Clean curves that generate multi-span surfaces?

I was doing a level 2 training manual exercise which involves sweeping 1 rail (see attached)
spans.3dm (132.0 KB)

The input curves for the rail are is minimal/clean as possible to describe the shape, but the generated surface has many spans. Why does Rhino build such dense surfaces when the input curves are clean? Is it because the rail curve is joined? If I were to explode the rail curve and sweep 1 rail, one at a time, for the individual curve segment, it would generate single span surfaces which could then be joined together for a cleaner surface, but this takes much longer…

Hi Justin – to some extent if depends on whether you choose to refit the rail:

but the simplest will come from doing one segment at a time -


I think I might expect the non-fit rails version on a series of acrcs to match that, which it does not on the smaller arcs - I’ll ask.


It would be cool if the rail curve could be exploded and then chain edges could factor the start and end point of input curves into building clean single span surfaces, so it is not a manual one-at-time surface build.

hmm… looks like something funny is going on here

the order of selection does impact the resulting sweep surfaces. That does not make any sense to me.
maybe this illustration helps:

if selecting the longer profile curve first, a clean surface is generated, although only between max. 2 profile curves.

Hmmmmmm, that is strange. Does it have something to do with the underlying curve direction (ShowDir). Could it be something to do with whether curves are selected in a clockwise or counter-clockwise chain?