Sweep One Rail Issue

Hi All,

We are attempting to create a parametric undulating surface that is divided into a series wood slats. These slats are cut from the surface in one direction (x-direction) with a 20mm gap between each slat. Each slat can have a potential width of 25,35,45 or 55 mm with a potential depth of 38,50,64mm creating a total of 12 different permutations. We have attempted to use the contours and jitter command to create a pseudo random pattern with the permutations.

I believe we have achieved the necessary math although I'm relatively new to this software.

Our problem lies in the sweep of the rail curve as it only seems to sweep for some of the contour lines instead of all of them. The files are attached and any assistance or general pointers is greatly appreciated. The surface is also derived within the grasshopper environment.

Thanks!

Attachments:

20151204-Ceiling Maths Flat.gh, 66 KB
Capture.PNG, 7 KB

Replies to This Discussion

Permalink Reply by Joseph Oster on December 8, 2015 at 4:53pm

Looks like you want one rectangle section curve on each rail? If so, I see two problems:

Too much grafting - I don't think grafting is appropriate here.
The number of rails doesn't match the number of rectangles.

I started playing around a little but quickly froze up Rhino/GH and my whole laptop! So not fun. :(

I'll leave it at that for now, maybe more later when I don't mind taking the risk of freezing up my computer. Really annoying that Rhino/GH hang up so badly like that...

Permalink Reply by Joseph Oster on December 8, 2015 at 8:25pm

OK, I looked at this again. At first, I looked at why the number of rails and sections don't match; I can see ways to fix that but it's complex and time consuming and I didn't want to bother...

So I took a different approach; I baked the rails and the rectangles so I could see and touch them. I found that at one end of the rectangles, there were multiple copies of three different rectangles that weren't attached to a rail. At the other end, I found the first rail had no rectangle associated with it.

So I deleted the extraneous rectangles and the one rail and wrote a simple bit of code to sweep the rectangles along the rails - no graphing needed, though I had to partition the lists feeding 'Swp1'. I sorted both the rails and the sections by 'Y', just in case, though that's probably not necessary (confirmed below). And guess what? I saw the same thing you did in your code; about 1/3rd of the rails in the middle worked and the rest did not!

So I looked a little further and discovered that the rectangles (sections) are at the end of the rail curves instead of the start - and all I had to do was 'FlipCurve' the direction of the rails and it worked fine!

So I went back and added the 'Flip' component to the rails in your code and it works too!

So what happens is that grafting both inputs to 'Swp1' has the same effect as the 'Partition' components I added:

And the sorting wasn't needed - CeilingMathsFlat_2015Dec8c.gh attached:

P.S. Instead of 'Flip', you could easily place your rectangles at the start of the rails by organizing your code a little differently - using the 'Endpoints S' output instead of 'Endpoints E'?

Attachments:

Permalink Reply by Mike Taylor on December 9, 2015 at 7:26am

My goodness. Thank you so much for the quick and effective response.

Like I said, I'm new to this and all this help is awesome. I'll be sure to post some images of the final ceiling when it's all said and done. You can see what you've helped us build!

RSS