Surface hole in closed polysurface...?

Hi forum,

I’ve been pulling my hair out on this one, I’ve just sent this ring off to be cast and they responded with a hole they found in the supposedly closed polysurface. Not only can I not select any of the edges around the hole but I can’t even redo the form as my boolean difference tool is now giving me boolean unions…I’m about to throw my laptop through a window, please help.

Cheers,
Stephen


Lawrence_Ring_2.3dm (5.0 MB)

Hi Stephen - there is almost certainly a small loop in the trim there, and most likely a fold in the surface of the number 3 there. Please post the file and I’ll have a look, or send to tech@mcneel.com with a link back to this topic in your comments.

-Pascal

Thanks Pascal, I uploaded it between the screenshots. I’ll email across too though

Hi Stephen - this surface

has folds in it - I see on untrimming it that it is very deep compared to the depth of the cut - when flowing these I would make the object only slightly deeper than they need to be for the depth you want. Another consideration is the inputs that generated the surface to be flowed - font curves are not generally very good for surfacing and flowing them will not be a happy thing in many cases - when in doubt, which is most of the time with font curves, clean the curves up before making surfaces - make sure all tangencies are correct - usually they are not - and where possible replace multiple tangent or near tangent segments with simple G2 curves. I say all of this not having seen the original curves or objects that were flowed but I am pretty sure I am right - they should be only deep enough to make the cuts, and made from better curves…

-Pascal

Thanks for your help, Pascal,

I wrote out the characters, extruded them, then replaced the spine before making the cuts on the master. This is the only character that I had an issue with, there is also a 3 on the opposite side that worked fine. For future reference, was there something that I did wrong, or should have included through that process?

Also, so I’m clear, to correct this issue, I should be starting again from the beginning or am I able to correct the issue by addressing the folds?

Thanks,
Stephen

Hi Stephen - the worst case, I think, is that you may need to re-do the 3. Can you post the curves, and the input to the Flow operation? I’ll take a look (in the morning)

-Pascal

Lawrence_Ring_3.3dm (5.1 MB)
Thanks Pascal,

I tried to untrim then mirror across the successful 3 from the other side which I think might have worked after I moved it across so the boolean difference was cleaner.

I’ve attached but I can’t see the same issue now so I think it’s completed.

Thanks again for your prompt and effective help!

Cheers,
Stephen