Issues when I do match surface

Rhino
,
1

7e157f73b1f3dfd112ea6f75f9cba07
7e157f73b1f3dfd112ea6f75f9cba071438×962 41.9 KB

I’m trying to achieve curvature continuity at A, B1, and B2. However, when I successfully matched at B1 and B2, A gets altered. And when I fix A again, B1 and B2 changed back.

I also tried using NetworkSrf and Sweep 2, but the results weren’t what I expected.

Does anyone have an idea how to resolve this? Thanks in advance!

Below is a link to the file.

2

Do you have the version 9 WIP installed? It has a new command, FillSrf , that can help in this situation. It looks like MatchSrf, even with MultipleMatches (did you try that?) has trouble to get it to curvature continuity.

Matching_Surfaces_FillSrf_MDvR.3dm (249.4 KB)

3

image
image2001×1539 148 KB

why do you have such a complex surface ?
working with deformable circles / arcs is a good approach for matchSrf (it does not like circular surfaces)
what s the maximum deviation from a perfect circle you can live with ?
visual = 0.1 (nobody will see it)
technical = 0.01 (nearly no product is product that accurate)
precise like a clock = 0.001
??

4

image
image2001×1539 193 KB

ok here is a solution that is not 100% Circular - deviation to real arc is lower then 0.1

note that _matchSrf does not like circular / radial surfaces.

it s an nice exercise that shows the limitations of current _matchSrf
And I still hope we get some improvements for V9

key steps:

  • only build one half, mirror it as final step
  • set up approximated Arcs (rebuild) with single Span CV4/D3 and CV6/D5
  • the upper curves are all scaled / sheared - copied versions of the base curves
  • all initial Surfaces are _loft
  • the bigger surfaces is adapted to the smaller stripe
  • for an initial match, there are helper-Surfaces (green) created with _extendSrf (merge = no), those helper-Surfaces are matched first
  • now the bigger surface is matched against those helpers (G2)
  • some manual CV massage with Gumball, MoveUVN, SetPt under the radar of _edgeContinuety

limitations of matchSrf

  • you can not combine G0,G1,G2 in one match
    • workaround: match all G2 Edges with G2, match all G2, G1 Edges with G1, match all G2,G1,G0 with G0 - painful and error prone.
  • you can not look CVs - to let _matchSrf only work with a partial set of CVs (at least I don t know how)
  • _matchSrf does not feedback the resulted deviation while the dialog is active
  • _machtSrf does not allow to limit the number of inserted Knots

… let s see what V9 brings.

image
image2001×1539 103 KB

matchSrf_sample.3dm (278.3 KB)

happy modelling

-tom

5

I’m not sure how MatchSrf relates to this, but this shape is something that Loose Loft does well.
Loose_Loft.3dm (271.5 KB)

2 Likes
7

Thanks so much for your patience and professionalism! It would be super helpful if you could show the steps with screenshots — it’s a bit hard for me to follow just from the text. Really appreciate it, and I hope I’m not bothering you.