Mesh From Lines

I've been trying to get a good way of creating a mesh from lines for a long time now.

I've just started to try the Starling tools which look promising.

I start with a set of lines / curves and pipe them, then use the the slFastMesh component and finally Weaverbird Smoothing.

Can anyone tell me why the attached definition will not work when all the pipe radii are the same? It almost works when the pipe radii differ by a very small amount.

Any other ideas on how to achieve a mesh from a framework of lines would be great!

Attachments:

MeshFromLines.gh, 4 KB
MeshFromLines.jpg, 165 KB

Replies to This Discussion

Permalink Reply by martyn hogg on January 4, 2014 at 11:25am

yeah, that's a nice article... it seems like I need to make the giant leap from playing with GH to writing code to generate meshes!

Permalink Reply by Aaron M. Ryan on January 4, 2014 at 11:53am

not necessarily you could do much of this in GH without resorting to coding...it might be heavier, but it can be done.

Permalink Reply by martyn hogg on January 6, 2014 at 3:15pm

Getting closer... Not a neat solution by any means and I still need to tweak it a bit and figure out how to subtract cylindrical meshes from each arm but it works for any of the corners on my geodesic dome.

Permalink Reply by David Stasiuk on January 6, 2014 at 4:27pm

Hi Martyn-

If you have starling, uto's meshedit tools and weaverbird, then you can do it this way. It's really the same logic that's used to solve the connection nodes in exoskeleton, but with the offsets and sizes manually controlled rather than calculated based on angles (so you'll have to manage that)...and it also has the recessed nubs.

What it does is build a convex hull on points defined by polygons along each strut. Then it removes all of the faces interior to each polygon, and fills these holes with the recesses that you want for your struts.

Let me know if you have any questions...

Attachments:

print_nodes.gh, 15 KB

Permalink Reply by martyn hogg on January 7, 2014 at 6:30am

Thanks David,

I looked at this but wasn't sure how to get the uto mesh edit tools... I think you have to register with them on their website? I will have another look though.

Would there be options to extend the arms along the recessed nubs do you think?

I tried the GH mesh tools too last night and got a much nicer mesh than the one I posted above.

More updates later tonight...

Permalink Reply by to] on January 7, 2014 at 6:49am

Hi,

you can access our meshedit tools either over our homepage http://utos.blogspot.com/p/downloads.html (you have to register to access the files) or over http://www.food4rhino.com/project/meshedittools (great page for all the food rhino could maybe need) where you also have to be member of the overall page to download the 3rd party plugins.

Permalink Reply by martyn hogg on January 7, 2014 at 8:21am

Thanks!

Permalink Reply by David Stasiuk on January 7, 2014 at 11:22am

Hi Martyn-

Yes, it doesn't take much of a change...the arms are all handled with variations on offsetting the polygons that define each strut, so here I just add a couple more polygons to be moved outwards...I see you're working in Rhino 4...so I've tried to clean up the image of the definition for you to better read it...hope it helps out!

Permalink Reply by Daniel Piker on January 7, 2014 at 11:44am

nice!

Permalink Reply by martyn hogg on January 7, 2014 at 12:53pm

Thanks, that's brilliant!

From a manufacturing perspective, its hard to control the size of the nubs... they end up conical with rounded ends so if I was to make struts to push into these nubs then i'd have to shape every strut end to fit.

Can you see a way to control the nubs so they remain as specified cylinders with no rounding?

I was thinking to either leave holes in the mesh and fill them in afterwards or do some kind of mesh boolean with cylindrical meshes.

Permalink Reply by David Stasiuk on January 7, 2014 at 2:25pm

I had been thinking about that as well...the subdivision doesn't allow for you to define seams or creases within the mesh. As a quick fix, you may want to create a cylinder along each strut at the appropriate depth and radius...and then for the subdivided mesh, you determine which vertices are inside of a cylinder. If a vertex falls within one of your strut cylinders, you pull isolate it, pull it to the cylinder, and then rebuild the mesh. But it's not particularly elegant...the "inside brep" component gets quite slow when you're testing so many vertices against multiple struts, but it works. There is still a funny bit of geometry at the end of the inside, where some of the vertices are pulled to the sides of the cylinder and others to the caps, but it's close enough: in any event, the sides of the channels are straight and follow the cylinders correctly.

Permalink Reply by martyn hogg on January 7, 2014 at 4:21pm

That's great! I might just print out some structures with this and if I can figure out how to add slots for panes between the struts then I'm halfway to a 3d printed geodesic greenhouse (with no door :-) )

I tried using more sides on the polygons, NOT joining the hole meshes to the rest of the shape, leaving the naked edges fixed during the weaverbird smoothing and then joined the hole meshes back to the smoothed mesh after smoothing. I think this might not be giving a perfect join though, and the mesh doesn't look as nice!

I'm also trying your process but trying to control the polygons so that the base of the end result is flat (to suit the cheap 3d printers) and possibly to include the slot geometry.