‘ELON MUSK’ PERPETUAL MOTION MACHINE:

Keep reading, it's not just the first part......

Basically, (FIRST METHOD) a heavy weight on train tracks at an angle has 100% conversion minus friction, which is pretty good if the friction is low, or if the input energy is cheap… But this is not the same as 100% efficiency, it is more like 0% efficiency. which is pretty good if the friction is low, or if the input energy is cheap… Although (SECOND METHOD), using a *weighted lever* to lift a weight to the top of a long track may be more effective.

(Other notes):
Also, this method is overall very cheap considering how easy it is.
(Even the first method) is much better for example than lifting weights on pulleys.

I estimate the second method if it uses a long lever with the counterweight, and a short distance for the lighter end, could create net-over-unity effects when the whole energy system is considered, as only close to equilibrium would be necessary to lift the weight, and part of it could be considered as leverage rather than mass.
Therefore, 100% mass - 10% (lifting in equilibrium) + 48% is about 98% efficiency = 138% over-unity minus friction.
That is at least 38% more efficient than not using the counterweight, and incidentally creates free energy if used effectively.
The advantage here would also be the great speed with which stored energy could be created.
—What is the most cost effictive means to store energy using today’s technology that has potential for mass adoption? (https://www.quora.com/What-is-the-most-cost-effictive-means-to-store-energy-using-today-s-technology-that-has-potential-for-mass-adoption/answer/Nathan-Coppedge)
“Only problem is it’s too slow.” —Elon Musk (?)
Then scale up! —Unknown
—Nathan Coppedge (June 30, 2019), ‘Elon Musk’ Perpetual Motion Machine (https://nathancoppedge.quora.com/Elon-Musk-Perpetual-Motion-Machine)

‘ELON MUSK’ PERPETUAL MOTION MACHINE:

UPDATE:
Basically, if the weight attached to the lever is more than 60% of the weight of the ball, additional weight added counts as over-unity, assuming low friction and 50% return of mass-energy in the ball while it returns.

Modification 1:

I have considered maybe the support on the return reduces the 100% to 50%. So here is a solution to that.

Use very slight change in altitude so that the ball can depress the lever when it returns to the plate using most of it’s mass.

Now take the difference between the total mass and the amount the lever can lift, which is something between 50% and 99.999% of the ball’s mass if we assume the lever is heavily weighted and that the ball can depress the plate automatically.

Now, look at the over-unity rating: it is equal to:

50% -10% + (61 to 99.999%).

Now, if the lever is weighted to the equivalent of above 61% of the ball’s mass, we STILL get over-unity in ideal cases. And then, we have up to another 38 units we could add to reduce friction.

If all energy can be extracted from the falling weight, these new equations give us about less than another 50% on top of that, which would give us between 101% and 188% over-unity. In practice however, the value is more likely to be 101% to 138% identical to the value given earlier with the less conservative equation.