Filling in corners seems to be a big problem. Last night I watched a number of videos, tried the techniques and was disappointed by the results.
Here is one that I came up with for this problem. I am wondering if this actually solves the problems of bad parts and parallel rails. I want to fill in this area. The bounds are an arc and an ellipse. The frame along the centerline is also an ellipse.
From the edges and the center frame I created a networksrf of the type that is supposed to be bad due to tangent frames.
My question then is: will this trimming of the surface to create a 4-sided opening cause problems (or will any other part of this process cause problems)???
Additional seam could be a problem - for example if you need to run a fillet somewhere at some point of modelling and it goes through that seam - it will fail (not always but in most cases).
networksrf is a pretty powerful command, but I’ve found in some instances not the best option, someone posted this video on another thread in the forum, I found it really good , it may or may not help
Hello, two options works for this kind of corner blend in my opinion:
1:
Curve Offset on Surface with equal distance on both edges
Split at Iso + shrink
Curvature Blend Surface
2:
Split top Surface in the middle, shrink
Blend between both splits
Split & Shrink Blend
Rebuild m = 6,6 n = 5,5
partially tan match at all surfaces
match both blends with average
usually tangency is enough, because exact curvature won’t work in any case (edges do also need curvature connectivity, but results will be nearly curvature matched, so highlights should be alright.
Oh and prevent weighed (rational) nurbs at any cost, it will increase cpcount drastically.
, it may or may not help
