https://www.youtube.com/watch?v=0WbaRZlUhNM

Great find by Stefan, thanks for reposting on YouTube.
Did it have a thread yet?

Lovely how his design can be tuned for the perfect air gap, per magnet.

Anyone with a 3D printer ought to be able to do a decent replication attempt if he's shared the files (he likely has but I didn't check), even easier.

My mind immediate goes to finding a way to curve the magnet (radially and vertically) so the balls or rotor can stay in a flat round plane with the smoothest torque possible. Sping load might substitute gravity to an extent, but you want a motor to work in any orientation and allow for shaking during operation.

Any serious reason to doubt the validy of the presentation?

That's a fair comment.

It'd love to see that on a glass table and then balls added until they start to interfere.
A clever feed ramp or twist and stop might manage to get in two or three balls per magnet.

If anyone wants to waste his time, here are the original STL files.

But it doesn't work. Like any other SMOT, V-gate or whatever. It's never possible to close the loop.

Was discussed here a while back between a few builders
The perfect synchronizing of the spheres/balls
180 degrees out …


Was mentioned as odd


 

Was discussed here a while back between a few builders
The perfect synchronizing of the spheres/balls
180 degrees out …

Was mentioned as odd

Good point

Might be possible that the second ball would happen to enter at the precise
moment for them to be 180 degrees out and remain so ?

2. There is no reason why it should work, the energy present as attraction
peaks when a ball is nearest a magnet, both in approaching and in escaping.

3. If this could work, there would be no need for a high point in the ramp at all.

At first I thought it not very likely the balls could sink 180 out.
But upon thinking it through some more, if it works then why couldn't they
sink at 180 out ? So, I edited my previous post.
  But...
2. There is no reason why it should work, the energy present as attraction
peaks when a ball is nearest a magnet, both in approaching and in escaping.

3. The energy needed to pull or lift a ball up, to some specific height, is the same if along
a long ramp as if it is along a short ramp, either way.  All that counts is the height of the lifting.
The same applies to its falling / rolling down.

4. If this could work, there would be no need for a high point in the ramp at all.

5. If this could work the ball could be mounted upon a swinging arm and the magnets
placed around the outside circumference of its swing.  This does not work.

But I support anyone building experimenting, learning and sharing.

My mind immediate goes to finding a way to curve the magnet (radially and vertically)

Flexible plastic magnets will help you. In modern motors, this is mainly. It can be tied in a knot.

Things to consider:


Angle of incline on acceleration (towards center of magnet)


And angle of incline during deceleration


Also momentum at the center point to overcome the small incline
(longer horizontally, and more gradual vertical transition)
and if this is enough to overcome the reverse force that would
otherwise pull the ball back up the shorter distance (steeper vertical)