opdfgdfzbcbzvs<awefgthjgfgdsedas
There are multiple ways of achieving those shapes. Interesting to know is whether the ridge curves are planar or not. A top view of the model would be really useful to understand the logic better.
a) planar ridge curves: field of multiple overlapping circles which split each other into arc-like segments. Those segemnts are rotated and lofted
b) non planar ridge curves: multiple Boolean Differences
etc
Thank you Kaspar!
I am not sure what you mean by planar? The ridges start from level 0 ground and goes upwards, they are not the same lever which is obvious from the images.
Do you mind explaining with a quick sketch?
This is the closest I get but still not close…
The left curve is an arc (in plane) whereas the right curve is a 3d curve. Both appear to be an arch from the top view
its a combination of both, the long ridge is planar while the 2 short ridges are non planar arc curves.
is this what you mean?
The spheres you used for the BD share one plane and have the same diameter right? If you change their position in space along the z axis and their diemeter you will get the curved ridge-curve from the top view.
the last image was just splitted circle curves.
is this what you mean? (they are all same diameter)test.3dm (361.6 KB)
Hello - I expect you’ll need to mix different methods - assuming the ridge curves are what you start with, this approach might be useful:
DianaShapesMaybe.3dm (141.8 KB)
-Pascal
2 Likes
yeah but maybe do it more extreme? but not to end up in a voronoi like I did kind of; )
BooleanDifference_multipleSpheres.3dm (3.8 MB)
Thank you for your comment Pascal,
very close except that I am trying to make it 3 sided rather than 2, how would I go on to do that?
Hi Diana - about the same, you just need to be careful about matching the surfaces across the creases for Tangency, with Average surfaces set, That will make a mess along the original ridge curves, so then go back and match for position to the ridge curves again.
DianaShapesMaybe2.3dm (101.0 KB)
-Pascal
Hi Pascal,
So I tried the basic one first which you showed in steps, I drew a new curve and extruded side ways in both direction, when I matchsrf it dose not turn out like your, do you know why?
Hi Diana - one crucial detail missing from my explanation: When you choose the surface to match to, make sure the ‘CurveNearSurface’ option is set to On. Do not pick a target edge (Enter) and then pick the plane as the target surface.
-Pascal
Thank you Pascal,
Now it makes sense.
Just one other question if I want to close of the openings as you can see from the rendered view, so its also similar to the other sides, how would you do it?
I did patch it but it looks to flat and abrupt
Hello - you can do the MatchSrf trick again on those edges - it will pull away from the ridge of course - the best is, if you want the ridge to fade into the plane, then draw the ridge curve that way to begin with (Last two or even three control points on the plane).
-Pascal
1 Like