Does anyone have any idea how to make the geometry on the bottom half of these pictures in MoI?

I have tried doing a curve intersection between a helix and a revolved curve (a sort of sideways )( shape) and then sweeping a tooth profile along that intersected helix with a flat sweep. I'm getting a profile where the "teeth" lean in one direction like this: /// instead of being in-line with the surface normals, centered around the middle of the helix like this: /|\. If I do a freeform sweep it just makes garbage.

I need to sweep along the helix with the profile leaning in the direction of the surface normals, but I'm not sure how I accomplishing this in MoI.

Thanks,
FDP

Hi FDP, maybe an approach that scripts the positions of the cylinders and booleans them similar to this:
http://moi3d.com/forum/index.php?webtag=MOI&msg=8430.1

- Michael

Here is an interesting paper, antibacklash, but it has two rollers per tooth, either cylindrical, conical, or spherical.
There is quite a lot of trigonometry and matrices...
There are 22 references...

https://www.researchgate.net/profile/Mark_Horstemeyer/publication/274581068_Parametric_Study_of_Meshing_Characteristics_With_Respect_to_Different_Meshing_Rollers_of_the_Antibacklash_Double-Roller_Enveloping_Worm_Gear/links/593248a245851553b6a9dd24/Parametric-Study-of-Meshing-Characteristics-With-Respect-to-Different-Meshing-Rollers-of-the-Antibacklash-Double-Roller-Enveloping-Worm-Gear.pdf

- Brian

A single roller, touching on both sides, is not going to roll...

Hi FDP and Friends ,
for the construction of this Rotating Helicoidal Tree
this is my little contribution for a possible solution:

1- files in 3dm with manual construction, with indicated
the names of the objects, for an easier understanding of the steps taken,
and the final result
2- files in 3dm (curves and reference profile) and file.nod , with the final results:
in this last case I used my last Macro
Helix on Ref. Curve
As usual Files at this link:
http://www.mediafire.com/file/3b823z33g7g4x96/Helicoidal_Tree.zip/file

Have a nice day to all
al

Hi FDM,
I created a model following this video https://www.youtube.com/watch?v=fBtyCWyZvRo
The fact that the helix is created in single loops is not perfect as they need to blend.
Hope this helps.
Barry

Michael, bemfarmer, speedy and Barry-H, thank you all so much for your help. I haven't had a chance to go through this today, but it looks like I have several good answers here. I will let you all know how it works out later this weekend!
-FDP

Hi Al,
impressive nod thanks for sharing. The problem in question is the pitch of the helix must be equal around the curve but at the moment the pitch is equal over it's length. In the attach gif for 1 revolution of the cam the
roller moves 20 degrees. Perhaps a nod could plot the points required to produce the helix or do a Boolean difference every step to get the basic shape. I have produced a nod to do the basic movements but now stuck on best way to go ie:boolean or points any help would be great.
This maybe of interest. https://archivesmse.org/api/files/view/122245.pdf
Barry
Ps Added new gif manually created helix on hour glass shape.

Yeah, that nod is very impressive looking! I'm having trouble getting it to run though (so for now I can just marvel at its complexity). I have version v.1.0.rc3.2018.03.09 of node editor installed on MoI 3 Sept 18 2014. Do I need to be running a different version? I'm both new to node editor and haven't moved on to MoI 4 beta yet, I'll give that a try later today. I've been using Grasshopper over in Rhino instead of node editor, but NE seems much faster from the few experiments I've done, so I'm going to give it another look.

With this geometry one thing that is easy to overlook is that you want the following contact points for the rollers:



The easiest way to do this should be with two swept rectangles (representing swept cylinders), each representing one side/face of contact. I'm still going through the material you guys have shared and will let you know if I get the model to work on my end! Again, thank you all so much!

Hi FDP,
I have modified the method to achieve the helix's to loft.
In the youtube video the helix's where created by two radii and distance between them thus creating a conic helix. The method I have used produces a helix on the surface of the hour glass shape as shown in Photo. You will need to produce 2 Outer & 2 Inner hour glass shapes to create the Cam Roller Groove.
Cheers
Barry

Anti Backlash would seem to be of benefit.

It seems to me that the worm should be cut a little bit smaller than the space between two cams, and
then the left and right halves of the worm should be separated by (half) the space difference.
(Or equivalently, be cut with the separation initially.)

The center of the worm is not doing anything. Less than half of the worm is driving the cams at any time.
So performance is much reduced...

The geometry of the Double-Roller seems to not have these drawbacks...

- Brian