Hi,
Thougt I should show you an idea I got. In the pictures below I have simulated a Halbach cylinder whith a magnet array inside it. The magnet array inside is all polarized so they allways attract eachother, but at the top and the bottom the magnets are gradually flipped out of the ring as they rotate around the center. The upper part attracts the magnetic field inside the Halbach cylinder, while the bottom part is flipped opposite and repel the magnetic field. Inside this flipping magnet array there are an iron core to increase magnetic flux through the magnet array around it.
In my model I have approx 16Nm torque. This torque will be pretty much constant regardless of the flipping magnets position, as they flip in and out as they approach or leave the top or bottom. The model size is about 4 inches in diameter and 5 inches deep. N32 magnets are used.
There are weights that is used to force the magnets out of the magnet array an into the attracting or repelling magnetic field. These weights are fixed to each magnet in the array. Hopefylly the weights and the forces inside this motor is working WITH eachother, and nor AGAINST eachother :)
Any questions, please let me know. Any comments and "why this don't work" are ofcourse welcome :)
Br.
Vidar
The one thing I noticed right away was that you have an overall odd configuration of 1.333 which is a minimun must have. You must allways have 2 or more pulling or pushing one thru each gate.
I will study this more when I get some time to do so.
It is a very clever design that deserves being played with.
One question I have is: Are the arms fixed or do the pivit?
Quote from: nightlife on February 13, 2010, 09:12:08 AM
The one thing I noticed right away was that you have an overall odd configuration of 1.333 which is a minimun must have. You must allways have 2 or more pulling or pushing one thru each gate.
I will study this more when I get some time to do so.
It is a very clever design that deserves being played with.
One question I have is: Are the arms fixed or do the pivit?
The weights are fixed to each magnet, but I see now this will only work best in a static situation - no rotation due to the possible centripetal force that will not make a right torque to flip the magnets - maybe in slow motion.
Anyway, there are in this "frozen" picture at least 10 magnets that is forced in the same direction. All 5 magnets in the bottom halv part is pushed towards the right, All 5 magnets in the top half part is pulled towards the left. The two remaining magnets in the rightmost and leftmost side are doing nothing - just waiting to move and to be flipped, so they can do their job.
The question is how the magnetic force between all these magnets in relationship with the magnetic field from the Halbach cylinder will counterforce the rotation. Because I must harness energy from the rotation to flip the magnets if the weights do not work
I see it like this:
The upper magnet is being helped to flip out by the Halbach field, while the lower magnets is prevented to flip by the Halbach field.
On the other hand, the upper magnets are prevented to flip back again as they move counter clockwise from the very top due to the Halbach field.
And the bottom magnets are helped to flip back again as they move counter clockwise from the very bottom due to the Halbach field.
The question is: Will all those forces needed, to flip the magnets in the properiate positions, cancel out, so we are left with only the forces that push and pull the magnet array around inside the Halbach field?
I will look closer into this too, but it takes some time to do all the calculations; in FEMM as the torque is allways calculated around point zero (X=0, Y=0). So I have to move the whole design around to find the flipping torque in each and every magnet.
Vidar
Any flux movement of the Halbach array would delete the arrays designed affect if not evenly done so. This takes away from what is needed.
The more I think about this, the less apt to playing with it I become.
Good luck
I do hope to post a much easier way soon. I am still testing my theory.
I'm currently working on something similar, I think you would do better if didn't approach this so 2 dimensionally, try keeping the poles in the arrays on the same plane (think into/out of the page) in or around (outer's better - homogenity) the circle then you can use your stators to push/pull
Quote from: nightlife on February 13, 2010, 10:00:39 AM
Any flux movement of the Halbach array would delete the arrays designed affect if not evenly done so. This takes away from what is needed.
The more I think about this, the less apt to playing with it I become.
Good luck
I do hope to post a much easier way soon. I am still testing my theory.
The halbach cylinder is fixed. I have made a better/new drawing which shows the path (white circle) the leading outer corner of each magnet in the rotor is following.
Maybe it doesn't help much.
I found that the force needed to flip the magnets as they go is counterforcing. Exactly how much is not known as there are too many factors for my brain to handle ;D
QuoteExactly how much is not known as there are too many factors for my brain to handle
That is why I have chose not to play with it. My head hurts enough already. LOL
I finally figured it out. I found that the "flip-distance" versus torque, ended up in 100% of all the torque available is used to flip all those magnets. So therefor there will not be any torque left to run the rotor. What a discovery... f...ck!!
LOL ;D
Case closed.
Don't give up yet. Just remember that an array must be created to have an uneven field and the runners have to be unevenly numbered but an even combination must be added to the mix to make it work. ;)
I just gave you more then I should have and if you don't understand what I have said, it will be more clear after I am done building my proto type.
Quote from: A on February 13, 2010, 11:12:42 PM
Don't give up yet. Just remember that an array must be created to have an uneven field and the runners have to be unevenly numbered but an even combination must be added to the mix to make it work. ;)
I just gave you more then I should have and if you don't understand what I have said, it will be more clear after I am done building my proto type.
I see the picture - I think. Anyway, to make anything work, the magnetic density must be constant uneven around the runner to force it to run. So far I haven't figured out how to do this without electricity. I hope you don't go silent if you figure out your prototype does not work...
Good luck.
I'll go and play with my magnets again...
Vidar