[ashtonJ]

Dear All,

I am doing FSI modelling in an artery model. I want to use the following boundary conditions for the FSI modelling.

1- Inlet: Fixed (Stationary)
Outlet: Not fixed (Free to move)

2- Inlet and outlet not fixed (both free to move)

My FSI model is converged using the first boundary condition; however, for the second boundary condition, I have got the following error message:

In Analysis 'Flow Analysis 1': Export results are unavailable as no suitable boundaries exist for use as export surfaces. Mesh motion on a boundary invalidates its use as an export surface.

What I did:
In Structural part, I defined "frictionless support" for both inlet and outlet, then in CFX-pre, I chose "unspecified" option for the mesh motion of the inlet and outlet.

Could anyone help me please.
Kind regards,
AshtonJ

[p.galimutti]

that is not an error. it's just a warning message! you ABSOLUTELY need not to worry about it !

'friction-less support' means you can either use it for symmetry or constrain the motion only normal to the face. If your solid bcs represent correctly, your fluid part is correct (unspecified).

[ashtonJ]

Dear galimutti,
Thanks so much. For solid part, I specified frictionless support boundary condition at inlet and outlet and fluid solid interaction on the inner surface f the wall which is in contact with fluid. There is no boundary condition specified on the outer surface of the wall.
I think because the model is less constrained, its’ movement is large and that's why it doesn't converge. Is there any specific setting for FSI modelling that makes easier the getting converge.

Kind regards,
AshtonJ

Originally Posted by p.galimutti

that is not an error. it's just a warning message! you ABSOLUTELY need not to worry about it !

'friction-less support' means you can either use it for symmetry or constrain the motion only normal to the face. If your solid bcs represent correctly, your fluid part is correct (unspecified).

[p.galimutti]

No we can not get convergence by magic! there are options though like smaller time step, expert settings for mesh displacement etc.

but first you need to get your bcs correct!

[ashtonJ]

Thanks. What I want to model, is an artery which is completely free to move (no constrains) due to the fluid pressure. To do this, I specified frictionless supports at inlet and outlet and FSI boundary condition on the inner surface of the wall. Do you think these boundary conditions are correct to model the aforementioned movement?

When I fixed the inlet of the model and specify the frictionless at the outlet, it converges.

Kind regards,
AshtonJ

[p.galimutti]

friction-less support only restricts the movement normal to the face. say Z direction, and X & Y are completely free to move. so at your inlet/outlet faces you restricted motion in Z direction only, what about X & Y? So if you give friction-less supports at both the ends, buckling (or something similar) of the artery could happen.

i guess you need to do little bit of research on how arteries work. then you'll be able to visualize appropriate bcs. Once you're done with that i maybe able to help you how you can avoid convergence problems.

[ashtonJ]

Again thanks. I have done a research on different kind of FSI boundary condition in arteries.

What I am trying to do as my thesis, is to model pulsatility motion, cardiac motion and respiratory motion modelling in human coronary arteries. For the pulsatile motion, different boundary conditions were used, some people fixed the inlet and outlet of the artery and some people did not fix the inlet and outlet. What I am going to do is to investigate the effect of different FSI boundary conditions on the flow behaviour. I could converge for three different bcs, however, the last one with not fixed inlet and outlet has not been converged.

Thank you.
Regards,
AshtonJ