Modelling of rubber - composite hybrid material

[flo_dlr]

Hello

I am student from Stuttgart University doing my final thesis at the moment. Part of the work is to do research on joining two composites (CFC with Araldit LY 556 matrix) by an elastomere (EPDM). The three parts are joined by co-vulcanisation.
As the solver I plan to use Ansys or Permas. So far I have only basic knowledge of FEM.

* I am looking for hints how I can model this with FEA?
Particulary I am interested how the contact between the elastomer and the composite can be defined and which material model is appropriate for EPDM.

At the moment I am in contact with my supplier about the material data of EPDM, if he can not provide I have the possibility to do experiments.
It is probably accurate enough to model the composite homogenous.


Thank you
Florian

[Jorgen]

If the bond between the different parts is good enough, then you can create a single part that contains the three different materials.

I would start with a simple neo-Hookean material model the EPDM.

-Jorgen

[flo_dlr]

Hi,
thank u for the answer!

What do you mean with one part? Just create one big part and somehow mix the material properties of the composite and the EPDM?

I attached a little sketch it might help to understand better what I plan to simulate.

I think I should create at least three parts, separate ones for the composite and the EPDM, or?

Another question, I have the possibility either to use ANSYS or PERMAS. Do you have experience which one can do the job better?

Thank you, Florian

[manfred]

one comment!

. . . I think you should model each one of your materials as an individual body and then define appropriate conditions on the interfaces between those bodies. For example when perfect contact is in question "glue" the nodes of the body at the interface of the adjacent one etc. EPDM, I think, as already proposed by Jorgen, you may start with the Neo-Hookean constitutive model. But probably I did’nt really catch the point - what is the aim of your analysis? From the answer of this question it depents how the constitutive modelling has to be performed,

manfred

[flo_dlr]

Hello
thanks, the first step is to simulate the sample for axial tension and to at the same time experimental testing of the sample to verify the simulation.

Afterwards the simulation should be used to study the behaviour under long term dynamical conditions (vibrations) leading to fatigue.

Florian

[manfred]

. . . as far as vibration analysis is concerned for EPDM rubber you may use DMA (Dynamical Mechanical Analysis) data which can for example be down loaded from http://www.viscodata.de/. You may also finde a data set for ARALDIT at that web site so far as I can remember.

manfred