Composite Modeling 1

[john84]

How do I model [0/+60/-60] laminate in 3D with a fiber volume fraction of 21%? I want to evaluate its mechanical properties (Elastic modulus)?

Do I have to use transformation relations for [+60/-60] laminas?


Thanks in advance!!

 

[ESPcomposites]

You will have to be more specific in your goal:

Do you wish to model the fiber and matrix separately? Which modulus do you wish to obtain? Why is the fiber volume fraction so low? What is the point of modeling it in 3D?

Most likely a hand calc will yield an accurate/same result, so are you just doing this as a FEM exercise? The goal itself seems odd.

Also, since the laminate is not symmetric, you will need to use more care as to what the "effective modulus" is (i.e. you will be have extension-bending coupling). That is, unless the laminate is restrained from bending.

To answer your last part, it depends. You could do that, but you can also just create a coordinate system in the +60 and -60 direction and assign the basic ply properties from the 1-2 system.


Brian

http://www.espcomposites.com

 

[john84]

@ Brian Thanks for your comments.

I want to create matrix first and then the fibers and then assemble them together. I want to obtain the inplane elastic modulus of the laminate. Is there any way that I can model it as a 2D with the desired fiber volume fraction? I have several fiber volume fractions starting from 21% and going higher.

I would be glad if you would let me know how do I model the laminate in 2D or 3D with the desired fiber volume fraction?

Thanks in Advance....

 

[ESPcomposites]

I am still not sure why you need a FEM to do this?

To do it by hand you:

1. Get the ply properties via micromechanics (google rule of mixtures for E1). For E2 and G12, you are better off using Halpin-Tsai relations (though the basic mechanics of materials solution will work about as good).
2. Use lamination theory to determine effective engineering constants. There are many programs that will do this for you such as:

http://www.espcomposites.com/software/eLaminate.html

If you want to model it via FEM, you can do it in 2D or 3D, though 3D is probably impractical.

3D:
create the fiber and matrix separately, with the given fiber volume fraction. Note this is very uncommon. Basically build unit cells to represent the plies.

2D:
- Determine the ply properties first via micromechanics solutions.
- Use the shell element in ABAQUS. It will support a composite layup. Essentially, you will be doing the same thing as lamination theory in this case.

Brian

http://www.espcomposites.com

 

[john84]

Thanks Brian...

Well Determining the ply properties would be calculating the elastic modulus and Poisson's ratio at +60 and -60 by micro-mechanics theory and then inputting those values in the abaqus composite lay up(2D shell elements) module right?

 

[ESPcomposites]

No, you calculate E1, E2, G12, v12 via micromechanics. Then input those values with the layup (including orientation angles) and associate it to the element.

You can then obtain the effective props via the load-deformation result. But if the laminate is not symmetric, that may not be straightforward.

You may want to invest in a mechanics of composite book if you are going to be working in that area.

Brian

http://www.espcomposites.com

 

[john84]

Ok I got the procedure. It is not feasible to get stress strain curve of the laminate from Abaqus if I model it as 2D and have stress strain curve of individual constituents?

Thanks.

 

[cpinz]

Hi all,

John84 forget FEM (for now).Take the time to read a textbook on the subject.If , later, you will find tedious to do some matrix calculation, use one of the many simple programs based on CLT; they will do for you what ESPcomposites explained in his posts above....but, before, "back to the basics"