P.S. More elaborate code was needed to avoid using isocurves, which are affected by surface shape. Also, I had to check loft option ‘Align sections’ at the end (green group, not shown).
P.P.S. Oh, rats! I tried to avoid loft option ‘Align sections’ but failed. You can uncheck it and see what it does. But I just tried using it on your original shape and it alters the result. So it must be altering the result with the Rhino shape too…
OK, got it. After jumping through hoops to align seams (failed!), I found the fix was just to flip curves with a circle guide (yellow groups). So loft option ‘Align sections’ is not used at all.
In addition to that white group, I needed to override the Rotate ‘P’ input because the Rhino shape was not at the origin. I was oblivious to the fact that isocurves all have the same direction so didn’t see the flip curve issue until later.
But as I said, isocurves are affected by surface shape so they weren’t evenly spaced:
Which led to versions ‘b’, ‘c’, ‘d’ and ‘e’ I vaguely recall now that this issue of the top edge plane intersection curve being flipped has been reported and discussed before, in relation to contours.
But I struggled with it yesterday, wasting a lot of time going through four different versions to understand and fix the problem.