Lutec, De Palma, Adams and some others claim to have made overunity motor generator machines.
A refrigeration machine is an overunity machine. A heat pump uses a COP or co efficient of performance as a unit to describe its output. A COP of 3.5 means that with an input of 1 KW of energy, an output of 3.5 KW of heat will be delivered. This is a fairly normal COP. Instead of creating heat by resistance, existing heat is pumped from one area to another, utilizing the latent heat of controlled phase changes.
Both engineering and science tell us an electric motor connected to a generator cannot have an output which exceeds the input. Some very efficient motors are about 94% or a COP of 0.94. But electrons, like heat, may also be able to have a COP.
I think that if the bearings of a motor are eliminated then perhaps the efficiency could be
improved somewhat. With no bearings, an output shaft is probably impracticable, so this machine may need to be a combined Motor Generator.
My proposal is a vertical shaft with a taper roller bearing at each end, the taper rollers pointing down into the shaft on both bearings, with the shaft located only by gravity. A horizontal Rotor is fixed to the centre of this shaft.
At the outer circumference of the Rotor is an array of suspension magnets with poles facing vertically or axially. These magnets would be alternating. Another similar array is set out with poles facing radially outward, also alternating. These magnets would be surrounded by copper bars fixed to the frame. Below the air gap distance would be very close at rest, with the air gap above about 20mm, and to the side probably about 10mm.
When the Rotor is spun, the close proximity to the lower copper bar will produce a repelling magnetic effect, and this should lift the Rotor into a position central between the top and bottom copper bars. This will also lift the vertical shaft clear of the angled bearing face. Stability should be assured by the outer radial copper bars. The bars would need to be positioned to maintain lift and decrease drag. The builders of the Inductrac Train have shown that as speed increases, lift increases and drag decreases. This is opposite to what would be found with aircraft. Near these magnet arrays would be a steel balance ring containing some mercury.
Inboard would be another 2 rings of magnets. The inner array set out in a manner suitable for a 6 coil 8 magnet BLDC motor drive. The coils to drive the motor would be both above and below the Rotor making a 12 coil 16 magnet motor. The lower coils would be very close at rest, with the driving shared top and bottom when the Rotor rises.
Further out would be another ring of alternating magnets. These would be for generating only. Coils would be mounted below and above as for the motor.
An interesting aspect of this design is that it may be possible to incorporate and test a means of eliminating cogging in both the motor and generator.
This should provide a very efficient motor, and with fine tuning may be near to self running.
There is another possible source of energy generation available from this design. The repelling magnetic fields, primarily being used for electrodynamic suspension, should also be moving on the outer side of the copper bars. These bars are as far as possible away from the magnets, to create the least possible drag. If coils are placed directly behind the copper bars, any moving magnetic field will cut through the windings, and may create an electric current with no, or minimal, additional drag.
Some tests will need to be made to see if the moving field does in fact occur at the other side of the copper bar, and its strength. Information on the exact movement and shape of the electric eddy currents and associated magnetic fields is not easy to find. I suspect the only way to check is to run the motor with both volt meter and ammeter, and attach coils in various alignments to instruments and a load, totally isolated from the drive source.
My guess is that the current follows the direction of the movement of the nearby magnets, near to the surface between bar and magnets. It would then move to the back side of the bar and move in the opposite direction until rejoining the current at the front of the bar.