Prevent penetration 1

[vepawz]

I have a 3 layer stack-up which are all in contact at the beginning of the simulation. At the end of the simulation, they separate due to applied pressure. Now, if I were to restart this simulation with the same applied pressure, some parts of the stack-up are no longer in contact. How do I make sure that these separated parts of the stack-up don't penetrate each other for the rest of the analysis. The stack-up expands due to heating.

I can't define contact interactions since the separated parts of the interface are not actually in contact. Is there any type of constraint that I can apply so that the separated parts don't intersect?

 

[FEA way]

Can you attach a picture of this model ? By stack-up do you mean 3 separate bodies meshed with solid elements ? Contact will prevent penetration if the parts get close to each other. It's not only defined when bodies are already touching each other at the beginning of the analysis. It can also handle situations when there's initial separation between parts and at some point of the analysis they touch due to deformation. The only problem is that, if it's static stress analysis, contact with initial separation may lead to rigid body motions unless each of the bodies is constrained against them.

General contact capability may be the best choice for your simulation. Note that it can also handle self contact and even situations when failed elements are deleted from mesh and interior surfaces come into contact.

 

[vepawz]

Yes, exactly that.

A close-up of the right end.

How do I handle these rigid body motions in case of static stress analysis. I've been trying tie constraints until now, but i'm pretty sure that it's wrong.

 

[FEA way]

Tie constraints will form bonded connections between parts so basically they will behave like perfectly welded/glued bodies. There are several methods to solve the problem of rigid body motions occuring before contact initiates since it's a very common issue. These parts are very close to each other so I would use contact stabilization (*Contact controls, stabilize). Another options is to use boundary conditions to move the bodies into contact and then remove these BCs in subsequent steps.

 

[vepawz]

I need these layers to expand under constant pressure and increasing temperatures as the simulation goes on. I don't necessarily want the edges to be in contact throughout the simulation. They can be only when they have to be, So, I guess adding contact stabilization makes most sense but at what point does the extra energy added by the stabilization become unrealistic?

 

[FEA way]

When you use this option verify if it doesn't have significant impact on results by comparing ALLSD with ALLIE. ALLSD should be small but it's hard to give specific value. Also compare values of CDSTRESS and CSTRESS after the contact finishes establishing. CDSTRESS should be low.