Weaving tapes: Problems in connecting nodes and multiple contacts

[ExceedtheMax]

Hi guys,

I hope you guys can help me in this matter. I want to analyze a woven configuration made of weaving tapes under a constant load. The tapes are bonded to each other but I'm afraid the nodes are not connected between the tapes, yielding inaccurate results. There are also a number of contact areas due to the weaving configurations. How can I connect the nodes between the tapes?

For easier understanding, a picture of the weaving model is attached with this post.

Could anyone advise me on how to go about this? One advice I got was to just make a homogenous layer instead of modeling the weaving profile.

Thank you.

img

http://files.engineering.com/getfil...9d-4253-8d92-5e30c7becb46&file=woven_tape.jpg

]

 

[rickfischer51]

If the tapes are modeled in physical contact, you can select all an make a multibody part. That will give you shared nodes. Or you can use bonded contact. If there is a gap between the tapes, multibody part will not work, but contact will. You may have to adjust the pinball to ensure contact is modeled. If you go with a multibody part, it may be difficult to mesh, because from the mesher's viewpoint, it will see one complex part.

You show the tape making neat right angle bends when it weaves between the cross tapes. To get the response of a woven tape, you want to include orthotropic material properties, and the direction will change when the tape bends. You need to figure out how to account for that in your model.

I dont think the homogeneous layer will work because at any depth in the section, you will have tape running in each direction at different points in the layer.

Rick Fischer
Principal Engineer
Argonne National Laboratory