algorithmic modeling for Rhino
I'm trying to make a hyperbolic paraboloid using perpendicular parabolas on a grid system. I'm trying to make parabolas that are planar, but shift in the x,y axis.
The attachment shows what I have set up, but the curves I made are not truly parabolic. They need to connect with each other, as well as connect to the circles on the grid I have. Is there an easier way for me to do this?
-Colin
Tags:
If you cut a true hyperbolic paraboloid with planes parallel to the xy plane you will get hyperbolic intersections (by definition), not circular arc ones. How does this effect what you wish to do?
Chris
Here it is in Rhino. Create the hypar using parabolas, extract some UV curves (or use planes to get intersections) from the asymtotes up to about halfway to the circles, get the arc of the circle for the final curve then loft these. This gives one quarter of the final form. Next use mirror/copy to complete. Trim back along the intersections of the surfaces using isocurves and then loft. In the very center at the saddle point use network surface. This requires duplicating and splitting various curves. I'm not sure how you would do all of that in GH. Be sure to unhide in the 3dm to see construction lines.
Chris
In the description above I should have said that I used network surface rather than loft in the second sentence.
This is a much better result. Here I trimmed out the saddle section of the hypar and then used the same curves from the above method to generate the transition surface to the circles. I used blend to join up the pieces. It's not perfect but looks pretty good in the environment map analysis. This seems like it would be fairly easy, except perhaps for the blends, to do in GH.
Chris
This is EXACTLY what I was trying to do! Can't wait to see how you did it in grasshopper :D
Here's what I've got. I was unable to get a trimming solution in GH so you have to bake the trim curves and the final modified hypar together and trim in Rhino.
Chris
Thank you Chris! This has been extremely helpful :D
I think one of the problems in my file was that i set up the grid in a way that the parabolic function couldn't go to the points I wanted it to.
The way you did the model was easier than what I was doing, so thank you very much :)
Glad to hear that this was helpful.
Chris
hi dimitry,
i put a question about the reverse-way-to-make-a-hyperbolic-paraboloid-surface:
http://www.grasshopper3d.com/forum/topics/reverse-way-to-make-a-hyp...
i am so glad when u could have a look on this. thx!
Welcome to
Grasshopper
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
© 2024 Created by Scott Davidson. Powered by