I was wondering if someone could tell me how to remove the dangling curve that is created from a FilletEdge operation.
I’ve tried many different approaches, of exploding object, extending surface edges, untriming, trim, split, etc…but can’t seem to find a way.
It is part of a trimmed (fillet) surface, not a curve. Something went wrong in the trim during FilletEdge.
Did you change the absolute tolerance after FilletEdge?
No I didn’t.
Is there any kind of other operation that can be used to get a fillet then?
The only way I can think of is to just build the geometry with a profile and guide curves,
in order to get the rounded geometry, instead of relying on the fillet.
Hello- it’s hard to say without the input object intanct - reconstructing the un-filleted part here, the fillets worked as expected.
Dangling_Curve_PG.3dm (146.8 KB)
-Pascal
Hi, Yeah what you have works but only if you select the consecutive edges and apply the fillet to both at the same time.
However, if you select one edge, apply fillet, then select next edge and fillet again it doesn’t work.
My original example was an excerpt from a model which has more edges. This is a model for a client so I couldn’t post the entire thing.
The reason I was going with doing one edge at a time is because on the original model by doing all edges at once things get messed up. So my strategy was to do one edge at a time and ‘clean’ up as I go before doing successive edges.
I’ve uploaded another of the same model exposing more of the model with more edges. I’ve keep a pre-fillet stage on the first black layer.
On the blue layer I’ve tried filleting the entire ‘loop’ edge at once. As you can see this gives undesired results at points along the edges.
So again, trying to do one edge at a time also still gives undesirable results like with dangling details:
The red layer was two edges done separately in succession and as you can see the spurious result happens between them.
The green layer was done with the two edges at the same time, but then the end of the two edges gives a spurious result.
So I am at a loss in the best approach/sequence of doing the fillets.
Looks like I simply have to go back to the original geometry and create it differently keeping in mind I’m going to create the rounded edges rather than filleting.
There is another way to make the fillet and BTW its a more accurate way.
Just extract (use ExtractSrf) and delete the bad surface that filletEdge made and replace it using the FilletSrf command using the “extend” option. The fillet size that fills the hole left by the bad fillet is .618034mm.
After making the replacement fillet you will need to untrim the end that Rhino trimmed incorrectly and then use the Trim command to trim that end using the edge of the adjoining fillet as the trimming object. (Rhino will always do this trim on extended ends incorrectly so the user always has to untrim to make it right)
Dangling_CurveX.3dm (142.1 KB)
The blue layer approach is good since the same size fillet is being applied to all the contiguous edges of one face.
Referring to the image below, to be able to fillet the entire loop, delete the red faces, and untrim and retrim the main 2 faces and the fillet face with the 2 short (dotted) edges.
CleanedHalfFor_Dangling_Curve_v0.3dm (122.1 KB)
Thanks Jim…I tried this with along one edge and it worked at one end, but at the other end the extend option didn’t go out past where it needed too, and I couldn’t do any manual extending/trimming with either the filleting part or the face it was meeting - leaving a gap that has overlapping edges at some parts.
Red layer with cyan fillet.
Dangling_Curve_v1.3dm (1.5 MB)
Thanks Steven. I was wondering what you mean by 'fillet face with the 2 short (dotted) edges.
I couldn’t get a final trim of the two main large surfaces(that were untrimmed) with the fillet faces/edges(the ‘sides’).
So I tried using the extend srf on the fillet edges to get an intersection that could trim the large faces.
It worked on one side, but because on the other side there was a couple fillet sections that would not let me extend (yellow section on green layer).
Dangling_Curve_v1.3dm (1.5 MB)
As I explained in my previous post → Rhino trims the extended ends on fillets whenever the user asks for the extend option in the FilletSrf command. This trimming is not at all helpful to the user. What it means is the user has to untrim the end to be able to do anything reasonable with it.
So you need to run the Untrim command and click on that edge and then you will be able to do the trimming or extending. You will be able to trim the end without extending, but it is a good practice to extend a little beyond where you need to trim.
You should have been able to extend trimmed edge of the fillet but don’t do that, it will give you garbage that will likely create more time consuming problems later. You already have more than enough garbage in your model that will consume a lot of time as it is.
Also, in your original model (Dangling_Curve) the fillet size was .618034 and on current model (Dangling_Curve_v1) the fillet size is .6. So you can’t use the .618034 fillet to fill a hole that was trimmed for a .6 fillet when your tolerance is .0001. Incidentally, a looser tolerance will probably also make what you are trying to do easier unless you really need your model to be that precise.
I was referring to the now yellow faces. Originally, they each had 6 edges, 2 of them with lengths shown in the image above. Now, they each have 1 short edge. Their trims are probably giving _ExtendSrf a problem. You can use _ShrinkTrimmedSrf then _UntrimAll on them before _ExtendSrf.
A couple of other short edges have emerged since the v0 model. Dots for them are in
DotsFor_Dangling_Curve_v1.3dm (34.8 KB)
They can be found by _DupEdge all edges, then using _SelShortCrv.
I tried starting over from scratch. Using a FilletSrf with size of 0.5
Along the first Side Face I get a clean fillet(no dangling) and what appears to be the same results either using:
UnTrim>Trim or UnTrim>Extend>Trim.
The reason I tried both ways was because I thought it might lead to a different ‘setup’(How Rhino might respond) for attempting to get the second Side Face fillet.
But it doesn’t seem to matter as when I try to do the second Side Face Fillet I run into a dead end in which no combinations of Untrim, trim or extend edge allows me to produce a fillet in which the two fillets butt to each other.
I run into the situation in which I can trim one fillet but not the other.
In my new file, I’ve separated each step I did in it’s own layer.
I’m stuck at the last layer ‘F - Untrim’ which is the untrimming of the second Side Face created Fillet.
I’m wondering what might be the next steps to get a completed fillet that I could reproduce going along further the Side doing other faces.
you have some strange surfaces in the corners where you have fillets. If you start again with the first posted file, then see:
fillets.3dm (885.0 KB)
Thanks Gijs…however when I try to fillet around the back as well, it gets messed up ‘Fillet_around_back1.jpg’
I tried the same with another file that I rebuilt those two trouble surfaces like you did but with a different radius…but same thing happens…good fillet around the front, but going around the back it gets messed up.
