Different approach to blending a Y-branch. This workflow is focused on minimizing trimmed edges and achieving a natural flow between the blend surfaces.
Great method! Yes, it might need a bit of additional tweaking but overall the concept is really neat! * Theoretically when you wanted to blend (by extracting edge then pulling it to the other surface to trim it) and because the extracted edge is from a clean surface, namely from a blend surface, pulled to another blend surface, after the trim you could shrink the trimmed one back to that edge (ShrinkToEdge command). Hence, theoretically it could help with the Match/Blend surface.
Hi PashKuli, Thanks for the feedback. I just tried running the "ShinkTrimmedSrfToEdge" on that third blend, but it didn't change the controlpoint structure at all. Not sure If I'm doing something wrong, but it doesn't seem to help in this case. What did help, was to run "RebuildEdges" set to the file tolerance. This reduced the complexity of the trimmed edge and I was able to get a watertight join with just the fourth blend surface. No need to increase the degree. There's always little tricks to improve the result, please continue to comment. I believe that the comments from other experienced users are helpful to everybody. Thanks.
I remember this method from the olden days of Alias Design / Auto Studio when blend curves were a new thing (before Multiblend)... BTW: a good tutorial (no matter how short) always provides context - that could be WHY are we doing it in this specific way, what are alternative approaches, what seems obvious but will lead to problems later on etc...
Hi Mark, I am learning a lot from your tutorials! Thank you very much for these. I tried this with your method and then for fun tried it with xnurbs in Rhino, letting it create those 4 surface that makes up a T shape between the main cylinder surfaces at once by selecting 6 edges - it came out really well to my surprise - have you tried that, letting it fill all the missing surfaces at once ?
Cool method. Thanks for sharing.
Thanks PC Sim!
Great method! Yes, it might need a bit of additional tweaking but overall the concept is really neat!
* Theoretically when you wanted to blend (by extracting edge then pulling it to the other surface to trim it) and because the extracted edge is from a clean surface, namely from a blend surface, pulled to another blend surface, after the trim you could shrink the trimmed one back to that edge (ShrinkToEdge command). Hence, theoretically it could help with the Match/Blend surface.
Hi PashKuli, Thanks for the feedback. I just tried running the "ShinkTrimmedSrfToEdge" on that third blend, but it didn't change the controlpoint structure at all. Not sure If I'm doing something wrong, but it doesn't seem to help in this case.
What did help, was to run "RebuildEdges" set to the file tolerance. This reduced the complexity of the trimmed edge and I was able to get a watertight join with just the fourth blend surface. No need to increase the degree.
There's always little tricks to improve the result, please continue to comment. I believe that the comments from other experienced users are helpful to everybody. Thanks.
I remember this method from the olden days of Alias Design / Auto Studio when blend curves were a new thing (before Multiblend)... BTW: a good tutorial (no matter how short) always provides context - that could be WHY are we doing it in this specific way, what are alternative approaches, what seems obvious but will lead to problems later on etc...
Hi Mark, I am learning a lot from your tutorials! Thank you very much for these. I tried this with your method and then for fun tried it with xnurbs in Rhino, letting it create those 4 surface that makes up a T shape between the main cylinder surfaces at once by selecting 6 edges - it came out really well to my surprise - have you tried that, letting it fill all the missing surfaces at once ?
Can you make a review of XirusCAD does it good?