Since upgrading to Rhino 7, I have been having issues trimming two fillets at their intersection. Pic01 shows the two fillets intersection, Pic02 shows the curve I make at the intersection of the fillets, but Pic03 shows how the tip of the intersection is usually tweaked, so I press F10 to bring up control points and then use Setpt to smooth out the end, shown in Pic04. This used to work good in Rhino 6 but since upgrading to Rhino 7, it has become much more difficult. I try to smooth it out using Setpt more and more, but still no go. Would be nice if they would just trim together without making curves, but if we have to use a work around, would be nice if that would work. Is there any other suggestions?
Hello - please post a file with the surfaces you are asking about.
Attached is the model I am working on in v7 format. There are several of these intersections in this model. -THanks
Sample.3dm (251.4 KB)
Hello - the intersections short of the surface edges in some cases:
In another I am looking at, everything seems correct but trimming fails… still digging.
In that case, the cutting curve appears to have been extended a very small amount by a line - I am not sure yet why this matters but it seems to.
I think I just figured it out, my tolerance is too high. I just lowered it to 0.001 and they are all trimming now. Why did I not figure this out years ago?
Nevertheless it ought to work if the intersections are taken at that tolerance - that is a small but by no means unusual or too-small tolerance.
Thanks for digging into it and let me know if you find a solution, would be nice to just trim using surfaces rather than having to create curves to trim on.
Well, trimming with the surfaces will work if they as accurately enough placed to work with your tolerance, of course - if your surfaces slightly out of place compared to tolerance that will not work, and is expected and predictable. I was looking at the curve that goes edge to edge on the surface but does not trim, I do not see why. Pull also fails on this curve.
I noticed these have 3 stacked control points at one of their ends, shouldn’t Rhino find these or is this allowed?
Hi Gijs - it is allowed, in that the object is not invalid but it is a bad thing. What I cannot tell is how this happened here - Intersect does not generate this as far as I can see.
I see this all the time with fillets falling short of the real intersection, as you can see below. I turn on control points and move those 3 points to the intersection with SetPt. I can then trim the surfaces properly with the fillet. It works most of the time unless there are many points near the end of the fillet and a bit of a hook on the end. It’s been this way forever, I’ve just learned to live with it. I’ve never complained and just figured everyone else was doing the same and maybe one day it would be fixed. It was a small price to pay for such a great modeling program!
Hi Jeff - please post a file with the surfaces that generate the fillet, and the fillet you get from them, and the surface you are intersecting. I cannot see anything useful in the image.
Whenever I need to make these kind of transitions, I use FilletSrf without trimming to generate all fillets, then using the edge curves of the resulting fillet chain to trim the surfaces. But I never make fillets that end in a zero width, I’d rather make transitions like this: (the last XNurbs step can be avoided if needed)
Here you go Pascal. Let me know if you need anything else.
Thanks (77.0 KB)
Thinking the drag & drop didn’t work. Here it is again via the upload button.
3_Surfs.3dm (77.0 KB)
Ah, this situation -
- the way out is probably exactly what you said,
ExtendSrfwill get it very close. I see what you are running into now. It happens where the input surfaces come tangent to one another and I do not think there is a better way than SetPt if it starts out very close.
Alright. Thanks, Pascal!
Just glad there wasn’t some simple fix I didn’t know about.
Have a great weekend!