Help in removing self-intersections within a brep

I've been experimenting with some geometry in GH, and have created a twisting tube spiral shape. For my project, I need this to be one solid shape, but the way I have created it means it passes through itself multiple times. Is there any way in grasshopper to create a solid form just from the overall external shape, rather than having it intersect itself (almost as if I were making a mold). I'm really stumped on this one, although I admit I am not an expert by any stretch

I hope this makes sense!

Attachments:

Spiral SHape.3dm, 2.6 MB
Capture.JPG, 156 KB

Replies to This Discussion

Permalink Reply by Max Alexander on February 9, 2016 at 8:37am

I guess what I am trying to ask is whether there is a tool or method I am unaware of which could be used to create a form defined by only the external surface / skin of my shape

Permalink Reply by peter fotiadis on February 9, 2016 at 10:16am

Yikes ... this is ultra paranoid (therefor I like it).

Spend some minutes for a random super-doughnut generator > now what ??

I'll be back with some ideas(?) soon(?)

Permalink Reply by peter fotiadis on February 10, 2016 at 1:20am

Progress is made - but since there's no IntersectWithSelf method available for Breps/Surfaces [as is available for Curves] the "key" method used (not available as component) is the notorious - and not recommended at all - Brep.MergeBreps method where you do the following paranoid thingy: you merge a Brep (pipe's BoundingBox as Brep) with a flipped Brep (pipe): if this makes any(?) sense(?), that is ... he he.

Therefore the "solution" is performed solely via code.

However it's highly questionable what Rhino means with regard manifold "solids" ... or the fact that a untrimmed Surface (pipe) is evaluated as closed Brep (Rhino) and ... er ... hmm ... reported as untrimmed Surface (GH).

more soon (Nik and Djordje would like that).

Permalink Reply by Nik Willmore on February 10, 2016 at 1:32am

An exact solution would be thrilling indeed.

I note that Boolean union also works with two separate intersecting cylinders, even after the caps have been removed so they are just open tubes, and the result retains the open caps.

This fails for two copies of this same tube though, since the surfaces are coincident everywhere, not just at tube self-crossings.

Also a fail is Boolean union after removing the caps and splitting the tube along its length with two long snaking isocurves, here shown separated:

The Boolean union merely restores the original, akin to a join.

Permalink Reply by peter fotiadis on February 10, 2016 at 1:48am

But by means of a "black box" (as you said for the other "similar" case a while ago), he he.

Anyway you can translate it to P.

Permalink Reply by Nik Willmore on February 10, 2016 at 2:27am

Merely splitting each free loop works OK to allow a Boolean union, for the main body, but fails at the stem where there are no free loops, just lots of self-intersection.

Permalink Reply by peter fotiadis on February 10, 2016 at 3:23am

Hmm ... due to hideous delay time(s) I think it's worth trying Plan B (the HocusLocusPocus option, that is). Shown the "section" option (spot the "part" missing from the orange pipe that is NOT // to section plane: meaning the good news):

Permalink Reply by Nik Willmore on February 10, 2016 at 1:27am

The new Cocoon marching cubes plug-in affords a list of metaline field strengths along its length. I extracted the isocurve your model features along its length to approximate your missing center curve.

http://www.bespokegeometry.com/2015/07/22/cocoon/

Attachments:

Cocoon TEMP A.gh, 89 KB

Permalink Reply by Max Alexander on February 10, 2016 at 2:34am

This is all very interesting - I didn't realise it would be such a complicated issue! I almost assumed there would be an easy method I hadn't spotted.

If it is of any interest/use, I constructed the model from a spline, with a variably pipe running along it's length. Attached is the spline itself.

Thanks though, if nothing else at least the cocoon mesh appears to give me a workaround! Cheers

Permalink Reply by Max Alexander on February 10, 2016 at 2:43am

Forgot to attach

One method I was trying to work was to shatter the spline, make a tube of each segment, and then boolean union them. I cannot however get the variable pipe to work over the whole spline, and am not sure whether this method would work anyway

Attachments:

Spline.3dm, 67 KB

Permalink Reply by Martin Siegrist on February 10, 2016 at 4:01am

So, I also have an idea:

Imagine an array of spheres along your curve.

You can create a loop and boolean union the second with the first sphere, then the third with the result of the first boolean union and so on...

Of course this creates a lot of edges but it's fun to look at and I imagine this is some sort of 3D print simulation so these edges don't really matter...

Permalink Reply by Martin Siegrist on February 10, 2016 at 4:18am

An image for the concept in my previous post.

I once did the opposite for cnc simulation where I consecutively removed a sphere (ball end mill bit) from a brep...