enter image description hereHi guys, I already posted this on reddit blenderhelp but I think it kinda became a meme, and even though I fixed a few things my problem hasn't been solved.

I tried the bend tool and simple bend modifier with no success because it didn't give me enough control on the bend trajectory, I then tried the bezier curve modifier, I applied a bend on the X Axis, to have the wrist part curve, then a second bend on the Z axis to have the top part curve down too.

My problem is that I want the top part to be flat on the X Axis, because after a 180° Z Bend, the top part of the object now moves away from the X Bend trajectory, as it is now mirrored.

I tried applying Vertex groups on the bottom half so that the top wouldn't bend on the X Axis, but blender doesn't make a smooth transition between flat and bent parts, I get weird connection polys instead of a fluid twist (or untwist in this case).

Is there any simple way to bend an object in two different directions ? A bit like so https://blenderartists.org/t/c4d-like-bend-deformer/1146668

  • $\begingroup$ Completely rewritten the answer to address the need for the end of the fold to be flat. $\endgroup$ – Robin Betts Feb 24 at 20:36

You could go with the Bend, using limits and Axis/Orientation Empties in the modifiers, but IMO, you would get finer control over the curves using Curve modifiers throughout?

Excluding the flat region from the circular curve modifier

In order to keep the end of the folded-over region flat, it must remain unaffected either by the mapping of the entire object into the curve, or the subsequent deformation. The first is achieved by arranging origins.

  • The circle's origin is placed at its first vertex, and at World 0.
  • The object's origin is placed at the center of its X dimension, at its minimum Z
  • The object is then translated in X (effectively rotating it around the circle under deformation) until the part of the mesh we want to remain stationary is directly above the circle's origin.

enter image description here

The second is achieved by weight painting:

  • The circle Curve modifier (deforming in X) is aimed at a vertex group - it does not affect the unweighted region:

enter image description here

The bottom is curved, the top remains flat, and in place.

Adding a second curve modifier to fold over a section.

The folding curve modifier deforms in Z, its curve can be anywhere you like in X, but here shown with its origin in the same place as the circle curve.

enter image description here

You could flatten more or less of the flap with weight-painting.

enter image description here

| improve this answer | |
  • $\begingroup$ Hi and thanks for the input Robin. Well that's what I did, I used Bezier Curve modifier, but as I said, I'd like the top part to stay flat while the bottom curves around the wrist and have a clean horizontal bend for the top part.The bend tool didn't give me enough flexibility, I will have a look at the Axis/Orientation Empties in the modifiers. $\endgroup$ – Mathieu Neander Feb 24 at 19:20
  • $\begingroup$ I just wish I could have it bend around two cylinders, one placed vertically for the wrist curve, one placed horizontally, perpendicular to the first one, for the second bend. May be I need a CAD for this. I tried with Fusion 360, with no luck, there is no bend tool on this software ! While Autodesk has it in Inventor... $\endgroup$ – Mathieu Neander Feb 24 at 19:21
  • $\begingroup$ @NeanderMat see edit... Is this what you mean? This is just eyeballed. weight painting and adjusting the curve. It could be made into a perfect cylinder if that's what you need $\endgroup$ – Robin Betts Feb 24 at 22:23
  • $\begingroup$ Or.. you could even shrinkwrap to your 2 cylinders... depends.. if you need animated deformation.. $\endgroup$ – Robin Betts Feb 24 at 22:46
  • $\begingroup$ Hey Robin, thanks for the edit, the weight painting seems like it would do the trick perfectly ! This is what I was missing when I tried applying vertex group modifiers. The transition between the bent part to the flat one was totally messed up and not smooth at all. I am not 100% confident in my understanding of the first method you proposed though, I will give it a try and adjust placement of every different element. $\endgroup$ – Mathieu Neander Feb 24 at 22:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.