I need to somehow find the average surface of solid brep (meaning the middle between the top surface and the bottom one) and unroll it while keeping the lines marking the kinks in the brep. Any help would be greatelly apreciated.
Unroll Brep.gh (1018 Bytes)
You geometry is not internalized.
Is this for a press tool?
Take the bottom or top surface and offset it by half the thickness maybe?
I think you are supposed to use 1/3 thickness for sheet metal… Could be wrong and this might not be a sheet metal problem!
Yes it is a sheet metal problem. I have tried the same, but the thickness is not always the same and the geometry could be quite complex.
bend will always result in a small fillet instead of a sharp edge, filled radius depends on the metal type and its thickness (in the drawing thickness look like constant 2mm)
are you also looking for something that implements this fillet, or are just interested in the median planes?
I know. But I have to ignore the fillet altogether. I just need to be able to unroll these tipes of geometries, because I have hundrets of them. I will compensate for the rest afterwards.
got it 
I’m trying to understand why this fails on some surfaces which do not end up as planar… it might be something related to a not-wise use of cullPoints
I got simmiral results. If you turn on the initial geometry you will notice that some of the new curves are not in plaannar. Hence you can’t create a surface. That’s where I am bashing my head also.
this version works much better
BUT there’s one thing to say 
in this piece for instance:
this edge fails:
looking from backside: that face is seen as part of surface A, while it should be part of surface B
I don’t know if it’s good practice or bad practice, but I would have drawn that like this, without the red strip
this is one of the big troubles of reverse-drawing sheet metal
Absolutelly agree with you.
it’s anyway pretty easy to see if something is not working as it should (or if it’s producing weird results)
if you join the surfaces together and get more than one brep, then it means there’s something strange happening, for instance:
I went on similar root, but then turned the surface into a mesh and used Weaverbird to offset the faces to the average. Your way is way cleaner and clever. As far as the crapy breps, I will redraw them to fix this issue. Your help is absolutelly amazing!
don’t want to overextend this discussion 
and thanks for the kind words
I just wanted to point out that the “manual intervention” I proposed on the brep in my earlier post won’t work 
if you have a bend that travels just on a portion of an edge, this is how to make it be interpreted the right way: 45° chamfer
brep at branch {1} results in one single polysurface
it looks like it’s “sufficient” to just chamfer that portion of edge with distance = 2 units 
Yes, I saw that. Trying your sugestion right away!
