Why does the epsilon function in bc_cylinder.f90 gives half of the geometry ?

[airwarriorg91]

Recently, I modified the bc_cylinder.f90 file to export the points (xm,ym) where epsilon is 1 to a dat file. Then, I scatter plot the points to visualize the geometry. The points give a half-circle instead of a complete circle (plot attached). But the results of the simulation present a complete circle. Is there the assumption of symmetry in epsilon function about the horizontal diameter.

[slaizet]

Hi,
The epsilon function does not make any assumption of symmetry. There might be something wrong with your script to extract the points to visualise the geometry.
Best wishes,
Sylvain