Hello there!
I just came across the asymptotic curves grid shell pavilion by Eike Schling. I would like to know if there is a possibility to create a network of asymptotic curves on any given NURB free form surface…? I tried the Bowerbird plugin, but the output curves are not smooth. Also, if asymptotic curves are only possible with the minimal surfaces. Is there a possibility to incorporate that workflow as well?
I am attaching the images for your reference.
Thanks in Advance.
while i cannot comment on the specifics, his thesis might give some insight:
You could try to get the minimal surface with the kiwi plug in, which is also developed by the TUM.
Hey Baris!
Thanks for your response, I will start looking into it. Would it be possible for you to demonstrate this workflow where a freeform surface is created, then it is optimized to minimal surface with Kiwi and then the asymptotic curves are generated with Bowerbird, or any other possible approach for that matter.
Thanks in Advance!
Thanks Dan!
Will definitely give this a go!
There are KIWI example files. I would suggest you start simple, and try out. For me the problem always was, the inital surface. I guess there are also surfaces, which never will work nice. Also always a question of tolerances.
Maybe you want to share your surface?
Sure, here is the file for your reference.
BASE SURFACE.3dm (890.9 KB)
Thank in Advance! 
Hi,
sorry for the late reply!
I tried it with KIWI and had no success. Had in mind to try it further but forgot about it.
Now I see, you did a nice progress with Millipede!
May I ask what the final goal is? And why a mesh?
I would guess everything is also tolerance realted. If you are planning to build it, the joints will need some gap, and the wood/metal will maybe layout smooth by itself.
Also, I just realized the strips do not unroll flat, what maybe because the millipede outcome.
So, the goal is to establish a workflow wherein we can convert any freeform surface into a minimal surface and then extract the principal curvature curves or/and the asymptotic curves for panelization. I believe they have unique properties which might help us make 90 degree nodes.
So, that is why I was asking if we can make a mesh out of the millipede curves. Because then it is super easy to panelize I guess. What are your thoughts…?
ok. I tried also a bit on asymptotic curves, a while ago. I guess not every freeformsruface works.
For me mathematical decribed sufraces worked well, as well as some stuff I tried with KIWI.
Understood, also, I was thinking if it would be possible to trace the principal curvature curves with anemone…? I think these curves would do as well, for all the generic free form surfaces. Also then we can calculate the asymptotic curves with a deviation angle which might not be 45 degrees as it is in the case of minimal surfaces. What do you think…?
not sure why annemone. Bowerbird traces them iteratively. Maybe someone with more expertise has an idea.
