Hi,

I do not understand why you expect acceleration downwards on the right hand side of the wheel due to gravity?

I suppose the center of gravity of your wheel (with its balls all attached as you show) would be in the wheel axle,  i.e. the whole wheel is rotation-symmetrical, right?   If yes, then it is ok in theory that the external SMOT ramp would accelerate the wheel on the left hand side but gravity could influence this rotation only if at least 1 ball on the SMOT side would lose its weight then leaving the SMOT it would suddenly regain its weight on the right hand side.  This is what I think when you mention gravity acceleration with this setup.

The SMOT effect itself could keep the wheel in rotation IF the sticky point on the SMOT upper part could be of less breaking force than the total acceleration force the SMOT would give to the wheel.  (I do not think gravity is involved in this setup.)
I think much tinkering would be needed to find the correct sizes for the balls and magnet strengths and the correct distances between the upper magnets and the balls to minimize the braking force at leaving the sticky point.  It might be possible...

Gyula

EDIT:  Have you seen this video? very interesting and may give food for thoughts :

http://www.youtube.com/watch?v=Pxy8phi6I74 

SMOT Cravity Wheel

A smot ramp is slowly providing a steel ball potential energy. That means there is a little force that pulls the ball up a slight hill over relatively long time. But at the end of the ramp, the vertical force of gravity is greater than the small magnetic force from the ramp, and the ball falls down with great force in short time. Energy is conserved in this whole seequence. The math is simple: The ball starts and ends on the same level even though it seems that the ball has gained energy. You cannot close the loop and expect it to run forever.


Regarding the wheel you've drawn, the same principle can be explained. Each ball on the wheel will always end up on the same spot after one complete revolution. Therefor none of the balls will gain any netto potential energy that can be converted into excess energy.


Vidar

 
I  totally agree with your argument.   As you said  the distances would be the main criteria here.   The distance between the balls should be adjusted such that when the top ball leaves the track,  the bottom ball should be attracted towards the track. ie.,  any decelleration of  the top ball when it leaves the track should balanced by  acceleration of bottom ball which is attracted towards the track.
 

Unfortunately,  the balance (as you describe above) is not good enough in this setup :  when you have a balance achieved between deceleration and acceleration, then all you have done is you have compensated the negative effect of the sticky point.  What you need is a positive difference (that should include friction and air drag at least) between the two counter-forces to keep the rotor disk in a continuous rotation.
Re on the distance:  I agree, the distance between the ball magnets is one important thing, however I refer to the distance in space between the last (stationary) magnets of the SMOT ramp and the just leaving ball magnet too.  This may sound strange but just think the curve steepness of the SMOT ramp: it may have to be ,say,  ellipsiod-like at its upper end instead of a circle.
(This is a visual speculation from me.)

Greetings,  Gyula