Error while creating custom geometry in BC-Cylinder.f90

[xcompact3d]

Error while creating custom geometry in BC-Cylinder.f90

[mathrack]

Dear Gaurav,

I think the present issue / bug is not present in the DNS code available on github and is directly linked with the modifications you did. I will thus close this issue. You can open a similar topic in the discussion section.

Thank you,
Mathrack

[airwarriorg91]

Hello @mathrack ,
I understand that is does not seems to be bug with the code because usually it is working.

But, can you tell me why the values of value of nxi,nxf, nyi, nyf, nzi,nzf is changing while running the simulation. I suspected the error I am getting is due to index out of range. Ideally, the values of nxi,nxf, nyi, nyf, nzi,nzf should be equal to the values of set by the user in the input.i3d file, right ? But printing the values, presents it is changing. So, if I am able to crack this, then I think I can solve the problem.

I have worked on a python code which can be used to generate epsilon function from STL files directly, since the xcompact3d-toolbox is not working anymore. There seems to be some problem with installation. Libraries are not supporting each other. If we solve the error, then it can be used by the other xcompact3d users to import complex geometries.

Thanks,
Gaurav

[mathrack]

Here are the calls to geomcomplex (where there is a call to geomcomplex_cyl)

$ grep -ir call\ geomcomplex\(
genepsi3d.f90:    call geomcomplex(ep1,xstart(1),xend(1),ny,xstart(2),xend(2),xstart(3),xend(3),dx,yp,dz,one)
genepsi3d.f90:    call geomcomplex(ep1,xstart(1),xend(1),ny,xstart(2),xend(2),xstart(3),xend(3),dx,yp,dz,one)
genepsi3d.f90:    call geomcomplex(xepsi,1,nxraf,ny,xstart(2),xend(2),xstart(3),xend(3),dxraf,yp,dz,one)
genepsi3d.f90:    call geomcomplex(yepsi,ystart(1),yend(1),nyraf,1,nyraf,ystart(3),yend(3),dx,ypraf,dz,one)
genepsi3d.f90:    call geomcomplex(zepsi,zstart(1),zend(1),ny,zstart(2),zend(2),1,nzraf,dx,yp,dzraf,one)

[airwarriorg91]

Thanks a lot for that ! I will check them. As I understand the code in geocomplex subroutine will check which of the points are inside the geometry using the equation provided and set epsilon to 1 accordingly. This is how it works, right ? I am trying to the exact thing like just skipping the checking part and setting epsilon to 1 for the list of points i have. But I am not able to understand why the mesh size is changing.

Thanks,
Gaurav

[mathrack]

Sorry, I can not help you with the IBM part of the code. Please do some bibliography on the IBM, go to the discussions section and open some topics there. I will now lock this issue.