I am trying to create a ring for 3D printing. Part of the requirements for 3D printing is to not have any non-manifold geometry in your model. When checking whether I met this requirement, I found that my model contains two edge loops consisting of "Boundaries" non-manifold geometry.

Highlighted in orange are the non-manifold edges of type "boundaries" (Highlighted using Ctrl + Alt + Shift + M, only Boundaries is toggled so that I could narrow down what type of non-manifold this is): Non-manifold edges (boundaries)

When I zoom in and move some of these edges aside, I find that there exists a duplicate edge loop underneath!: Here, you see that there is a duplicate loop hiding underneath my loop

I undo that translation, reselect the non-manifold edge loops through Ctrl + Alt + Shift + M, and I "Remove Doubles". Here, I find that the hidden duplicate loop no longer exists, but now my non-manifold edges are of type "Non Contiguous": Non-contiguous edge loops

I read that non-contiguous edges are due to face normals being in opposite directions(?). Turning on face normals, I get this: Face-normals

Honestly, I am new to 3D Modelling and Blender, and am not sure how to correct my issue from here. Do you guys have any advice?

Edit: Here is the .blend file prior to the modifications mentioned in this post:


1 Answer 1


In this case:

  • Select all vertices, and CtrlV .. R remove doubles
  • Then, with all still selected, CtrlN, make normals consistent.
  • $\begingroup$ At first I tried this with just the non-manifold geometries and it didn't work, but explicitly following your instructions and doing this to all of the vertices did the trick. Thanks!! $\endgroup$
    – Logan
    Oct 6, 2018 at 17:52
  • $\begingroup$ Sadly, this no longer works. In 3.5, it broke me mesh (thankfully undo works better these days!). :) $\endgroup$ Apr 29 at 2:26
  • $\begingroup$ Hi @SeanRobertson.. sorry you had a break! I just retried this, (with updated shortcuts: M for merge, and Alt-N for normals) on the original heart mesh.. and it worked for me. It''s not exactly a very general Q or A, though.. more one of those 'fix this particular case' questions. Not sure it's very useful to the archive. Maybe it has to go. $\endgroup$
    – Robin Betts
    Apr 29 at 6:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .