Periodic Boundary Contitions

[omoya]

Hi all,
I'm trying to simulate the behaviour of and unit cell of bi-metal (metal substrate with a film of other metal) against thermal strain. I wanna to use Periodic Boundary Conditions to have a discret volum of control (using a *EQUATION Interaction Module) but I haven't been able to apply this.
ABAQUS always returns a message error about that the DOF it's not possible to be constrained because it's not active in the model. What type of element I have to use? Is it possible to apply via kernel or I need to edit the inp file?

Thanks!

 

[corus]

A dof not active in the model could mean you've restrained that freedom so it no longer applies or you've just got the wrong element type. It's probably the latter. If you've used a thermal element then you can simply edit the inp file to change it to a solid stress element. Similarly with shells.

corus

 

[omoya]

Thanks corus for your comment. I was thinking that but not sure.
About the elements, I use the standard solid stress element (no reduced integration) that you can select via the kernel, but this ones have a only 1,2,3 DOF active... The problem is that I need to use several MPC (*EQUATION BC) for apply a Periodic Boundary Condition (defining the displacements & rotations of the unit cell faces symmetrical over middle planes) and I need that the 4,5,6 DOF to be active...

omoya

 

[corus]

Solid elements don't need rotational degreees of freedom, only shell elements have rotational degrees of freedom as they have no physical thickess. Solid elements have a thickness and hence restraining two nodes translationally also restrains the element rotationally. If you have a film of metal on the surface of a solid then use the 'skin' feature in the property module to apply a layer of shell elements to the solids. As the nodes are concident there is no need to tie the nodes toegther.
Other than that I don't know what is meant by a 'periodic boundary condition', even after checking the manuals so can't answer your question any better.

corus

 

[omoya]

I´m really thankful to you for you clearing comment. You are right about the rotational dof for solid elements so I believe that I was doing something wrong applying the MPC for simulate a Periodic Boundary Condition behaviour.

I'll try modeling the thin film material as you says.

Thanks a lot!

 

[bfillery]

Dear omoya

If by "Periodic Boundary Condition" you mean you want your modelled segment of the bi-layer to behave as if it were a part of a much greater semi-infinite or infinite medium, then you may want to apply a combination of symmetry boundary conditions and coupling such that all planes remain plane during the analysis. To do this you will only model half the unit cell with appropriate symmetry boundary conditions, and all other remaininng surfaces will be coupled in the direction normal to those surfaces. To do this, for each surface, create an assembly reference point, and apply a surface based kinematic coupling constraint on those surfaces with the constrained degree of freedom corresponding to the direction of the surface normal.

bfillery

 

[omoya]

Yes bfillery! This is exactly what I need!
I proposed to the tutor of the project to do this with axisymmetric shell elements and in the boundary put in infinite element CINAX4 (may be another way to do the same) but he impose the using of PBC.

Thank you to explain how I can apply this method.

 

[corus]

Student posting isn't allowed on this site as you're supposed to do the work yourself. As such, no further comments can be made to the question.

corus

 

[omoya]

Sorry but I didn't know this site rule, so I'll try to do the job with infinite elements because anyone in the school have been able to tell me how I can use Periodic Boundary in ABAQUS CAE.