The following text describes three main varients of motor means used to power the Gyroscopic Inertial Thruster.

Any prime mover capable of producing the energy needed to maintain the orbital velocity of the drive spheres can be used, giving consideration to the operating environment and appropriate use of available technology.

The first power transfer means is the simplest, and most appropriate for experimental assemblies, as it is much easier to produce by anyone not having a full knowledge and production means neccessary to build an electromagnetic version.

A friction drive wheel is positioned near the center of the race to contact all drive spheres around the inside of the race so as to both hold them firmly in contact with the race, and provide a consistant spacing around the race. Although the balls will crowd at the spin end of an accelerating race, and spread apart at the nose, they will do so proportional to their momentary position in the race, giving a simple means to maintain balanced separation. This is a handy means to maintain positional relavance when the thing's shut off and in the hall closet {The motor means won't have to lug the great gob of static balls off of the bottom of the race and sling them into obedience!}.

The axis of rotation of the drive wheel will be slightly off center, given the fact that an open portion of the race will allow the balls to move to the outside of the orbital path, and the closed end will push the balls toward the center, to the degree of difference of the ball diameter to the inside diameter of the orbital path.

Of course a frictive ring of material (such as rubber, etc.) to transfer the motion to the spheres is needed, but should be as thin as needed to make up for slight out of round (oval) that may be inherent in a simple race, and not so large so as to magnify elastomeric loss of energy.

Experimentation is needed to confirm (or refute) the gravity tapping generator, and I suspect that by orienting the nose and tail parallel with the ground, at right angles to the center of gravity, efficiency will be maximized.

The crowding of the balls at the tail side of the race have a much longer dwell time in the acceleration field, and their fall is slowed by the conversion to spin, gathering potential.

If acceleration (deceleration) at right angles to the produced thrust is the most efficient input orientation, right angled turns will be the most efficient for recapturing energy of motion. Hmm, sounds familiar again.

There are two main types of electromagnetic drive means.

I call these the Electro-mechanical, and the electro-dynamic.

The electro-mechanical means is based on standard electric motor theory. Individual coils are distributed around the race, and are energised by a timing and control means to pulse these coils in advance of the arrival of each ball around the race in calculated positions. Control means must take into account the spacing distortion at various acceleration settings, and provide for individual advancement and retarding of balls that get out of sync.
^
The electro-dynamic version of the motor means is potentially the best and most interesting.

As you know, (if you don't then you're probably scratching your head over most of this collection of documents) a wire carrying a current becomes polarized with a magnetic field circling the electron path. By applying an opposing potential on either half of an electrically separated split race, the balls will act as conductors, becoming circularly polarised just like a wire does.

The race halves can be either permanent magnets, electromagnets, or a combination of both. The polarisation will be one pole circularly pointing to the center of the race and the other radially pointing away from the center, and both halves having the same polarisation, that is they will want to push each other away. This places a magnetic field radially crossing the space the balls travel in.

Standard motor theory explaines the tangential force that will be applied to the balls in the race, driving them around the race proportional to the current that is moving through the balls.

A consequence of magnetizing the balls is that an opposing field is developed between the balls, which may aid in keeping the balls from banging into one another, a potentially disasterous condition.

A control means for juggling these whirling dervishes in the race can be accomplished by segmenting the electrical contact surface of the race to allow for individual adjustment of orbital velocity and position.

A hybrid between the two electromagnetic motor types is the use of a simple race (not segmented) for the electrodynamic version, while adding individual coils about the race to accomplish individual adjustment, and perhaps 'overdrive' if needed.

I believe the hybrid version is probably the best application of motor means in that the race is much cheaper to build and easier to maintain (resurface), as well as providing a more solid rolling friction.
^
Here's an experiment you might try. Start with a magnetically polarized race of material of sufficient mechanical and magnetic 'hardness' (permanent magnets), place within the race non-magnetized balls that are built of magnetically hard material (will hold an imposed magnetic charge), and then apply a current through the balls to get them moving in the race. The balls will gradually aquire a permanent polarization that is in the same direction as the electrically imposed field.

Will the gyroscopic forces of the spinning balls be of sufficient strength to prevent the now permanent magnet balls from flipping over? If it is, will the balls continue rolling in the race without power input? If I knew for sure, it wouldn't be any fun to explore now would it?

IMPOSSIBLE is just a word, it shouldn't be a paradigm!

This concludes this section of the document collection.

I hereby declare that the contained information is true and correct to the best of my knowledge and ability to extrapolate.

4 May 1997 David Eugene Cowlishaw
Edited 30 May 1997