geometry question

Hello everybody.

I am working on some building that we are drawing plank lines on a
surface because we want to build it with wood.Still don't have the right
shape, but I am preparing grasshopper to get some plank lines, then
offset them at the surface, and finally loft them (will do more things
later, like the joints in the picture).

I used .rhp file called toycar.rhp, that I think came from David.
I am quite happy with the result, but i have a question.

As you see, I unrolled the surfaces. But the thing it is that they are
not straight. And i think it is because the surface is not a revolution
surface, and as it has beeing modified the plank lines will not be
straight.

Am i right? People at my work say it is because i did them wrong

Attachments:

plank.zip, 1 MB

Replies to This Discussion

Permalink Reply by taz on April 14, 2010 at 10:37am

A quick sampling of people looking at plank lines or geodesic curves seems to indicate you can make it work with freeform surfaces. There may however be limitations. I'm not quite sure.

http://eat-a-bug.blogspot.com/2009/05/geodesic-shell.html

http://biosarch.wordpress.com/2009/07/22/plank-lines/

http://thegeometryofbending.blogspot.com/2009/05/surfaceturtle-and-...

http://www.grasshopper3d.com/forum/topics/one-more-push-before-the

http://www.grasshopper3d.com/photo/photo/listForContributor?screenN...

Permalink Reply by javier zaratiegui on April 14, 2010 at 10:46am

The plank lines i got with the toycar.rhp attached file. but the thing i am asking if would be possible to get straight surfaces or if those are only from spheres, cones and so.
I am not sure if i am clear.
I got the plank lines, i got them offset, i unrolled them and can print them. the thing is if they could be straight, or that depends on the surface

Attachments:

toycar+grasshopper.PNG, 203 KB

Permalink Reply by Mark Cabrinha on April 15, 2010 at 12:36pm

I remember looking at the toycar a while back. Are you using geodesic curves in the latest version of grasshopper? You can get geodesics (AKA short path, or plank lines) on most any surface, it just may not be the result you were expecting. How you develop the lath geometry from those single line geodesics is very particular. For example, doing this "manually" in Rhino, I used the fin command to extrude out the geodesic from the base surface, and then loft from that extruded fin to both sides of the fin. This should unroll has a straight surface.

Permalink Reply by javier zaratiegui on April 15, 2010 at 12:46pm

thanks!
will have a try tomorrow.
got to finish other things first.
thanks everybody!

Permalink Reply by Mårten Nettelbladt on April 14, 2010 at 2:47pm

Hi Javier, like Taz says, I think you can find plank lines in more or less any surface.

When you say you offset the plan line, what does that mean? I have a feeling that this may cause the error. Only the centre line of the plank line will actually touch the surface so you'll have to be careful how you offset it.

You could try another approach:
Create "cross section lines" along the plank line and loft them to give your plank line width.
Make sure the cross section lines are always:
- perpendicular to the plank line
- extending on both sides of the plank line (centre point of cross section line = on plank line)
- lying in the tangent plane of the surface

Good luck!

Permalink Reply by javier zaratiegui on April 15, 2010 at 2:47am

i didnt made me understandable.
if you see the png file i attached, there is writen "unrolled surface is perfectly straight"
i get my unrolled surfaces but are as in the picture inside the zip file i uploaded first.
so the thing i wonder if i did something wrong or this is because the surface is not a primitive.

Anyway, thanks everybody

Permalink Reply by Mårten Nettelbladt on April 15, 2010 at 5:16am

The surfaces in the png image are straight not because their plank lines were better than yours but because of the way those plank line surfaces were generated. They were generated by lofting cross section lines like I described above. However those plank line surfaces are NOT correct developable surfaces. Please have a look here:

http://thegeometryofbending.blogspot.com/2009/03/single-curved-cone...
http://thegeometryofbending.blogspot.com/2009/03/more-cones-testing...

RSS