Join strategy.3dm (911.9 KB)
Now that I’ve learned how to take best advantage of join, I have 3 small naked edges left over. I’ve tried trimming behind those edges, but that doesn’t work, likely because they don’t actually come together. For these particular edges, I’m not sure what tools to employ to get rid of them. As you can see, most of the similar edges in the model joined just fine. thanks, rex
Hi Rex - in wireframe you can see the surfaces under those corner fillets are not actually trimmed back. Here it is fixed up, but it is a little bit of a nuisance to do, I’ll make you a little clip shortly so you can see some of the things to look for.
Join strategy_PG.3dm (591.9 KB)
Here’s a clip -
note I switch to Wireframe to see the structure of the thing more clearly, and I switch to parallel view when I need to zoom in really close.
The command near the end, testRemoveAllNakedMicroloops will not autocomplete on your system - type the whole thing.
-Pascal
Hello,
It looks like you could use fillet on this model on the top edge instead of trying to (sweep)? I may have used a different strategy in using mostly Booleans for this but it could be an alternate method if the corner fillets are acceptable.
Pascal, I made it through your procedure. Thanks much. I’m not sure what that final untrim border accomplished though or why it was necessary. thanks
Also, I can’t find testRemoveAllNakedMicroloops on the Mac version. Is it there? If not, how do I get rid of that open point? I use both Mac and Windows. thanks, rex
Also, when it came time to trim the two side surfaces at the edges of the little compound triangle, why was it necessary to trim with a curve rather than being able to use the triangle surface?
The reason I use the curve when trimming with fillets is that fillets by definition just barely touch the adjacent surface - when tolerances and surface to surface intersections come into play, it is not uncommon for the curve of intersection to be incomplete and/or take a long time to calculate. Since I know the edge curve is in fact the thing I want ti trim with it is quicker and more reliable to specify the edge curve using the CRV filter.
-Pascal
By monkeying with the surface control points (delete points and SetPt) the existing trimmed edge will move around. I wanted to have a new clean accurate trim.
-Pascal
Pascal, I often have this kind of problem, and it is nice to have this video to use when I try to fix them.
However I can’t say I understand totally what each step did. I won’t ask you to explain step by step here, but it would be useful to have a video with narration sometime to explain each of the steps. I think I would learn a bunch generally about the subtleties of surface editing.
For instance I have no idea what testRemoveAllNakedMicroloops does or what created the conditions that made it necessary.
Nick
Yeah… this is a bit of test command magic. It does pretty much what the command name suggests - it finds those pesky, hard to fix ‘point’ naked edges and where possible rearranges the trim, microscopically, to eliminate the looped curve.
-Pascal
It’s there, but as mentioned, being a test command it does not autocomplete- you must type the whole thing. Correctly…
(copy/paste works)
-Pascal
I now have a closed polysurface and am a lot smarter. Thank you!
Rex
Hello, five years later. By doing myself some research about the same issue (naked points)
I came across this topic.
I could not resist, and I did myself the solving process on the same file. Works fine (I use Rhino 6, so there is the command ‘‘remove naked macro edges’’)
I do have one question though: Nothing appears to be naked on the design. But when you zoom in, the curves seem to do not touch or interact.
So, I wonder Is this something only just visible? Or is still a real issue considering 3d printing prototypes?
Thank you
Spiros
You are zooming in to high enough magnificantion that you are seeing round-off errorw in the display graphics calculations. If the curves originate with intersecting surfaces or similar then sometimes there will be gaps smaller than the absolute tolerance setting due to how NURBS math works… Use CrvDeviation if you want to verify the curves intersect…
Ok I see,
I thought ‘’CrvDaviation’’ is used for overlapping curves and not for distanced curves.
Thank you David
Assuming the absolute tolerance is appropriate to the end use of the model. The basic rule is absolute tolerance should be at leas one order of magnitued smaller than the accuracy required. So if your manufacturing systems needs a tolerance of 0.1 use an absolute tolerance of 0.01 or 0.001. The disadvantage of too small a toleranc is the model may become excessively “heavy” with many control points.
https://wiki.mcneel.com/rhino/faqtolerances
CrvDeviation returns the minimum and maximum distance between curves and marks the points of minimum and maximum distance.
Much appreciated David, I will do my homework!
Hi, I did everything I could to join surfaces. I even used the phrase “remove all micro edges.” but there are two dots. I want to make a shell using a specified thickness. Please refer the attched model and give me a advise
900mm V10.3dm (475.5 KB)
@Emmanuel_Fernando you have a lot more issues in that model, for example:
It looks like you have forced joins with JoinEdge, is that correct?
If I explode and _RebuildEdges, I get this:
What can I do to fix it. Pls let me know in steps
