What you appear to be investigating is being able to lift a mass to a higher GPE than GPE lost cost of that lift. This I believe is one of the reoccurring 'disconnects' I talk about regularly. In order to get greater height gain than height lost (for the same mass etc) you would have to design a device that breaks Archimedes Law of Levers [ALOL - f1 x d2 = f2 x d1]. And Law of Levers is predicated on a constant 'g' acceleration (because it's weight force and f = m x a), which is conservative precisely because it is constant (at earths surface).
The real reason the supposed M.O.W. wheel doesn't work is the same reason "height for width" (even though I don't like the term) mechanisms don't work:Law of Lever is static.
What result would yield using static in dynamic situation?
Gravity Torque is measured by the horizontal distance from the axle. Gravity energy is measured by the vertical distance. When a weight drops more than its counterpart, PE is lost and KE enters the system. When a weight rotates around a wheel, it falls just as far as it rises. If weight movement is perfectly efficient, we will have a net loss equal to frictions.
Leafy, That is not a small conversation. Ultimately, your dynamic lever will convert a fixed PE to KE and back again. Leverage applies statically and dynamically. As Fletcher mentioned, Extra energy has to be consumed somewhere to get a gain in KE, GPE, Momentum. So unless an energy source is external or a finite fuel, energy has to be created, requiring a break in known physics. So Bessler apparently found a chink in the physics armour, or he was a fraud.
Where do you theorize that is coming from in your design Leafy?
Pequade here always confused energy with leverage(as he still does on OU.com), and it is easy to see why. It is a pity I have not had time and resources to test more of his theories.