Visualing surfaces/interfaces in 3D

[fu351]

I am currently trying to model the lens in 3D in my optical model system by tracing the rays and using the interception points(the coordinate when a ray enter/exits a surface/interface) to create a mesh to plot.
However this has the issue of taking a long time when trying to create an accurate mesh since a lot of rays had to be traced to get enough interception points.
Is there a way to model these surfaces directly by extracting their profiles instead of using the interception points?

[mjhoptics]

Hello @fu351,
Yes, I think you can do this in ray-optics without invoking ray-tracing.
The Surface class has an instance of a profile as an attribute. The sag() method on the profile can be used to give the z coordinate of a point x, y on the surface.

N = 5      # mesh dimension
mesh_list = []
for i in range(1, sm.get_num_surfaces()-1):
    mesh = []
    surf = sm.ifcs[i]
    sd = surf.surface_od()
    for xj in np.linspace(-sd, sd, num=N):
        mesh_row = []
        for yk in np.linspace(-sd, sd, num=N):
            if surf.point_inside(xj, yk):
                z = surf.profile.sag(xj, yk)
            else:
                z = None
            mesh_row.append(z)
        mesh.append(mesh_row)
    mesh_list.append(mesh)

There's probably a more numpy-efficient way of doing this, but this conveys the idea.
Hope this helps.
Mike Hayford