Magnetic repulsion converts into attraction

[Low-Q]

Vidar,
       Get ready for some 'constructive critisism'.
You have about 6 'like' facing magnets, very close together, going 'upward', untill they start to repel each other ------- correct?  The rest of the magnets, are 'fanned' out.
Unless, EVERY ONE of those close magnets, REPEL each other ----- AT THE SAME TIME ----- having the 'extra
magnets, would be a waste of time.  Only one pair of magnets repel each other at ANY ONE TIME!

SPANG.                           (BILL.)

Thanks for the critisism. Good to know that some actually look into the design and come up with some points, cons and pros about it.

The closed magnets does not repel eachother towards left and right. They are aligned in the same way as in the first video in this thread (My practical experiment with these magnets).

The idea with the 5-6 "closed" magnet pairs moving upwards is to simulate one solid magnet. I do this in order to prevent the magnetic path on each of those magnets to be spread all over the place, but instead be aligned angular to the magnets direction. This will weaken/take away the force that is required to escape the shield. Otherwise, separate magnets will attract to the shield, and prevented to escape from it, because the magnetic flux also point in direction towards the shield. The "pseudo solid" magnets flux lines does not. Instead the magnets separates into a "fan" away from the shield.

I am sure you have discovered that in the center of a very long magnet, there is very weak magnetic attraction and repulsion. This is because the magnetic flux is more uniform and straight within a given space around that center. At the ends of that magnet, the flux lines are very curved and there is big difference in flux density from one point to another (Which is the reason why there is greater attraction and repulsion at the tip of a magnet bar).

The magnetic shield is covering half the "motor". This shield is guiding the magnetic flux directly from N to S via the hub. This guide will therfor make the approaching magnets "blind" to eachother, and therefor not repel eachother. Then they mesh together into a "pseudo solid" magnet, which in turn will let each magnet escape the shield effordless, and gets ready for a new repulsion.

There will allways be "side by side" repulsion between the magnets all the way around, but that force is allways compensated at the "same point" on the opposite side. This repulsion is not what is suppose to run the motor. Only the repulsion is between the magnet tips in the moment the magnets spread out in a "fan". There will initially be great repulsion, but weakens as the magnets gets farther apart. So there is not only ONE magnet which repels at ONE time. It happens gradually. At least what I think, and has seen in the simulations.

I don't say it will work. I have just an idea. When I get some more space to do my hobby, I will make some prototypes - initially to learn more about magnetism, - and learn how naive I am when it comes to magnetism