STRUCTURE-FLUID interaction (and not the contrary) with Comsol

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
Sign up Log in
2 replies · 8K views
sunset451
Messages
4
Reaction score
0
Hi,
I've got a strange doubt.I model an elastic diaghfram that move water in piping system (a simple pump) so I've got used the 3D FSI multiphicis. Note that in Comsol's examples is the fluid that move the solid structure: i use the same procedure but for the opposite case.
I create a simple geometry with a thin elastic rectangular solid (the diaghpram) directly attached on a cube(the fluid subdomain). I set on solid fixed constrains on four laterals buondaries and a surface load on upper wall.

So I've found an absurd: if i leave the default setting on the boundary condition on the fluid-structure wall (that corrisponding at the "structural velocity" time derivates of solid displacements (ut,vt,wt) ) the fluid doesn't see the perturbation.
BUT if in the same field i set DIRECTLY the SOLID DISPLACEMENTS (u,v,w) the model works perfectly and it rapidly converge. BUT Dimensionally is an absurd, why it works? Please help me to understand.
 
Physics news on Phys.org
sunset451 said:
Hi,
I've got a strange doubt.I model an elastic diaghfram that move water in piping system (a simple pump) so I've got used the 3D FSI multiphicis. Note that in Comsol's examples is the fluid that move the solid structure: i use the same procedure but for the opposite case.
I create a simple geometry with a thin elastic rectangular solid (the diaghpram) directly attached on a cube(the fluid subdomain). I set on solid fixed constrains on four laterals buondaries and a surface load on upper wall.

So I've found an absurd: if i leave the default setting on the boundary condition on the fluid-structure wall (that corrisponding at the "structural velocity" time derivates of solid displacements (ut,vt,wt) ) the fluid doesn't see the perturbation.
BUT if in the same field i set DIRECTLY the SOLID DISPLACEMENTS (u,v,w) the model works perfectly and it rapidly converge. BUT Dimensionally is an absurd, why it works? Please help me to understand.

I was in error... for activate the fluid motion i set to the interface a pressure type boundary condition with (Tax_smsld,Tay_smsld,Taz_smsld)...so now the problem is that the fluid is strongly turbulent and the solution converge only for too small force applied to the diagphram... i need to understand the "artificial diffusion" options.please help me.
 
Hello my friend,

In my case I´m trying to simulate the flow paths of a gearbox with a pair of spur gears... Right now I´ve been trying to follow two directions... rotating machinery module on chemical engineering and fluid-structure interaction on Structural mechanics... works fine for chemical module but it will be difficult to simulate a gear pair with that approach... and on the last one I haven´t been able to simulate fluid movement due to the gears... Can you explain me if you could fix and get fluid movement due to a solid body?

Thanks in advance !
Jesus