I suspect it’s because those faces have draft angle issues:
I think you could simply extract the lower surfaces and make new extrusions.
thank you Gijs
… wasn t there a nice post of @davidcockey ? - did you delete it ? / or maybe I am wrong.
I think the main issue is, that the helix-like-surface is a ruled surface, but not developable.
all developable surfaces are ruled.
not all ruled surfaces are developable.
see also the gauschen curvature that should be 0 at the entire surface to be developable.
@Gijs - I don t think drafting is an issue here, looks like we’re talking about a 5-axis waterjet-cutting (are we @bob5 ?)
yes Tom, 5 axis waterjet, just that the red surfaces seemed suspisous, that’s why the drafting came up
@bob5 you sent this in on tech support as well
if it is 5 axis cutting, then indeed it is not a drafting issue. Most likely it is then an issue with the surfaces having torsion.
Tom, how do I make the suface developable?
L. Land-V2.3dm (1.3 MB)
This is what I’m trying to acheive
I guess - from a strictly geometrical point of view - it is not possible with the given edges.
even more guessing: also not as long as you stay with the helix as one of the curves.
from a craftsman point of view:
…but i think it is possible to deform the sheet metal after cutting in the desired shape. (as the steel will expand / compress)
if you have access to one of this great 5-Axis machines - why don t you cut the part first with some add on, bend it, and then back to the machine, cut the edges and holes in a 2nd run ?
this will give a much nicer result, also because the holes are not the weakest position (with most bending / streching) while bending.
??
1 Like
@bob5 has multiple, similar posts about this topic, which is annoying. I previously posted in another of his threads: Flattening a 3-d helix - #16 by davidcockey
Due to the thickness of this part and the fact that it will be formed hot, there is no reason to be very concerned about starting with an exactly developable shape.
Why Gaussian curvature is not a good way to evaluate how close a surface is a developable shape: Verifying developable surfaces - #7 by davidcockey
Gaussian curvature and developability - #5 by davidcockey
Bob5’s part has a width to thickness ratio of 4:1. That means it when a phyical part it is bent within the elastic range it will deviate substantially from a developable shape due to Poisson’s ratio effects. The top and bottom surfaces will be anticlastic (double curvature in opposite directions). For the bent shape to be close to developable the width to thickness ratio needs to be much large such as a piece of sheet metal.
To greatly compound the deviation of Bob5’s part from a developable shape when fabricated the part will be made using a hot forming process. This means there will be large, inelastic deformation. The developable shape assumption is based on elastic deformation. With the hot forming there can be stretching, compression or both, and the final shape will depend on the details of the process.
Without expert knowledge of the forming process and more detailed analysis using the results of UnrollSrf or UnrollSrfUV on the not exactly developable surface may be a reasonable starting point. But I would consult an expert on the forming process before investing much time or money in trying to optimize the bland to be cut by waterjet.
After the hot forming the holes will be distorted considerable from cylindrical.
(Update - This is incorrect.) A developable surface is possible with the given edges.
1 Like
@Tom_P is correct and I was mistaken in my previous post. A developable surface is not possible with the given curves.