This is the first time I have gotten a simple wheel to accelerate under real world conditions.
My thanks to Mikhail Dimitryev for inspiration, but no motors nor clutches neccessary.
Parts:A 50kg non magnetic material wheel set on a mount with a high end bearing
At 15º intervals bolts with ( 360/15 = 24 ) free hanging iron weights of 250 grams each attached via 25cm of string or other suited material
Weights (represented by spheres) cannot collide
Between the two and three ´o clock position a static magnet is placed in such a way that the weights come into its field at around the two fourtyfive position (2:45) and leave the field at around the three ´o clock position. The balls that come into the magnetic field are slightly pulled towards the magnet and immediately after leave the field as the wheel rotates. This small offset is enough to cause a permanent imbalance in the wheel and causes it to rotate and accelerate in a clockwise direction.
The force the magnet exerts is simulated by the following formula: and(body(b) .p.x > 0.080,body(b).p.y > 0.200)
where b is the objectnumber of the iron ball
p.x. is the position of the ball on the x axis in the model
p.y is the position of the ball on the y axis in the model
the ball comes into the sphere of influence of the magnet at x > 0.080 and y > 0.200 (in this model).
To visualise this I have highlighted four balls in red, blue, orange and green and the respective forces on those balls once they come near the magnet. Note the activation of the force between the two and three ´o-clock position and the deactivation near the three fifteen position, thus simulating the ball entering and exiting the sphere of influence of the magnet. The force the magnet exerts is set to Fx=0.05N (x-axial), in real world you can move the magnet closer or farther to vary this force. I assume the optimal distance is the one where any force is exerted.
The force is uniform and at the same location for all yellow balls in the same manner as shown for the colored balls.
No matter how low the force is set in this simulation, the wheel rotates CW and accelerates until contraints break (infinity).
Gravity exercises upon the minor but persistent imbalance caused by the magnet.
Caveats: Real world one would not expect to have such a force as a constant, but due to the pendular motion of the weights rather a varying force. Still, I would expect to see motion and a self starting accelerating wheel using these principles. Also friction on the bearing may be an issue.
Accurate gravity and air resistance set to high in this model.
I am conservatively optimistic about this model and will be building this. However I gave most my parts away and do not even have a decent bearing left. See? Never give stuff away or throw it out. You never know when inspirational bugs bite you.
AZ-Wheel Model attached. Please point out mistakes I made, I must be missing something.
AZ.