Hi Olga,
> Did you try the opposite side of the first ring and the other ring to the right?
Yeah, I did try both sides of that one ring, it seemed to fillet fine over here. One side has edges broken up into 3 pieces (this can happen not always just because of smoothness always but due to the "seam" edge structures on closed tube-type shapes that are booleaned). But just select all 3 edges on that side and it should be fine.
Here is the ring with 4 edges selected (3 on one side a and one on the other):
Then applying a fillet of radius 0.75:
Everything seems to be working well there for me on that ring, are you trying to apply a radius of larger than 0.75, or are you not selecting all 3 edges on the one side?
If you look at the bottom of the ring there, you can see that there isn't very much room left there when using a fillet of radius 0.75 - very much more than that and the fillets would start to run into each other.
This is the first ring that I tried to fillet - the one that is to the lower left in the top view.
The tight curvature of the other one that I was referring to was this area on the ring that is located higher in z than the others:
You see that shaded line there? That is an area where the surface is bent at a tight radius, that tight bend in the surface will greatly limit the radius of fillet that can fit in that spot. If you want to apply a fillet around there the curvature of the surface will have to be more gradually curving instead of going through such a tight bump right there.
The other ring in the middle has some kind of malformed area in the very bottom of it which may be interfering with its filleting:
If I turn on surface control points, I can see a kind of messy structure there:
Fillet is a sensitive command and cannot really deal very well with surfaces that have pieces that kind of fold around back on top of themselves and self-intersect. That seems to be what is happening down there in the bottom tip of that ring surface.
Was that possibly made with another scaling rail sweep? I see that you may not have had an actual corner in the scaling rail there, but it looks like the scaling rail may have been come to a kind of very steep down-turn at the very end there, that can also create some difficulty for making a well-formed tip, having something like a very steep "cliff" profile is similar in function to having a corner...
You may want to switch to Network for some of these things where you want to come to a kind of flat-shaped tip rather than sweep + scaling rail.
- Michael