Mechanical Analysis of Woven composite in ANSYS

[rmadk.dce2014]

After successfully exporting step file of a woven composite from Texgen, I want to analyse it in Ansys. Importing the geometry and simply assuming a bonded contact between yarns-matrix/ yarns-yarns doesn't give me similar results as voxel mesh export in Abaqus. I am manually assigning yarn and matrix properties. Where am i wrong?

[louisepb]

Hi,

If you are exporting a step file then all you have is the geometry. You will have lost any information about yarn orientations and volume fractions which, I assume, will make a difference to your results.

Best regards,
Louise

[rmadk.dce2014]

I am using ANSYS Workbench just because of its automatic contact option in connections. ANSYS automatically recognizes all the contacts between yarn-yarn/yarn-matrix which can be modified with a single click. In ABAQUS, each contact pair interaction has to be defined by specifying each of the individual surface pairs that can interact with each other. This is quite cumbersome in case of 3D textile composite. Can you help me if there is any simple way for contact interaction in ABAQUS?

[louisepb]

Hi,
I'm afraid that I'm not an Abaqus expert so may be of limited help. The voxel export assumes perfect bonding between the yarn and matrix and therefore doesn't specify contacts. This is fine for prediction of elastic properties. If, however, you want to simulate debonding or something else then you may need to specify the contacts. I don't think that there is an automatic way of specifying the contact surfaces in Abaqus but someone who knows more about Abaqus may be able to advise you better.

Sorry not to be more help,
Louise

[rmadk.dce2014]

Thanks for your reply. You are always helpful.

[rmadk.dce2014]

In the paper by Yan,S et.al.,"EXPERIMENTAL ASSESSMENT AND NUMERICAL ANALYSIS OF 3D WOVEN COMPOSITE T-JOINTS UNDER TENSILE LOADING" Figure 11 is as attached. How can one model a whole T-Joint using TexGen? Only the RVE's at different locations can be modelled as far as i know. But they draw this figure of T joint with same colour combinations as in Texgen?

[louisepb]

Hi,
Apologies for the delay in replying. There seemed to be a problem with the Forum last week and I couldn't get it to let me submit an answer.

Yes, this T-piece was modelling using TexGen. Shibo Yan wrote a script which defined the yarn path and cross-sections for each yarn. Because the textile is not periodic in all directions (only along the length of the T, looking into the screen as shown in the illustration) the standard periodic boundary conditions included in the TexGen export cannot be used. I believe that he used a voxel export and then specified the required boundary conditions in Abaqus.

The standard textiles created automatically by TexGen generate repeats and domain for a periodic textile but this isn't a requirement of a TexGen model.

I hope that helps,
Louise

[rmadk.dce2014]

Ahaan.Great. Thank you for replying.