New to Abaqus, Please help 1

[ajroman]

Hello All,

I have recently started working with Abaqus and have been assigned the task pertaining to 3D composite materials from the unit cell size level. We have used an outside program to import the exact unit cell model into Abaqus, so all the preprocessing is completed. The problem lies within the spacing between the fibers of the materials, none of the fibers are in contact with one another. The model we have developed needs to be "squashed" to produce a realistic replica of the unit cell. This model also needs to be periodical, essentially allowing itself to connect with another similar unit cell from all sides.

Could you please lend any guidance or advice on solving this problem. I have been looking into mostly the contact section of the user manual and I am still running into trouble. Which analysis is better to run, implicit or explicit? Please let me know if anymore explanation is needed. Thanks for your attention/help.

 

[rstupplebeen]

Are you trying to 'squash' dry fabric so that the fibers touch each other?

Set up cyclic symmetry on the faces of your unit cell.

Explicit Vs implicit: I always start with implicit and go to explicit if necessary. It will depend on the level of deformation for a problem like this. I hope this helps.

Rob Stupplebeen

 

[ajroman]

Yes, the fabric needs to match the same thickness of a dry mat, therefore, the warp (longitudinal yarns) and weft (lateral yarns that weave throughout the warp yarns tying the multi layers together) need to deform and mesh together as they would in the manufacturing process.

I am having the most trouble setting up the contact/interaction settings. My initial methodology was to set up boundary conditions such that the bottom of the unit cell would be fixed in all directions. I would then apply a pressure to the top pick yarn (warp yarns that have the same y-axis; columns) and allow the model to compress onto itself to obtain a natural mesh. I have been experimenting with a simplified version of the problem (a column of blocks) to nail the process down. I feel like the problem lies within the initial clearance distance between yarns, its too large b/c the job aborts due to too many increments. Would this be a plausible method?

Thanks for the periodical tip, once I get the compression step down, I'll be sure to try it. Same goes for the implicit suggestion, the reason I asked about explicit was mainly b/c I read that some contact problems may be easier to solve due to a general contact algorithm in explicit analysis.

 

[rstupplebeen]

In practice does the fabric increase in area upon compaction? If so the bottom being fixed may not be appropriate.

I would model 2 rigid plates and have the fabric between them. Fix the bottom plate and translate the upper plate until you achieve the desired gap. You may want to turn on general contact if you can afford the performance hit or you need to define every surface pair that might contact. You will probably also need to turn on automatic stabilization to remove the RBM. I hope this helps.

Rob Stupplebeen