Solved! by wh. See the answer in context.
Here's my approach.
I've made changes to spindle and ball screw, and made an array of the balls.
In the end I tried to turn the spindle and move it through the ball screw, but no luck. I suspect, that this is another sample of Inventor's unsatisfying constraint capabilities.
Remark: I know, that normally the balls are moving, too, but I neglected that, because of simplification. And I know, that normally the spindle is axially fixed and just rotating, and the ball screw does the axial movement. I'm in doubt, if this version would be running better.
EOP of parts needs to be moved down. Files are Rel. 2012
Really it depends on what you want to do with this.
The easiest solution is a curve driven pattern of Workpoints in the screw part and then use that pattern to place the balls in the assembly.
If your end goal is Dynamic Simulation then a bit more work is required.
See Tips 43-47 here http://home.pct.edu/~jmather/SkillsUSA%20Universit
the slot and the ball are the same size (no allowance for clearance and manufacturing tolerance) - we can't make perfect parts in the real world.
the circle sketch for the coil is not perpendicular to the start of the helix - therefore the slot is not round in cross-section and there is interference between the ball and the screw and nut.
Start with a Natural start condition (off the end of the part far enough that only the helix is cut or use a single line Coil to create a surface body coil to use with Sweep. This will be needed to place and pattern the WorkPoints, so that is how I would do the part.
The blue "line" is your circle profile on edge view.
The purple curve is the Helix.
Notice that the blue line is not perpendicular to the start of the helix.
The orange "line" is a Workplane created perpendicular to the start of the helix.
very few people catch this mistake that results in incorrect geometry.
I am using 2013 - so you would not be able to open my file even if I posted it.
In this image there are two circles - the 6mm circle you used with Coil to cut the face shown in orange.
I cut back overtop that feature (red) with a 5.99mm circle and notice that it cuts the sides more even though the circle is smaller.
My circle is perpendicular to the helix, yours is not.
I made a mistake earlier when I said one solution was Natural on the start of the Coil. That is the default. You would need a flat start to have the circle perpendicular to the coil curve. But don't use that technique unless you start the Coil well off of the part, and in any case it will give you results that are harder to understand than simply doing a Coil surface of a line and using the edge of the surface body helix for a Sweep.
I just realized I left out some steps in that other document - see this one
Access a broad range of knowledge to help get the most out of your products and services.