[nyflyer]

Hello all,

I'm trying to use the cyclic boundary condition to simulate periodicity, and the channel395 case has this implemented. However, looking at the channel395 case under post-processing, I notice that the boundary patches that are supposedly cyclic are not the same as you can tell by the pressure contours of p=0, and velocity magnitudes (see the attached screenshots.)

According to the boundary file in polyMesh (and the blockMeshDict) they should be cyclic, as below:

    inout1_half0
    {
        type            cyclic;
        inGroups        1(cyclic);
        nFaces          750;
        startFace       181700;
        matchTolerance  0.0001;
        transform       unknown;
        neighbourPatch  inout1_half1;
    }
    inout1_half1
    {
        type            cyclic;
        inGroups        1(cyclic);
        nFaces          750;
        startFace       182450;
        matchTolerance  0.0001;
        transform       unknown;
        neighbourPatch  inout1_half0;
    }
    inout2_half0
    {
        type            cyclic;
        inGroups        1(cyclic);
        nFaces          750;
        startFace       183200;
        matchTolerance  0.0001;
        transform       unknown;
        neighbourPatch  inout2_half1;
    }
    inout2_half1
    {
        type            cyclic;
        inGroups        1(cyclic);
        nFaces          750;
        startFace       183950;
        matchTolerance  0.0001;
        transform       unknown;
        neighbourPatch  inout2_half0;
    }

The boundaries on the pressure (0/p) as as follows:

 inout1_half0
    {
        type            cyclic;
    }
    inout2_half0
    {
        type            cyclic;
    }

    inout1_half1
    {
        type            cyclic;
    }
    inout2_half1
    {
        type            cyclic;
    }

That being said, is this is a bug in OpenFOAM or am I just not getting something right? From what I understand, the same solution should be on both cyclic patches (i.e. no equations are solved for this part of the matrix and the solution is just re-produced.)

What do you think is going on? Is this not the "periodic" boundary condition that I am expecting? If the cyclic BC is not the "periodic" boundary condition, then does OF have an implementation of the periodic boundary condition?

Moses

[nyflyer]

The difference is even more pronounced when looking at the "sides" patches, as you can see from the picture that neither the pressure or the velocity contours match across the sides.


Note: I'm not talking about the discontinuity, but rather, the similarity (or differences in this case) between patches are that are facing each other.

Moses

[nyflyer]

So I can't still figure this out if anyone has any insight. The channel395 case uses pimpleFOAM.

Is this a bug?

Moses