Homopolar Generators - Unanswered Questions and Design Details

[BinaryMan]

I thought about the different configurations of a homopolar generator and some of Tesla's comments on these devices. Asking the right questions may help to clarify whether this class of generator has certain special properties. I will present some of the information I've found online, my questions, and my current ideas or theories. I would appreciate more complete answers if anyone has an idea or experience with it so that I can fully understand how different designs could work.

[1] Logic of the Design.

In order for a device of this class to exhibit overunity properties, there appear to be 3 challenges to overcome.

[1.1] Eddy currents must be minimized since they create drag or oppose the stator magnetic field even when no current is being drawn ("passive drag").

[1.2] The current must be drawn from the rotor in a way that does not create drag or oppose the stator magnetic field ("active drag") and perhaps even increases the field strength of the stator.

[1.3] The mechanical friction must be minimized between the rotor and the load circuit contacts as well as within the rotational assembly.

[2] Stator, Rotor and Circuit configurations.

There are actually many possible configurations that I have not yet seen explored. I will examine each part seperately.

[2.1] The Stator's purpose is to generate a magnetic field for the Rotor; it can be a permanent magnet or an electromagnet.

[2.1.1] What is the limitation in field strength of a permanent magnet? Neodynium seems to be at about 1.2T. This requires no power to maintain.

[2.1.2] What is the limitation in field strength of an electromagnet? With an air core you cannot really drive enough current or coil it enough to get an appreciable field strength. What about electromagnets with a core? Some materials can reach 1.8T with an applied field of only 0.05 gauss. This means that perhaps 0.1 amps of current and a reasonable number of turns of coil can fully saturate it. This requires an insignificant amount of power (I think) to reach field strengths above most permanent magnets, but is sensitive to opposing magnetic forces.

[2.1.3] Electromagnets have the advantage that their fields can be adjusted. It's easier to put them into position initially when they aren't turned on (try placing Neodynium magnets close together during assembly). The current from the generator can be run through the coils to increase the field strength or just to power the device, so I think electromagnets are the best choice for a Stator.

[2.2] The Rotor's purpose is to generate a voltage potential by traveling perpendicular to the Stator's field.

[2.2.1] The most common choice seems to be a conductive disk or cylinder such as one made of copper. This is usually depicted as the classical representation of the device.

[2.2.2] However, in Tesla's "'Notes on a Unipolar Dynamo" he suggests having some turns of wire attached to the Stator leading up to the contacts at the edge. This reveals the potential "secret" (if any) to this machine, which is the possibility that the same current that powers the external circuit can also strengthen the magnetic field that powers the machine; this self-exciting behavior is the focus of the design.

[2.2.3] Expanding on the previous idea, what if the Rotor was in fact a rotating coil rather than a solid disk? It could be a standard coil or a bifilar coil. A parallel bifilar coil generally increases the field generated from the coil, while an antiparallel bifilar coil should generally eliminate its own field. If the rotation of the Rotor and the winding of the coil were combined correctly, any current through the coil should actually mimic the original magnetic fields of the Stator and strengthen it. I would think that this also eliminates most of the eddy current problems.

[2.2.4] There is the experimental case of using a magnet rotating in its own field essentially as both Rotor and Stator. I'm only mentioning it because it's a popular idea but I don't think it addresses all of the design issues.

[2.3] The circuit is usually thought of as containing just the load (say, a light bulb) to power with this generator.

[2.3.1] The important thing here is that the circuit should be thought of as consisting of the Rotor, Rotor mechanism, and Stator as well. With the Stator being an electromagnet, the generator output powers the generator field as well as the load. If the device can actually become overunity, this means that as the output increases, the field strength increases as well as the power running through the Rotor shaft; since the output power is proportional to the rotational speed and the field strength, the only real limit would be the capabilities of the mechanical parts and the ability of all the parts to handle an increasingly high current flow (and heat, etc).

[3] Geometry.

I wrote a program to test the relative field strength and vectors given different coil geometies. I found that two solenoids or flat coils do not create uniform magnetic fields along the surface of a disk between them, but it's fairly close at long as the disk is has a smaller radius than the coils. It appears that two hemisphere-shaped coils can create a uniform field across the disk surface. This is a relatively complex thing to accomplish, and doesn't suit well to using a core.

[3.1] The disk configuration is one type, but the drum type homopolar generator might have advantages. The principal is similar but the Rotor coil approach is difficult to adapt to this type.

[3.2] Are some of the properties of the generator derived from the non-uniformity of the magnetic field on the disk? Tesla's notes on the device kind of states that when only a part of the disk is covered by magnetic fields, it can still be made self-exciting. However, it will function a bit differently when the whole surface is covered, and perhaps especially if the field is more uniform.

[4] Ideas and Questions.

[4.1] Given what I have read, a good way to look at the device is that if it is an overunity generator it would most likely be a poor motor. You want a design where applied current to the Rotor produces no movement. Then you probably have a better chance of having the correct generator action.

[4.2] Why is it suggested in some places that connecting magnets to the disk Rotor creates a situation where there seems to be no counterforce (eddy currents)?

[4.3] Will using a coil as a Rotor or adding some coils to a disk Rotor result in current flow or not since the voltage potential is from the center to the rim of the disk? We are basically forcing a path that re-enforces the original field if possible. Drawing current from the center directly toward the rim I think results in a field that pushes against the original field and slows the disk (counter-productive).

[4.4] In testing any device for overunity operation, I would avoid batteries, permanent magnets, and meters. This is because a battery or magnet can make the device look like extra power is being created when in truth either the magnet is slowly being depolarized or the battery is boiling or otherwise destroying itself. Meters can also be decieving; rather, just try to power the machine from its own output (ie, perpetual motion) then add on an external load. If the machine can support both indefinitely then that's a good sign.

[broli]

This thread might be of interest to you...

http://www.overunity.com/index.php?topic=7391.0

Basically I'm trying to find the same thing. To come up with a design that either only works as a generator and not a motor or one that only is a motor with constant torque without any back torque regardless of rotation speed.

[Yucca]

Hi BinaryMan and All,

Good thread, a thread just for homopolar research and theory, brolis thread is good too.

[4.3] Will using a coil as a Rotor or adding some coils to a disk Rotor result in current flow or not since the voltage potential is from the center to the rim of the disk? We are basically forcing a path that re-enforces the original field if possible. Drawing current from the center directly toward the rim I think results in a field that pushes against the original field and slows the disk (counter-productive).

I read an account of an experiment performed where a homopolar rotor was made out of a pancake coil (not bifilar), a single spiral of thick insulated copper wire, one end on the axis, the other end terminating at a continuous loop around the rim.

The experimenter was hoping to see a higher voltage generated by the homopolar action, which would be good because working with these low voltages it´s very easy to incur electrical losses. Anyway his result was exactly the same open circuit voltage generated as when a solid disc was used.

[4.4] In testing any device for overunity operation, I would avoid batteries, permanent magnets, and meters. This is because a battery or magnet can make the device look like extra power is being created when in truth either the magnet is slowly being depolarized or the battery is boiling or otherwise destroying itself. Meters can also be decieving; rather, just try to power the machine from its own output (ie, perpetual motion) then add on an external load. If the machine can support both indefinitely then that's a good sign.

I agree with it being difficult to meter these things. But permanent mags should be ok for experimenting, it´s pretty difficult to shake a neo up, they can take alot of stick!

One idea I have had for searching for OU in a homopolar is this experiment, which negates the need for electrical output measurement:

Make the simplest homopolar genny, a good sized spinning N52 nickel coated neo disk mag, secured to a high speed motor by a metal shaft.

There should be 1 inch or so of free shaft between motor and neo disk.

Place a metal ballrace on the shaft between the drivemotor and disk.

Construct a static housing out of metal, could possibly use soldered food tins. The housing needs to sit on the bearing and be held still externally, possibly use copper pipe soldered to a hole in the flat face of the tin enclosure. This copper pipe is a push fit on the ballrace outer surface.

So we have a neo disk spinning within a metal enclosure that stays still.

Now we fill the enclosure with a saturated electrolyte solution, whatever is chemically stable and gives highest conductance. The bearing could be the common type with neoprene guard seals, these seals would prevent almost all leakage because the system would not be under pressure and we could arrange the apparatus vertically so the bearing was at the top of the container.

Now when we spin up the disk, a current will flow through the electrolyte between the magnet and the tin suround, which will cause heating of the electrolyte. Of course current will also flow in the mag, the shaft, the bearing and the tin encloseure which will also cause heating.

If we lagged the tin surround with lots of glassfibre mat and foil layers then it would be possible to test for OU using simple calorimetry taking into account the volumes of all the fluids and metals. So after a timed run we could calculate how many Joules the system had gained.

The good thing to this approach is that nearly all energy out could be captured and measured, frictional heating (including electrolyte turbulence), electrical heating, vibrational heating etc.

If the shaft power was known then COP could be calculated fairly accurate. Especially if we took the thermal performance of the system into account by doing some static cooldown runs to measure thermal leakage.

Also one could seperate mechanical effects from electrical by machining a metal disk identical to the magnet and spinning that and performing the same calorimetry.

Well that´s an initial idea for a possible experiment that might reveal some interesting info.

note:
A thermal insulating coupler would be needed between the drive motor and the apparatus.

[BinaryMan]

I read an account of an experiment performed where a homopolar rotor was made out of a pancake coil (not bifilar), a single spiral of thick insulated copper wire, one end on the axis, the other end terminating at a continuous loop around the rim.

The experimenter was hoping to see a higher voltage generated by the homopolar action, which would be good because working with these low voltages it´s very easy to incur electrical losses. Anyway his result was exactly the same open circuit voltage generated as when a solid disc was used.

I think that is good news if the shape of the electrical path on the Rotor doesn't interfere with the voltage potential generated. This means that a beneficial path can be created; in particular, the use of a coil that essentially acts like the original magnetic field when current flows through it.

I agree with it being difficult to meter these things. But permanent mags should be ok for experimenting, it´s pretty difficult to shake a neo up, they can take alot of stick!

It's not that it can't be done with permanent magnets, but rather that a machine based on electromagnets is scalable; the generator can use part of the output to increase the magnetic flux density. It also directly tells us if the Rotor is fighting back since that would affect the electromagnets.

Now we fill the enclosure with a saturated electrolyte solution, whatever is chemically stable and gives highest conductance. The bearing could be the common type with neoprene guard seals, these seals would prevent almost all leakage because the system would not be under pressure and we could arrange the apparatus vertically so the bearing was at the top of the container.

Yes, the refinement of the design might include a conductive liquid to ease frictional forces. I usually imagine that the application of the voltage in such a machine is for something where high current is efficient, such as splitting water.



I tried also to make a diagram which describes my intention; the blue arrow shows the direction of current flow through the Rotor. In a normal disk, the current draw would cause a field which opposes the intended field, while if a certain coil is used then the current draw causes a field which does not. I'm not entirely sure what the physical result would be. What we want is for the current draw on the Rotor to have no effect on the magnetic field; this would mean that the current can be used to power both the field and the rotation.


As far as I can tell, the equation to get the voltage for a disk is:

voltage potential = angular velocity * field strength * disk radius^2

(source: http://www.physics.umd.edu/lecdem/outreach/QOTW/arch11/q218unipolar.pdf)

Question: Why is the thickness of the disk not used in the calculation? Would a thicker disk generate more voltage as long as it's still think enough to experience the magnetic density?

The field strength is limited primarily by the material used. 1.2T for rare earth magnets, and 1.8T for certain electromagnet cores. In a larger scale device, it might be possible to exceed these based on the sheer size and thickness of the coils, but in test devices I don't think this would happen. The angular velocity is limited by the power output of the motor being used to turn the Rotor. The radius is variable and scalable.

The key to an overunity device of this type is that the power output from the voltage generated is greater than the power consumed by the motor. I think the field generation isn't the main concern since magnets can be used, and electromagnets of the correct type wouldn't take very much power to saturate the magnetic core. A device where the loop is established through all the components doesn't need an initial power source; a simple nudge should cause the machine to power itself and speed up until reaching maximum velocity.

I'm still curious about a magnet rotating through its own field; if the current is drawn in the coil fashion, perhaps this might work also.

[Yucca]

Hi BinaryMan,

The to split water then sufficient volatge must be achieved, I don't know the threshold off the top of my head, but it is 1.4v or so i think. This would take a largeish diameter or a very high speed.

It's not that it can't be done with permanent magnets, but rather that a machine based on electromagnets is scalable; the generator can use part of the output to increase the magnetic flux density. It also directly tells us if the Rotor is fighting back since that would affect the electromagnets.

I agree that electromag excitation would be good for scalability, and the whole notion of self excitation is alluring. A permanent mag would always be welcome though, perhaps augmented by an electromag, after all it would provide some flux for absolutely free. If you wan't to detect if the generator is (fighting back) then just monitor the input power to the motor drive. If you mean is the rotor is diminishing the mag field then just glue a hall probe to the permanet mag back, any shifting of average flux will be seen.

With regard to building an OU device straight away, maybe that's rushing it a little. If the calorimetry experiment I mention does yield OU then of course it's a path to consider, if not then it's  building for the sake of it, which I suppose can be enjoyable in its own right.

Homopolars are not that well documented so it's a field ripe for research. Bruce De Palmas "N machine" is said to be OU, just that it had terrific losses to heat, I haven't read anywhere whether calorimetry was performed on the device, it probably should have been.

I know it's boring to do the drudge work with calorimetry, many would prefer to jump straight in and close the loop, but the initial drudgework is required to better understand what we're working with.

The problems I see it with homopolars and building them for excess electrical output is that even the slightest electrical resistance gives large losses. Perhaps with a superconducting disk and brushes then it would be easy, but with even oxygen free copper it has great heat losses.

I might conduct the calorimetry experiment, it won't give us useable electric output power but it will reveal some more fundamental understanding of closed system COP.

Question: Why is the thickness of the disk not used in the calculation? Would a thicker disk generate more voltage as long as it's still think enough to experience the magnetic density?

I think a thicker disk could give higher available current, voltage would stay the same, the equation you quote is only for voltage so the thickness is not included. As you mention to run a thicker disk might require a stronger magnet to make the average field strength be the same throughout it.