Magnetic OU principle, You should really take a look at this !

[AhuraMazda]

@charlie
You want to look at this video:
http://www.youtube.com/watch?v=oWePP4MLUD0&feature=channel

[Butch LaFonte]

Step #5 is the same as step #1 butch  .

Just like butch said when you close in they will pull on you so that's energy gain. You'll just have to give that energy back when you spread the magnets back open. On the other hand you can use a clever trick to not feel any push/pull by doing the below thing. Basically you have a set of magnets that always repel each other attached to this main apparatus. I was thinking of ways how to store the energy effectively when the poles come to each other. Springs wouldn't work nicely as their forces are linear with distance. I need something that had a 1/distance relationship. Then it struck that one could just use magnets in repulsion. That way the attraction from the main apparatus will get canceled by the repulsion of magnets attached to it. Same poled electrical charge could also be used if you worry about demagnetization.

Broli,
See this link for Balance Fixture video from 10 years ago. It's an oldie but goodie.
https://www.yousendit.com/download/U0d5b2VDVnNFd2NLSkE9PQ
Butch

[broli]

Yeah that's pretty much the whole idea. Good to see you have all the knowledge packed, I was still in elementary school back then  . In your video it seems like attraction wins for very close distances that's why it sticks but repulsions wins for large distances. You'll indeed need a very subtle shaped magnet to overcome these two behaviors.

[broli]

Butch I have a rotor design in mind that uses the balance experiment you did on youtube. Although it does not use the fanner repulsion but can be incorporated. But I just wanted to keep it as simple and effective as possible. Sure you can convince a lot of people that step by step there's a net energy gain, but everyone wants to see a self resetting runner in the end  .

The design I have is 3d in nature so I'm going to give sketchup a go and try to make a rendition of it.

Edit: Design is attached below. I have to admit that it has many unfinished parts as I got too lazy  . Even the spacing of the guiding hoops is not correct. But this is the main idea. First of all from that view it's supposed to rotate CW. Now as the spaced out washers come from below the magnet they get attracted by a big force as proven on your balance experiment. The distance between the hoops should remain constant until the moment the washers arrive at the middle of the magnet. At this point the distance between the hoops will start to shrink very rapidly thus bringing the washers together. But the magnet will still want to keep them in the field. But as you've experimentally proven the force is weaker than they were spaced out and thus the washers manage to escape. While coming back the hoop distance expands again and pulls them apart and the process restarts.

[Charlie_V]

@broli

Thats a good idea.  I like your idea of balancing the pull force with opposing magnets, I think that could work!