotations within each yarn the then rotate the constitutive tensor (orthotropic) for each element within each yarn. Do you know Is this information available in the .ori file? (I don't see a pattern to use this information), and if it so, do you know how it is organized? Thank you very much again.
louisepb
Project Leader
Posts: 998
Joined: Tue Dec 08, 2009 2:27 pm
Location: Nottingham

Re: Meshing options and cross section of the yarns with doma

Post by louisepb »

Hi,

The .ori file contains the yarn orientation information.
Across each row the first entry is the element number, the next three are a vector representing the fibre direction and the last three are a vector perpendicular to the fibre direction.

Hope that helps,
Louise
UOMO
Regular
Posts: 45
Joined: Tue Feb 24, 2015 10:02 am

Re: Meshing options and cross section of the yarns with doma

Post by UOMO »

louisepb wrote:Hi,

The .ori file contains the yarn orientation information.
Across each row the first entry is the element number, the next three are a vector representing the fibre direction and the last three are a vector perpendicular to the fibre direction.

Hope that helps,
Louise
Yes, I realized about it... Thank you very much for replying Louise.
lambert8809
Regular
Posts: 14
Joined: Fri Oct 07, 2016 5:09 pm

Re: Meshing options and cross section of the yarns with doma

Post by lambert8809 »

Hi Louise,

If my understanding is correct, .ori file is used to define materials local coordinates as described in Abaqus document. According to the abaqus document, the first three can be used to define e1' which is actually the same as fiber direction, while the next three is a point "b" which located in the e1'-e2' plane. So do you mean the point "b" is located at e2' axis in TexGen when you say "the last three are a vector perpendicular to the fibre direction"?


Thanks
Xin
louisepb
Project Leader
Posts: 998
Joined: Tue Dec 08, 2009 2:27 pm
Location: Nottingham

Re: Meshing options and cross section of the yarns with doma

Post by louisepb »

Hi Xin,
The first three coordinates define the x axis of the local coordinate system, the fibre direction. The second three are generated from the cross-product of the first vector and the 'Up' vector used in TexGen and thus gives a vector perpendicular to both which defines the x-y plane of the local coordinate system.
Best regards,
Louise
Post Reply