Folded Diamond Pattern

Replies to This Discussion

Permalink Reply by Laurent DELRIEU on August 5, 2017 at 1:00pm

You can look at this the quads are also folded in 1 direction

http://www.grasshopper3d.com/forum/topics/free-form-and-holes?xg_so...

Permalink Reply by Birk Binnard on August 5, 2017 at 6:58pm

Hi Sarah - I've got a GH file for something very similar to your photo:

If I had used a zero value for the fillet radius on the twists the result would have been exactly what you are looking for. The key to making this geometry is to start with a single twisted shape like this:

and then add it to a second one that has the twists done in the reverse direction. That is exactly what the first image is. 

I design things for my 3D printer but I have not printed this one yet because there is a fundamental problem with the way STL files handle what's called "overlapping shells." And that is exactly what this geometry is - 2 overlapping shells, each twisted in the opposite direction. I've been looking for a solution to this problem, but I'm not sure there is one.

If you want to produce a real part with this type of geometry just let me know and I can post a simplified version of my GH layout that does this.  And if you are considering 3D printing it, this blog page might give you some insight about the issues involved: Do the Twist

Permalink Reply by Birk Binnard on August 5, 2017 at 11:44pm

Just for fun I tweaked the parameters a bit and got this:

Permalink Reply by Sarah C on August 8, 2017 at 11:14am

Hi Birk! 

Thank you for replying to my question!

Could you show me how did you achieve this? That would be very helpful!

Permalink Reply by Birk Binnard on August 8, 2017 at 11:36am

Hi Sarah - the method I used is very simple once the untwisted shape has been established. However, because the final result is a combination of a left twist and a right twist the issue of overlapping shells is a major problem. What happens is the file saved by Rhino is not what you would expect - at least not if it is saved as an STL file.

Here is a picture showing a cross section of the part I showed in my first post about half way up: 

As you can see, the resulting geometry is not at all what you would expect. This is because, for some reasons I don't understand, the STL file seems to contain all the geometry from both the left and right twisted  models, even though only the desired geometry is displayed in the 3D views I showed.

My GH file has components that generate a bottom and top surface, as well as an inside surface, all of which are needed for 3D printing. Consequently it is fairly large. In a couple of hours I'll post a simplified version of the file, but if you are considering 3D printing this we should probably have an offline discussion since there are a large number of GH parameters to deal with.

Permalink Reply by Sarah C on August 8, 2017 at 11:48am

Birk,

I'm still a little confused on how did the left and right twist will produce the folding planes. But I will check out your gh files perhaps that will help me more on understanding it.

Thanks again!

Permalink Reply by Birk Binnard on August 8, 2017 at 12:57pm

Ok Sara - here you go.  I cut out everything except the components that create the basic geometry. I also put in a bunch of comments to help you understand what's going on. Feel free to email me if you have any questions

Attachments:

• sara1.gh, 32 KB

Permalink Reply by Birk Binnard on August 8, 2017 at 1:19pm

Oh my!  I'm not sure why it worked, but I added back the components I used to create the top & bottom surfaces to the GH file I uploaded here, sliced it, and the results were what they were supposed to be. For some reason the overlapping shell problems I encountered before did not occur. I don't know if this is because the filleted curves were gone, or what. But I'll try printing the part (which will take a day) and then post results back here.

Also, given that my original, untwisted shape is basically a surface of revolution I'm guessing the flat surfaces resulting from the double twist are not truly flat plane surfaces. If this is true and you are needing truly planar surfaces then my method will be useless for you.

Permalink Reply by Sarah C on August 8, 2017 at 5:19pm

It looks very similar to what i am trying to achieve but i'm glad to learn another method to this. I might just try to panel around to achieve the plane surfaces.

Permalink Reply by Sarah C on August 8, 2017 at 4:22pm

Thank you so much for your time!! The comments are really helpful in the file.