CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   blood flow - aneurysm (http://www.cfd-online.com/Forums/openfoam-solving/114982-blood-flow-aneurysm.html)

openfoam_user March 21, 2013 03:57

blood flow - aneurysm
 
Hi OF-users,

I'd like to model the blood flow in the aneurysm.

Blood characteristics are:
- density: 1080 kg/m3
- kinematic viscosity: 3.7037e-06 m2/s

Which solver do you recommend for such a computation ?

I want to start with a steady-state solver.

Best regards,
Stephane.

akidess March 21, 2013 04:23

You specified the viscosity as a constant, thus you can use simpleFoam. In the future, you can easily find the answer to such a question here: http://www.openfoam.com/features/standard-solvers.php

openfoam_user March 21, 2013 04:30

Hi Anton,

OK.
But with the simpleFoam solver you don't need the density. Does it mean that density is equal to 1 or not ?
You only set the kinematic viscosity.

Best regards,
Stephane.

akidess March 21, 2013 04:43

If the density is constant, the equations can be rewritten such as that you do not have to explicitly specify it. Implicitly it comes back in the use of kinematic instead of dynamic viscosity. You can find details in any fluid mechanics textbook.

openfoam_user March 21, 2013 06:24

Hi Anton,

Now it works. I put laminar as RASModel in the RASProperties file.

Previously I had kEpsilon as RASModel with turbulence off.

So it means that a laminar computation is not equal to a turbulent computation with turbulence off !!!

Best regards,
Stephane.

JR22 March 21, 2013 06:44

Blood is non-newtonian also. Does it matter at this scale and conditions?

openfoam_user March 21, 2013 07:06

Hi,

I think you can assume blood as a Newtonian fluid in (large) arteries.
But in small capillaries it must be considered as non-Newtonian fluid.

Regards,
Stephane.

GerhardHolzinger March 21, 2013 07:22

Quote:

Originally Posted by openfoam_user (Post 415457)
Hi Anton,

Now it works. I put laminar as RASModel in the RASProperties file.

Previously I had kEpsilon as RASModel with turbulence off.

So it means that a laminar computation is not equal to a turbulent computation with turbulence off !!!

Best regards,
Stephane.

IF you look at the source code of simpleFoam, you will find, that simpleFoam always uses a RAS turbulence model.

This are some lines of simpleFoam.C

Code:

Application
    simpleFoam

Description
    Steady-state solver for incompressible, turbulent flow

\*---------------------------------------------------------------------------*/

#include "fvCFD.H"
#include "singlePhaseTransportModel.H"
#include "RASModel.H"
#include "simpleControl.H"
#include "IObasicSourceList.H"

You see, the header RASModel.H is included instead of turbulenceModel.H. E.g. compare the first lines of code of simpleFoam with the first lines of pimpleFoam.

The file createFields.H of simpleFoam contains the following lines:

Code:

singlePhaseTransportModel laminarTransport(U, phi);

    autoPtr<incompressible::RASModel> turbulence
    (
        incompressible::RASModel::New(U, phi, laminarTransport)
    );

A RAS turbulence model is create/ used in any case, so the switch turbulence of turbulenceProperties is inactive.

openfoam_user March 21, 2013 07:55

Hi Gerhard,thanks for your useful information.

Best regards,
Stephane.


All times are GMT -4. The time now is 21:57.