CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Is icoFoam a good solver to model blood? (https://www.cfd-online.com/Forums/openfoam-solving/230328-icofoam-good-solver-model-blood.html)

ramrebol September 18, 2020 12:16

Is icoFoam a good solver to model blood?
 
Hello.


I would like to model the blood inside an artery (aorta). I'm considering the walls of the artery fixed, and the blood as a newtonian fluid, because the aorta is not too small. This fluid is described by the Navier--Stokes with divergence of the velocity equal to zero.


I was using icoFoam solver, but now I realize that icoFoam is for laminar and transient fluid only, but I suppose that the blood inside an artery could be a swirling or turbulent fluid. Do you think icoFoam is a good choice for this model? Or do you suggest to me another solver?

Bloerb September 19, 2020 09:32

use pimpleFoam for unsteady and simpleFoam for steadyState flow. If you need to model FSI check out solids4foam on bitbucket.


icoFoam is basically just a stripped down version of pisoFoam -- so the option to add turbulence is removed to learn how solvers are coded in OpenFoam. And pisoFoam is a stripped down version of pimpleFoam...hence pimpleFoam for unsteady flow. icoFoam is however also sufficient in your case. It does however not have many additional options pimpleFoam has, which might be helpful later on. In both of them you can add various viscosity models like Casson and Carreau-Yasuda to model non newtonian effects of blood flow. For more complicated things like seperation effects inside blood vessels etc in certain diameters you'd need to switch to multiphase eulerEuler variants. For example red blood cells sticking to walls etc if you want to model them as a second continuous phase.


Keep in mind that blood is mostly laminar and can be non Newtonian in behavior. Turbulence is possible, but you shouldn't use the simple Re number. There are variations for pulsating flow...so it might be lower than the typical ~2300 for pipes. If i remember correctly ~900.

mkraposhin September 21, 2020 09:42

Quote:

Originally Posted by ramrebol (Post 783138)
Hello.


I would like to model the blood inside an artery (aorta). I'm considering the walls of the artery fixed, and the blood as a newtonian fluid, because the aorta is not too small. This fluid is described by the Navier--Stokes with divergence of the velocity equal to zero.


I was using icoFoam solver, but now I realize that icoFoam is for laminar and transient fluid only, but I suppose that the blood inside an artery could be a swirling or turbulent fluid. Do you think icoFoam is a good choice for this model? Or do you suggest to me another solver?

Hi, I recently published QHDFoam solver, which was developed for low-Re (less than 10000) incompressible viscous flows - see https://github.com/unicfdlab/QGDsolv...gitef-dev-1912. In some cases it behaves better than standard PISO/SIMPLE algorithms. I hope, it would be useful for you.

Please, write if you have any issues about compilation or running. I also have MSc students who can help you to manage with setup.

ramrebol September 21, 2020 23:28

Thanks mkraposhin, I going to revise your work.


All times are GMT -4. The time now is 21:46.