Cell Size Dependence of final geometry?

[ghost]

Does cell size play any role in how the final geometry looks? In the following code, I have kept the simulation domain size constant at 64nm x 64nm x 3 nm. I have varied the no. of cells in x- and y-direction, which in turn varied the cell sizes. No of cells along x-, y- and z-directions are Nx, Ny and Nz. I have kept Nz constant at 1. and varied the ratio between Nx and Ny. I am trying to make an ellipsoid structure with long axis along y-direction.

// does final geometry depends on Nx/Ny ?

// Define gridsize & cellsize Ny := 128 Nz := 1 sizeX := 64e-9 sizeY := 64e-9 sizeZ := 3e-9

start := 0.25 loop_step := 0.25 stop := 4

for ratio := start; ratio <= stop; ratio += loop_step{ Nx := ratio*Ny SetGridSize(Nx, Ny, Nz) SetCellSize(sizeX/Nx, sizeY/Ny, sizeZ/Nz)

//Define geometry edgesmooth = 10 ax_major := 18e-09 ax_minor := 12e-09 main_struc := ellipsoid(ax_minor,ax_major,3e-09) setgeom(main_struc)

//Define regions defregion(1,main_struc) save(regions) } But I obtained following results- The 1st image correspond to Nx/Ny = 0.25 and the last image correspond to Nx/Ny = 4 and so on. From images, it is very clear that the long axis is not along y-axis. I will be really thankful if you could point out the cause of these erroneous results. Regards, AM

[JeroenMulkers]

This is not unexpected behaviour. The user is helped here.