Contact set is not really new ..

[WHolzwarth]

.. and still has lots of bugs. I can show some more.

I've tried adding only one of the outer indexers to the contact set. But even this additional member failed.

Walter

[admaiora]

Maybe you are right. But at first sight there is not a perfect coupling.

Central blue part diameter 92.999 mm

Indexer diameter 93 mm

Solidworks imported geometry?

[WHolzwarth]

Yes, SWX. I'll have a closer look.

[WHolzwarth]

Hmm. The outer indexer is not perfect. See picture.

I'll do a perfect one. Let's see.

[WHolzwarth]

Hmm. I've grumbled some time with the fact, that Feature Recognition is overwriting data in other open files. AUTODESK should know about it.

See attached an updated dataset with same problems.

Kudos are provided for everyone with a solution.

[JDMather]

Do you have a YouTube of this thing working or link to original source - I have found other serious interferences in the assembly.

[WHolzwarth]

It's from there, Jeffrey:

https://grabcad.com/library/multiple-geneva-indexer-1

[WHolzwarth]

Well, I too found a collision, and changed a dimension to a better value. But still no motion of the outer indexer.

[JDMather]

When I (or I should say - my students) get around to working with this, we will not fool with Contact Sets (we will go directly to Dynamic Simulation Environement.

But a couple of things I have learned about Contact Sets -

 - I don't generally try to move with mouse unless very simple assembly, instead I Drive Constraint.

- do not try to make "perfect geometry" perfect parts - allow typical manufacturing tolerances for clearances between parts. I would not expect a 10mm pin to fit into a 10mm slot.  There should be clearance - just like the real world.

[WHolzwarth]

Hmm, Jeffrey. As very often, different opinions can arise. Here's what I think:

- My first attempts are mouse movements in similar tasks. If I can see good response there, my next step is driving constraints

- I've seen bad response of contact sets with non-optimal parts. This is a very good sample for faulty behavior.

- Surely clearance is needed in reality. But in theory, I'm modeling in most cases with zero clearance. That means 10mm pin in 10mm hole or 10 mm slot. The nominal dimensions appear in my drawings, too, combined with appropriate tolerances.

Well, I had to do some changes for these parts, moving in Inventor. But in the end, I got it. Moving of constraint Winkel (or Angle) from 0 to 1800 degrees should show it.

😉 But I didn't succeed by now with DS. There's some stuff left for you or your students.

Walter