CFD Online Discussion Forums - Problem with cyclicAMI and simpleFoam

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)

- - Problem with cyclicAMI and simpleFoam (https://www.cfd-online.com/Forums/openfoam-solving/142206-problem-cyclicami-simplefoam.html)

Problem with cyclicAMI and simpleFoam

Dear Foamers,

i am learning how to calculate rotating machines in OpenFOAM. I have a problem with the cyclic boundary condition in the rotor domain. Attached a picture that shows the cross section at the midle of the rotor domain of pressure contour with very high or low (unrealistic) values close to the cycly BC.

Basically I have done the following steps.

1)The mesh I converted to OpenFoam using "Fluent3DMeshToFoam".

2)Then I defined for the cyclic boundary conditions "cyclicAMI" in the boundary file as below

...

p1-rotor
{
type cyclicAMI;
inGroups 1(cyclicAMI);
matchTolerance 0.0001;
transform rotational;
neighbourPatch p2_rotor;
rotationAxis (0 0 1);
rotationCentre (0 0 0);
nFaces 5871;
startFace 6734366;
}
p2_rotor
{
type cyclicAMI;
inGroups 1(cyclicAMI);
matchTolerance 0.0001;
transform rotational;
neighbourPatch p1_rotor;
rotationAxis (0 0 1);
rotationCentre (0 0 0);
nFaces 5871;
startFace 6751983;
}...

3) In the fvOptions the corresponding angular velocity on rotor domain.

MRF1
{
type MRFSource;
active true;
selectionMode cellZone;
cellZone rotor;

MRFSourceCoeffs
{
origin (0 0 0);
axis (0 0 1);
omega 77.5;
}
}

The residuals after 4000 iterations are for pressure just below 10-e3, velocities below 10-e4 and k and omega below 10-e5.

Does anyone knows why my strange results chose to the cyclicAMI boundary condition?

Thanks in advance,
Mat

It seems that you are trying to use OF MRF module, as you are using the fvOptions dict. My suggestion would be to try setting your cyclicAMI patches as non-rotating ones. This might improve strange behaviour near the cyclicAMI boundaries.

Quote:

Originally Posted by Bluewater (Post 511847)

Dear Foamers,

i am learning how to calculate rotating machines in OpenFOAM. I have a problem with the cyclic boundary condition in the rotor domain. Attached a picture that shows the cross section at the midle of the rotor domain of pressure contour with very high or low (unrealistic) values close to the cycly BC.

Basically I have done the following steps.

1)The mesh I converted to OpenFoam using "Fluent3DMeshToFoam".

2)Then I defined for the cyclic boundary conditions "cyclicAMI" in the boundary file as below
...
p1-rotor
{
type cyclicAMI;
inGroups 1(cyclicAMI);
matchTolerance 0.0001;
transform rotational;
neighbourPatch p2_rotor;
rotationAxis (0 0 1);
rotationCentre (0 0 0);
nFaces 5871;
startFace 6734366;
}
p2_rotor
{
type cyclicAMI;
inGroups 1(cyclicAMI);
matchTolerance 0.0001;
transform rotational;
neighbourPatch p1_rotor;
rotationAxis (0 0 1);
rotationCentre (0 0 0);
nFaces 5871;
startFace 6751983;
}
...

3) In the fvOptions the corresponding angular velocity on rotor domain.

MRF1
{
type MRFSource;
active true;
selectionMode cellZone;
cellZone rotor;

MRFSourceCoeffs
{
origin (0 0 0);
axis (0 0 1);
omega 77.5;
}
}

The residuals after 4000 iterations are for pressure just below 10-e3, velocities below 10-e4 and k and omega below 10-e5.

Does anyone knows why my strange results chose to the cyclicAMI boundary condition?

Thanks in advance,
Mat

Hi Taxalian,
in fact the cyclic boundary conditions have to rotate with the domain defined as rotational domain. The model consist of one blade (see picture posted in my first post) that spans 72 degrees (so the real whole model has 5 blades). As you suggested I have defined in "fvOptions" file at "noRotatingPatches()" field the two periodic patches (they are defined in "boundary" file as "patches" but with a "createPatchDict" I tranform them to "cyclicAMI", correct me if this is a wrong procedure, because I do not know how to address directly "cyclicAMI" in "noRotatingPatches()").

Code:

 

MRF1

{

    type            MRFSource;

    active          true;

    selectionMode   cellZone;

    cellZone        fluid_rotor;



    MRFSourceCoeffs

    {

        nonRotatingPatches (p1-fluid_rotor p2-fluid_rotor);

        origin      (0 0 0);

        axis        (0 0 1);

        omega       77.5;

    }

}

The "createPatchDict" looks:

Code:

patches

(

    {

        name            p1_rotor;

        patchInfo

        {

            type            cyclicAMI;

            matchTolerance  0.0001;

            neighbourPatch  p2_rotor;

            transform       rotational;

            rotationAxis    (0 0 1);

            rotationCentre  (0 0 0);

        }

        constructFrom patches;

        patches (p1-fluid_rotor);

    }



    {

        name            p2_rotor;

        patchInfo

        {

            type            cyclicAMI;

            matchTolerance  0.0001;

            neighbourPatch  p1_rotor;

            transform       rotational;

            rotationAxis    (0 0 1);

            rotationCentre  (0 0 0);

        }

        constructFrom patches;

        patches (p2-fluid_rotor);

    }

);

I do not understand why for example under 0/* I have to address the cyclicAMI with the names that I use in "constructFrom patches;" of "createPatchDicht", since from the moment I have redefined them as "cyclicAMI" it has already a new name (i. e. in my case p1_rotor and p2_rotor).

If I look at the results now as expected the two cyclicAMI do not rotate, they behave as walls (velocity is 0.) I would expect an easy way to get the cyclicAMI rotate in the proper way), there are any idea? I am doing something wrong?

Quote:

Originally Posted by taxalian (Post 512083)

after some trials I do not see how to get that the periodic bondary conditions (defined as "cyclicAMI") in the rotational domain do what is expected, look the two pictures attached, where it is clear that they rotate in the contrary direction. This clarifies also the strange pressure contours closed to the periodic b.c. I attached in my first post. Does anyone in this forum work with this kind of problems, or "cyclicAMI" inside a "cellZone" that rotates and defined as such in "fvOptions" with "MRFSource"?