Composite Modelling - Boundary condition problem

[karabiber]

Hello,
Recently, I have read an article about prediction of composite properties from a representative volume element. I am trying to do same things to get the same results, but so far I could not achieve. Here is my question:

I have 3d object (x:100 y:100 z:50) and 47% fiber, 53%matrix. In the model,

Left surface of my object is x symmetric.
Bottom surface of my object is y symmetric.
In the article, it is written that on the Right side of the object, there is a boundary condition which I am not able to understand! I think error comes from this boundary condition!
I have a certain displacement on the top surface (as another boundary condition)
When I do my analysis, the data works.

And then from data manager, I get my forces on the top surface (first ODB field output, reaction forces of the nodes at the top surface, then operate on XY data to combine field output).
I got all forces acting on the nodes, then I find Stress by calculating myself, and then I know the strain on that surface. From that I get elastic modulus.
My elastic modulus is always lower than the composite. There is 100GPa difference due to this RIGHT HAND side boundary condition.

I am attaching the file. If someone can tell me how to arrange the boundary condition of the right hand side surface, I would be really glad.

Any help is already appreciated!

Thank you very much
Berkan

file:

http://rapidshare.com/files/207845213/3d_al_b.zip.html

or :

http://rapidshare.com/files/207845362/3d_al_b.zip.html

 

[karabiber]

Hello, please download from these links:
Updated File
:

http://rapidshare.com/files/207845867/3d_al_b.zip.html

or:

http://rapidshare.com/files/207846239/3d_al_b.zip.html

 

[karabiber]

The closes result that I could get it this one.
By putting a reference point on the edge of right surface, and then constrain it to the whole surface.

Now there is another problem with the meshes.
File

http://rapidshare.com/files/207857746/3d_al_b.zip.html