Hi Lara,

re:
> I tried an other way...but the result was exactly the same. I made a offset of the two big circles
> and moved it for 1 degree together. Sweep along the segment.

I'm sorry but I still don't understand what the problem is.

So you've got this:


Looking at these edges by themselves:


In the Top view, the distance here on-screen:



Will be different than the distance here on-screen:



So yes that is true but it's just how projection of a rotated shape works.

If you have a regular truncated cone like this:



If you rotate it:




Then view the rotated cone in a top view projection it will look like this:






That's normal and it's just how geometry works with projection.

If you wanted the projection to look like this:



That would need a slanted cone rather than a regular cone:



- Michael

Hi, Michael.
I assume that it is necessary so.

Hi WN, yes it would be a kind of simpler case if the cuts were straight up and down like in your example. In Mala's case they are slanted.

- Michael

Michael, what's wrong with this option? The difference is there but less.

Hi WN,

re:
> Michael, what's wrong with this option? The difference is there but less.

Well it's different in shape from the original design which had these profiles on parallel planes:





In your version those spots are not on parallel planes and so not equidistant as measured in 3D. But that's the kind of thing you would need to do to make the top-view projection of it to look equal, similar to the slanting cones that I showed above.

It's not clear to me what the desired result is - should it be equal spacing in 3D space, or should it be equal spacing of projected shapes?

I think Mala will need to give some more information on that.

- Michael

Thank you Michael.
Yes, we are waiting for what Mala will say.

Thanks a lot for discussion. Best is I try to make it clear what the goal is.
Hope the drawing makes it clear.

Hi Mala, I'm very sorry but I don't understand the drawing at all.

- Michael

Hi Lara.
Can you tell me which method is closer to the goal, or none of them is suitable?
If you want maximum help, you will have to try to explain.

Hi Lara,

OK, Silikone is a rubber-like substance?

A winkle is a type of sea snale, so I will translate it as a bevel, or a beveled ring, or a chamfer,
or chamfered edge surface.

You require that the chamfers be the same width?
And/or the same angle with respect to the outer torus?

The cross sections of the smaller torus-like "ring" will probably be different than each other,
perhaps non-circular, perhaps non elliptical, to match the chamfers.

The slant cuts are at different angle directions with respect to radial lines from the center of the big torus, which caused the different bevel widths.

Make the bevel cones the desired uniform width. Then make the smaller pseudo-torus.

-Brian

Bobbel has numerous translations from Dutch.

Extend the cone-like chamfers, to make them wider, then cut them at the desired uniform width.

Then make the smaller ring to match the new chamfer edge.

-Brian

I imagine that the "cones" are oblique, and not true geometric cones?

I made a drawing. Hope this will make it clearer.

As yet explained above your sections are inclined and not perpendicular to the tube of the ring!
So that can't be circular but ellipsoïds!

So if you want the same measure between & along the 2 "perimeters" internal / external: you must draw 2 ellipses !

If I understand what you want, I think Moi3D is not done for this.

You want to have this constraints :
- having a tore with 2 diameters of sections.
- having joining size regular in measure

Moi3D is not a constraints based modeler.

The result you want to have depends on a constraints resolving. The solution depends on :
- diameter of big section,
- diameter of small section,
- diameter of the tore (or ring). The main difficulty is here. ***
- size of the junction you want.

*** in fact, as you explain it, doing it with a cylinder is easy. But with a non linear path, you have to cut the small tore (or the big one, it depends of wich one is the "reference") with different inclinaison. It's section become en elipse.


You don't have a universal solution. I think, each solution depends on the complete set of these parameters.

Perhaps I'm wrong, but I think it's not possible to give to Moi3D and let him processing to have a solution.
To do it, you have to calculate yourself all values and give to Moi3D the right measurement you have calculated before.

Sorry, I was wrong...

The WN's method gives a soluce...

I have tested and you can test it with this file.
The test uses a big size of junction (3 as radius)


The modeling is done at the end with a sweep and two rails section by section.

Sorry but where are the similar gaps between the different diameters in these soluce ?

Is this what you are after lara?

Nearly a circlular cross section, and hypoallergenic.

- B

Hi Lara,
So this: """"""How would you place at beginning the cutting objects in order to avoid the distance differents at the lofts? Ideas?"""""""

From this post and image: """""""http://moi3d.com/forum/index.php?webtag=MOI&msg=10096.37""""""""""

Is answered this way... Notice you spoke about how the result is identical, "Under and reversed" in that area...

That is because your chamfers are going in 2 different directions, so the up and down side of the "slant" is opposite.

To get the symmetry in the place you want, with equal proportions, your "cutting objects" should be a "Mirror" of each other (Not a copy). This reverses the "slants", so they are opposite, and produce results that "face and duplicate" each other...

"Copying" the cutting objects will have the "reversed direction chamfers" produce "reversed symmetries"

Let me know if this is too hard to follow.