How to Join Surfaces to evaluate them as one Surface.

Hey Guys,

I have a Problem. As you can see in the Rhino file, I have three surfaces which I want to evaluate in Grasshopper. I tried to join them with merge, Brep Join... in Grasshopper and also tried joining them in Rhino but GH is evaluating the single surfaces individually. Is there any solution to join them? so GH will evaluate as if its just one Surface. Hope this question will help other users too.

Best regards from Germany

Attachments:

GH.PNG, 23 KB
Join_Surface.PNG, 149 KB
Surface_Join.3dm, 81 KB

Replies to This Discussion

Permalink Reply by Joseph Oster on September 26, 2017 at 8:14am

Your three surfaces are internalized in this GH file:

Attachments:

evaluate_surfaces_2017Sep26a.gh, 32 KB

Permalink Reply by Joseph Oster on September 26, 2017 at 8:39am

By the way, 'Top' view is the correct orientation for the 'MD Slider', navigating the three surfaces:

Attachments:

evaluate_surfaces_2017Sep26b.gh, 34 KB

Permalink Reply by Joseph Oster on September 26, 2017 at 9:00am

Permalink Reply by Michael Pryor on September 26, 2017 at 8:15am

it will evaluate them as 3 as you noticed. Also it will evaluate the untrimmed version of your trimmed breps as you probably noticed already as well. What exactly is the goal. What are you trying to evaluate?

Permalink Reply by Beka on September 26, 2017 at 9:04am

Thanks a lot for the Reply Joseph and Michael. Thx for the file Joseph that Looks very interesting. Im a beginner in GH an will figure out what your definition does. @Michael: im playing around with doing some hole structures on surfaces. I already wrote a definition that works pretty well on single surfaces. I tried it on that three surfaces and failed. However im still not happy with my definition. Im sure there are better ways to do it. You can see the definition and examples below.

Attachments:

Permalink Reply by Joseph Oster on September 26, 2017 at 10:53am

You didn't internalize a surface... so I connected an arbitrarily large 'PlaneSrf' and now I can see the nice spiral patterns of circles. I can 'Project' (graft!) them onto the three curved surfaces:

Any reason you used a surface earlier to create those patterns?

Attachments:

Hole_Structure_2017Sep26a.gh, 60 KB

Permalink Reply by Beka on September 26, 2017 at 12:35pm

Thats great! Thank you Joseph. The patterns are for automotive loudspeakers. My next task will be to let an equal gap/distance between the outer line and the last holes on the surfaces, so they look nice :)

Permalink Reply by Joseph Oster on September 26, 2017 at 1:54pm

Practicing with bigger holes results in fewer holes, smaller lists, faster 'Project' and 'SrfSplit' benchmarks. Ignore holes below a minimum size. Make sure algorithm works at low-res before dialing up to max-res. When projecting (or pulling) curves to a surface, hole shapes will stretch when the source curves and target surface(s) aren't on parallel planes, at "surface normal" to the axis for each hole.

Otherwise, in general, this is projecting or mapping from one frame of reference to another, or in this case, to three target frames of reference (curved surfaces), each covering just a small region of the source pattern.

Wait, did you try a curved "rectangular" surface as the source 'Srf' with your code? That might work well. Standby... check this out; I joined the 'Untrim' surfaces, selected edges to get rails and sections for 'Sweep2'. A single surface, which I internalized as a 'Srf' param and used as input for your code. Interesting! Still need to trim and 'Pull' the holes...

Attachments:

Hole_Structure_2017Sep26d.gh, 88 KB

Permalink Reply by Beka on September 26, 2017 at 11:44pm

Thank you. Im will take a look on your Definition today and try to understand what it does. I didnt try a rectangular surface yet. To avoid stretching the holes i just projected the spiral lines and circles and took the intersection Points for the Center and the normals from the source surface. I believed the hole shapes wont stretch that way.

Permalink Reply by Joseph Oster on September 27, 2017 at 9:28am

Rebuilding the three surfaces into a single surface might be a complete distraction, not worth the effort. The rebuilt surface I created with 'Sweep2' is slightly different from the original three pieces, and there are practical reasons for using multiple pieces, in some cases.

Attached file has several internalized surfaces, trimmed and untrimmed, 'Pulled Crvs' and 'Holed Triangle Srf'.

P.S. Note that 'SrfSplit' took 3.5 hours for 2278 holes!

Attachments:

Hole_Structure_2017Sep26e.gh, 13.4 MB

Permalink Reply by Michael Pryor on September 27, 2017 at 8:09am

GH has MergeFaces but not the one we really want "_MergeSrf". One option if they really need it in gh is the lunchbox plugin has a component for launching any rhino command in gh. Or just do that part in rhino.

Permalink Reply by Michael Pryor on September 27, 2017 at 8:53am

I wouldn't say it is mainly for pattern creation. I use gh daily for tasks that I definitely would not want to do manually and rarely is it for patterns. However, I do agree it is not for everything and surface modelling (for the most part) is def one of those cases. For me gh / coding is to defeat repetitive processes and calculations. Generally, I ask my self 3 questions before jumping into gh / script. 1.Will this take longer to script than do manually? 2. If so, will I need to do this same task many times in the future(then maybe is worth investing the time in the script.) 3.How many cases are there to worry about and is covering those cases worth the time to script.