A Laymans View.

[Fester]

After scouring ALL the pages of the mechanical section, in an effort to be brought up to speed on what has been attempted and such. i thought I would post some thoughts theories and a design( in words). Keep in mind I am a very mechanically oriented person, and the complete opposite electrically. Bear with me this will be a long post.

THE STICKY SPOT:
Every single design has the dreaded sticky spot which degrades the cyclical action. So the first thing i decided to do was examine this sticky spot, using website based picture of magnetic fields. A square magnet has 4 equal sides when viewed in 2d. Its field is much the shape of a 4 leaf clover. The "leaves of the clover" come together at the mid point of the face of the sides. The dreaded point is that spot where the 2 poles come together. Where the magnet is both pushing and pulling at the same time. Also to note the strength of the field is also weaker at this point. This is the same as 2 magnets, not touching and being adjacent to one another.

Observation: Every device I have seen tries to eliminate the sticky spot. I have only seen one that tries to use the sticky spot. The video I watched was a guy who had 2 mags on a CD and used his hand to rotate the magnet 1-3 degrees. Back and forth, depending on the poles.  Mechanically this is not alot of force. He was using the sticky spot to his advantage by performing a mechanical movement which is less energy than lets say gearing for example. Versus trying to flip or rotate and magnet on it ownright.

Theory/opinion: I do not believe the standard circular rotation will ever be able achieve getting past the sticky spot. The circle by nature causes the fields to overlap and distort. Making it very difficult to control the field.

WHAT COULD WORK:
Better suited device: I believe the pendulum is a more efficient device. A "Newtons Cradle" is a good demonstration of how efficient the device is.
How can this be done: We all know what and how a pendulum works. For a reference point, assume the pivot of the pendulum is the intersection of a x-y axis. The arm will start at 45 degrees positive x and swing towards negative x. At the end of the "bob" should be a magnet with a single pole side exposed. The arm swings towards the sticky spot of a cylindrical magnet. Now on the way over friction and wind will reduce the angle the arm will achieve when near the apex of negative x. So what we are trying to do is us the field to assist the arm in reclaiming the 1-2 degrees of swing it just lost( 1-2 assumed for a well bearinged pendulum).
To use the inertia of the pendulum and apply a sleight magnetic force, should be enough to get the magnet to reclaim its 1-2 lost degrees. As well as lose inertia and use gravity(weight of bob) to break the magnetic hold. Now remember that guy with the CD i talked about? cylindrical magnet should be able to arc in its plane 1-3 degrees. And arm could be fixed to the bob to bump the magnet to switch poles. Bumping a loss of energy you say? true but we have already reclaimed our loss from the swing so it shouldnt really matter. We should be bumping back at our original 45 degree angle. After the bump occurs same pole force is applied as well as gravity will now kick in. The start of the swing down should make the cylindrical magnet reset its self. and the pull should be negated by the new interia that is created and the amount of push we got from the magnet should equal the pull its trying to do as the pendulum rips away.
If the swing in and bump works as i believe, the pendulum is showing excess energy. The bump is a loss but needed. But in a vacuum with a mag bearing for the pivot this set up would actually be stopping more inertia at the apexes, because less resistive forces are applied.

Im going to start to build a pendulum. Trying to find an engine timing wheel i can borrow and incorperate into my build to show swing arm angles. Also im going to have to order bearings of some sort. either RC bearings or armature bearings from an electrical supply shop. Maybe a junk yard alternator for its bearings.

Comments always welcome. I dont have the software to design etc. If someone wants to do up an animation or whatever, feel free.

[Cloxxki]

That's basically what Howard Yu has done, it seems. Youtube his video's. He posts on forums as well.

I've not seen replication with claims of gain, although he does claim a gain himself.

[Fester]

I viewed Yu's  set up. i believe it holds the same flaws as all the other stator rotor set ups. the array of magnets at the top. His wind vane idea could work. But if you watch a wind vane in action it never stabilizes. It will have a sleight wobble to it. when the vane hits the field it will stabilize but this is lost work.  And as i stated in my original post turn a magnet to 180 degrees or more, is ALOT more work than a 1-3 degree turn. His build is like the others trying to eliminate the sticky spot instead of utilizing it. I guess the term utilizing I should define. The 2 poles meet there. There is much less work involved to do pole switching in this spot. He is going to be turning a magnet 180 degrees, that is alot of wind drag to try and overcome. Also 179-177 more degrees of radius drawing friction. His device may work, but there is too much waste involved in it. And mind you i dont have a build complete yet, so I can only speculate. I do agree Im kind of going down the same path.

[Cloxxki]

I don't think wind is an issues with such a small vane and such a large smot track. The vanes just holds on to a bit if air, it doesn't really offer someone refeshment standing enxt to it, more big air disturbance. In stead of wind a vane with tail weight might work. Stick it all in a vacume. If the twist of a magnet, supposedly outside the filed, would cost more energy than the smot track offers, then think of something better than a smot track.

If you are sure that it's realtively "cheap" to turn a magnet when at dead center being attracted to a opposite member, and get good repulsion as a result, you are SO in business. Surplus energy on the way in, some energy to flip the magnet, and good energy being repelled. I am not buying it until I see it, but if you can do that, all of the world if off building magnet engines.

[Fester]

If you read my write up its cheaper to turn a magnet a total of 6 degrees versus 180 degrees. 1-3 from positive to 1-3 negative. It will be coming in with momentum and a sleight attraction get reclaim the energy lost during swing in attract mode, impact then repel away. The repel will be turned off when the swing starts back and the arm takes tension off the rotating magnet. then attraction sets in and pulls the rotating magnet back to original starting spot. The pull from resetting should be an offset to the push that has already taken place. Leaving a full 45 degree angle once again to free fall. In rational words it makes sense. Math however trumps words in rationality. I'm good at math, but no where near that good.