I have an IGES file with naked edges that I need to repair. One problem is this trimmed radius surface, which needs to be joined on all three edges, two will not join.
My usual approach would be construct a new surface using patch or edgesrf but this produces a “heavy” surface. It will do the job but I’d like to find out if there is a fundamentally better approach. I guess I’m asking, is there a way to match a trimmed edge?
Apologies if this is a very basic question. I have been using Rhino for a number of years but am entirely self taught; I’m know there are many gaps in my knowledge and am attempting to fill them.
Well, if the corner partial sphere isn’t close enough to be within tolerance to the original edges after trimming, I suspect the only way will be to remake the corner using Network or EdgeSrf (which should make more or less the same surface). I would avoid Patch.
Its good to know that there isn’t a better way to resolve this than the approach I usually take. I will avoid the Patch tool, thanks for the tip. I have only used it a handful of times when all else fails.
The point structure of the surrounding surfaces proves that they aren’t as light as they at first appear. That explains why an as-light-or-lighter surface for the corner isn’t able to join all three. Are the surrounding surfaces (that you show with points on) tangent or G2 to the ‘flatter’ ones between them?
As Mitch suggests, I’d use Network, but brace myself for something heavy-looking that may well have one or two ugly ripples in it.
The surrounding “flatter” surfaces are tangent G1 to the radius surfaces, as far as I can tell (I used to have access to Alias and, if I recall correctly, it would tell you what the continuity condition was at surface boundaries. I never have found an equivalent in Rhino. I usually project curves across the joint and measure their continuity).
Thanks for the suggestion of network. It actually produces a reasonable surface, certainly good enough for this job (to give an idea of scale, the radius on the adjoining surfaces is only 1mm). It does create a singularity, which I’d rather avoid but I can live with it. It all joins together with no naked edges.
I did also try a method taught to me a few years back, which involves splitting the surface into three four sided patches. It works quite well but I wouldn’t want to rely on this method as it is rather time consuming. Still, when all else fails it does the trick.
If it’s a radius, I first try to use a real spherical surface in these cases. A 4-point sphere would be the starting point and then trimming with the three surfaces edges.
If it’s not a real radius, I then try with a sweep1.
