Here's my first crack at using MOI in a serious manner to model some Ferrari Challenger rims. My biggest issue at the moment is the surface profile '0' seam that MOI generates. For some reason, this initial seam is not a friendly trim area and I'm curious how others battle around it. I don't have this issue in with my workhorse CAD software.

Hi Jason - every closed NURBS surface has a seam so it isn't really something that can be totally avoided.

One thing that can get tricky is when a seam is not quite lined up with edges from adjacent surfaces. It is particularly tricky when they are pretty close to being lined up so that it is hard to notice the small edges.

That's what you've got in this case - it looks like the main issue is that you can't get that one piece filleted?

If you select what looks like the main edges there:



If you zoom into those 2 circled areas above, you can see there is actually a little tiny additional edge in each of those spots:



If you select those 2 little tiny edges in addition to the longer edges, then fillet will work on it, (radius 0.3 here):




There are some things that I should be able to do to improve this in the future, but for the time being when you have a seam like that and you're not getting a proper fillet, zoom in to the seam area to see if there are any small edges in there.

- Michael

Thanks Michael, unfortunately I was aware of the little edges so I wanted to know what to do about fixing that. Unfortunately seeing that I can't avoid going across the seam, I am not sure how to best deal with it. However no excuses, I'm still not very prolific in MOI and precision placing and mirror with continuity so I'm sure that attributes to it as well.

Note, I didn't put a fillet in because I couldn't get it to match the 0.5 fillet I have with the rest of the pattern. Perhaps what I should do is model only a pattern portion which could avoid this issue altogether.

Hi Jason, yes a pattern approach where you focus on one slice that is replicated can get some more control over revolved seams, since that doesn't end up with just one area of the structure having a different topology.

But it should be possible to get rid of those little edges in the case you have here with a bit more precision in just a couple of the pieces.

One thing that I would recommend is to build this object centered at the origin so that you can make more easy use of the axis snaps to make sure pieces are fully aligned, and it is easier to use the origin snap as a center reference point.


Examining your curves a bit here, I took a look at this curve:



And zoomed in closely to the circled area to see how it was aligned to the revolve profile (which is the vertical line in this view):




That one looks good, it is lined up right on it.

Now same thing on the second inner profile (I think these were lofted together to make the protruding surface part?):



Zooming in a bit:




It looks like that is probably the source of your small edges right there - the endpoints of the segments of that smaller piece are not aligned horizontally with the revolve axis.

If that piece was aligned, it should made for an exact join between the revolve seam and that adjacent protruded surface seam, avoiding creating any tiny edges.


Another separate issue than that is curve segment tangent alignment. If I take that original curve I showed above, and use Edit/Separate to break it into individual segments, then delete half and turn on control points:



Creating an extended horizontal line through the first point and zooming in there:




You can see that those points are not aligned horizontally with one another - that means that the curve tangent is not exactly perpendicular to the mirror axis, which means the mirrored parts are not totally smooth, it would help with filleting if those 2 points were initially horizontal (you can use Transform/Align to help ensure this), making those meet up in a smooth manner. Things that meet in a sharp manner tend to need more difficult fillet processing to try and match the ends or to attempt to put in corner pieces.

I would like to make some of these things more automatically adjusted in the future, but for now it is something that you need to watch out for. It tends to help out a lot for filleting in particular to have the original curve framework more fully aligned up.

Hope this helps,

- Michael

hehe, I knew I was guilty of some poor curve layout... I did mention that as a possibility right? ;)

Regardless, thanks for picking this one apart. I will have another go at the wheel.