Following the level 2 training but encountered troubles

  1. I made the last 2 surfaces of the scoop using NetworkSrf and Sweep2, finding that the NetworkSrf one cannot be joined.

    After tuning the continuity option and adding more network curves through BlendCrv, it still doesn’t work.

    I know the training is for Rhino 5, something must have changed in so many years, but it’s hard to know what happened actually. I will be appreciated that if someone can explain the details

  2. Another question is, I choose different continuity options when Sweep2. It looks similar, however, when I quadremesh them, one is good, one is bad, what leads to this difference
    Scoop_join_mesh_troubles.3dm (18.1 MB)

Hello - try a networkSrf like this:

Note the two ‘helper’ curves created witf BlendCrv >Edges, for tangency.

That said the ‘floor’ of this thing picks up higher than the surrounding surface
image

I would fix that first. These curves are the culrpits:

Shoot for really simple and similar, one to the other, accelerating curvature. These are degree 5, 6 points curves.

Then you can make a nice clean loft between them, matchSrf to the front bit etc etc.

-Pascal

1 Like

Thanks for your help.

Doing NetworkSrf in this way returns surfaces that can be joined.

For the second question, do you mean drawing a degree 5 curve then making the ‘floor’ lower than the main surface? It makes the curvature graph smoother. (which also applies to other n+1 curves). The outcome looks good, especially to the end of the flat lines.

Still, when I sweep2(networksrf functions well now) with the continuity option curvature on both rails, I finally get a bad mesh after quadremesh. I think the “curvature” or “tangency” continuity is the culprit.

Tangency makes no mistake, curvature has different behavior. Why I should use tangency in this circumstance?

By the way, a high degree curve is not ideal for subd, is there an applicable way to draw a control points curve with more than 4 points in degree 3 and keep away the peak in the curvature graph?

Hello - the fillets at 5 mm around the edges are tangent to their inputs, so you want the new surfaces that are extensions of the fillets, to also be tangent. If they are G2 then where they meet the tangent fillets, there will be a conflict and the surface will get messy if it needs to match both.

-Pascal

1 Like

Thank you so so much for this. You are a life savior