Vidar,
Placing the color in the slot only shows what you already have.  You need to view this as field lines not as colors or poles.
If you put the end view of the slotted magnet in FEMM you will see the field in the slot flowing in the magnets direction, where the field outside the magnet flows around to the back of the magnet (opposite direction of whats in the slot)

What this makes is a far reaching field and a strong local field in the opposite direction. The problem is going to be moving from the far reaching attract, into the far reaching repel, because the repel influence will be there before you reach the neturalizing slot.

So it will likely repel the junction befor you get to it. I think you can avoid this by making the slot a bit wider and use a smaller magnet that can further hide in the slot from the far reaching repel area.

Or using a diametrically polarized cylinder magnet, where after the attraction when it reaches the neutral point in the slot, you could rotate the magnet 180 and have it repel out another attraction side. It should rotate with no force at the netural point. This will avoid the far reaching repel problem because it does not exist!
This is what you are showing?

Something like that, yes. For example, when the moving magnet is far out, it will be "seen" as a normal magnet that is attracted by the stationary magnet, and the far reaching field will dominate the force between them. And vica versa. As the moving magnet is forced into the slot, the attracting forces will gradually decrease, and finally be zero when the magnets are meshed together.
The far reaching field will still attract to the larger diameter, but the strong local field will repel the smaller diameter. The last field want the magnet to turn 180 degrees due to the repulsive force, but the far reaching field will not due to the attraction.
However, the diameter of the outside part of the moving magnet is larger, so probably the torque required to turn the magnet 180 degrees will probably be greater than the torque inside the local strong field. At the same time, the local force is stronger, and hopefully will compensate for the smaller diameter and its torque.


Vidar

@Vidar:


Sounds like promising experiments    I'm looking forward to what you discover and share with us.


truesearch

Vidar,
 
I am to busy to work on this at this time but I couldn't resist doing some modeling on this setup. This is what I have found to be the main problem and it should be able to be compensated for.
 
The field in the gap is very strong and works at close distances which results in a non linear effect where the field outside the gap is large and has a linear effect on a rotating diametrically polarized cylinder magnet.
 
During rotation at the netural position, the outer field provides nearly constant torque. The inner field's torque drops off quickly during rotation because the poles move from the close tolerance netural position to the center of the gap at the 90 deg. rotation. This loss lets the outer field again dominate the rotation and resist rotation.
 
I think at this point the solution will end up being the shape of the magnet entering the gap to correct for the loss during rotation. This will probably perform better using a square or an oval magnet, at least for the shape entering the gap.
Once this is corrected for, I see no other problems in it's operation.
 
I was testing the action as a simple piston configuration with the magnet entering as being pulled into position, then rotated, then pushed away.

Thanks for working a bit on this when you have time :-)
I try to experiment with the initial setup - the sliding magnets. Not easy to come to any conclusion yet. I will post a new video of this test this weekend or the first days next week. First I must go buy more of these ferrite magnets (Those you find inside magnetic locks for cabinet doors). They are cheap to experiment with :-)

OK. I'm not sure how you picture the "clamp together". Do you mean clamped together as the bar I have drawn in reply #3?


The fields between the moving and the stationary magnets are vertically aligned (South and north up or down), so it should not weaken the magnets the same way as when the field is opposing eachother directly south towards south for example....


I'll wait to see what Bill meant about this. Thanks for the input.



Vidar

Vidar:

Sorry for the delay in my reply.  Powercat pm'd me a while ago and I just now read and replied to his message.  I did not read all the way through this topic yet so, you may be past this....but just in case....

A few years ago I too two ferrite disk magnets and glued them (superglue) with like poles together.  (I do not remember if it was NN or SS)  Instantly, to my great disappointment, the two together acted like a single thicker magnet with one pole on each end.

My thought process was that it took work to bring them together and, the glue was constantly doing work keeping them together so, something cool might be observed.  I still have them stuck to my fridge and they still only have one pole on each end of the stack.  I hope this helps in some way.  I considered my experiment a failure but, I did learn something.

By the way, to cut the ferrite magnets you do need diamond tooling and, very important, a lot of coolant.  The coolant could be just water but you need to keep it cool to protect the diamond blade, AND, if you locally heat any part of the magnet above the Currie temperature....guess what?  No more magnet.  Just a hunk of ferrite.  It could be remagnetized but, that might cost more than getting new ones.

Best of luck to you on your new design here.  One never knows for sure unless you try.

Bill

Vidar,

The concept of the diametric cylinder rotating between two ring magnets could work. You just need to think of it in reverse, the important part is the part inside/ between the ring magnets. The part outside is used only for balance to allow rotation.

The main problem with the original concept is that the cylinder outside the ring magnets caused more rotational force than the part inside the ring magnets. Change this to always operate inside the ring magnets and rotate when enough has moved outside to balance the rotation.

Sorry I don't have time to do a sketch right now so I will try to get something in a few weeks weeks if you don't understand.