Repulsion Motor UAER — Drum vs Disc Part I

Introduction

The somewhat cryptic title of this post refers to the "Unenergized Attraction - Energized Repulsion" concept that I've discussed before, as used in both the Kure Tekko motor with its drum-shaped rotor (and its repelling stator magnets), and my version of it with its disc-shaped rotor.

In this and the next post, I'll compare these two versions of the concept, including some magnetostatic modelling of them.


Drum design — good and bad features

A bad feature of the drum design is that a lot of the electromagnet's magnetic flux goes more or less directly between adjacent inner edges of its pole pieces, and so that flux is largely wasted — it has little influence on the rotor magnets. With the disc design's toroidal electromagnet, essentially all of its flux must go across the full airgap between the faces of its pole pieces, and thus it interacts better with the rotor magnet.

A good feature of the drum design is that not only does its electromagnet repel the rotor permanent magnets, but modelling results show that it also attracts a portion of the backing steel in the region immediately behind these magnets. With no backing steel, this feature is not possible in the disc design.

Drum design modelling, without repelling stator magnets

Fig 1a   Drum design model, without stator magnets

Figure 1a is from a model of a drum design, flattened out into a linear configuration. This was done to remove any possible ambiguity in the results, since only forces in the vertical Z-direction are now relevant to the analysis.

In this first model, stator magnets have been omitted.

Data (all dimensions in millimeters):—

Rotor Magnets (blue): Each 96 × 60 × 8, NdFeB35.

Magnet Backing Steel (grey): 130.1 × 25 × 960, M19 steel.
(The backing steel is essentially "infinitely" long in this modelling).

Electromagnet Core (grey): Cross-section 57.15 × 38.1 = 2177.4 mm² 
Core is based on E300 transformer laminations (halved by shearing).
Pole pieces are 96 × 60 × 5, with 0.5 airgaps to magnets.

Electromagnet Coils: Current injection plane area (solid red) = 2 × 19.225 × 105.3 = 4048.8 mm². Total excitation 16400 amp-turns.

The magnet pair is first attracted in to the unenergized core, until there is zero vertical displacement between them, as shown. The electromagnet is then energized as noted above, while the magnet pair is repelled through a vertical distance of 96mm out from the core. The magnet pair then continues to move away from the core until the force between them has diminished to zero.

Fig 1b   Modelling results, drum design without stator magnets

Results

Figure 1b shows graphed results from this modelling.

The dip in the attraction force is probably genuine — I have seen it in several other models. The negative spike in the repulsion force beyond 96mm displacement could of course have been eliminated by continuing the excitation out to say 150mm of magnet displacement.

As modelled above, the total output energy is 46.603 joules.


Drum design modelling, with repelling stator magnets

Fig 2a   Drum design model, with stator magnets

Figure 2a is from a model of the same drum design as before, with a pair of repelling arc-shaped stator magnets and their backing steel added. These magnets are 1500 surface radius × 60 wide × 8 thick, starting tangential to the electromagnet pole pieces, extending 500 in the Z-direction. Their backing steel is also 130.1 × 25 cross-section, M19 steel.

Fig 2b   Modelling results, drum design with stator magnets

Results

Figure 2b shows graphed results from this modelling.

In this case the moving magnets are first attracted-in to the unenergized electromagnet core as before, and are then repelled-out under the stator magnets with the same 16400 amp-turns excitation over the same 96mm magnet displacement as before. They then experience further repulsion from the stator magnets until that repulsion has diminished to zero.

The total energy gained is 47.743 joules. This is only 2.45% higher than when the stator magnets were absent. Note that the increase is all in the attraction-in energy, while the net repulsion-out energy has actually decreased slightly.

Conclusion — stator magnets not worthwhile

The overall conclusion from these results seems clear enough: only a very small increase in total energy is gained at best with the stator magnets added, making the considerable extra cost and complication of adding them not worthwhile.

As is only to be expected, the stator magnets act to "smooth out" the mechanical energy delivered over the repulsion part of an operating cycle, but they do not add any net energy.

Other more cost-effective methods could be used to smooth out the mechanical energy delivered. The simplest method would be just to add a flywheel.

Of course, the more modelling that is done, the more secure will be the conclusion drawn from it. Ideally more modelling should be done for longer excitation durations, and for stator magnets generally closer overall to the moving magnets. However, this modelling would take more time than I can currently spare. (So it becomes one more item for my ever-increasing "to do" list!)

I'll look at the modelling of a comparable disc version of the UAER concept next time.

The Kure Tekko Motor

Primary References:—

Popular Science Magazine June 1979, p80

Japanese Patents JPS55144783, JPS55136867, JPS55115641, JPS55114172, JPS55111654, JPS55106084, JPS55071185, JPS55061274, JPS55061273, JPS55053170, JPS55053160.

Fig 1   Popular Science magazine June 1979, p80

    

Fig 2   Popular Science magazine June 1979, p81

Early Disclosure

The first of several very interesting Japanese electromagnet/permanent magnet motors to 
be disclosed in the West was the prototype shown in Figures 1 and 2 above. Evidently the result of considerable design and development work, and with a radically new operating principle exploiting forces of magnetic repulsion, it was made by automotive components manufacturer Kure Tekko, a.k.a. Kure Tekkosho K.K. Several clues indicate that it was designed to be a free energy motor, although it was not specifically claimed to be one in the Popular Science article.

The Kure Tekko motor is the best example known to me of a genuine magnetic repulsion motor that was professionally designed and manufactured. As far as I know, there has not been any really careful independent examination and comment on what was reported about it. Let's now do that:—

Same principle

An essential part of this motor's operating cycle is the attraction-in of permanent magnets to an electromagnet's unenergized core, followed immediately by repulsion-out of the magnets from the electromagnet, now energized. This, in the drum-rotor Kure Tekko motor, is exactly the same principle as proposed in my disc-rotor "UAER" device (see my posts of 13 February and 5 March 2016).

For the record, my UAER concept dates back to my 1950s childhood. In about 1957 I built a crude prototype of this idea (which would not have delivered any free energy — I was not specifically trying to do that back then). However, I cannot prove that I built that prototype now. The Popular Science Kure Tekko article may well be the first time this principle has appeared publicly.

Patents and inventors

Most of the patents cited above can be regarded as precursor ideas. Only two, JPS55061274 and JPS55114172, resemble the final design described in the Popular Science article, and they have two sets of electromagnets with a 180º operating cycle, rather than the single set and 360º cycle of the final design.

However, when read carefully together, the full set of patents leave no doubt (in my mind at least) that the intention was to invent a free energy magnet motor.

It is often claimed that prominent magnet motor inventor Yasunori Takahashi invented the Kure Tekko motor. I have never found any hard evidence for that claim. His CV does not list any association with Kure Tekko, and he does not appear among the inventors named in the patents cited above.

Item (F)

The detailed colour illustration Figure 2 above would have been drawn from photographs of the actual machine, and also possibly from engineering drawings of it. In this illustration, the most obvious and important clue that this was indeed a free energy motor is item (F), which is described as a "small electric motor" that "crank-starts the rotors." It has a cooling fan. Such fans are used only on devices that run long enough to attain thermal equilibrium, which no starter motor is ever designed to do. Nor would it ever remain mechanically connected and rotating at high speed after starting, as (F) must, via its belt drive. Note that there is no over-running clutch or any other method of disconnecting the drive. Also, since the main motor was claimed to "start" at only about 200 rpm, the speed ratio between (F) and the main motor (as inferred from the pulley sizes) seems very wrong, if (F) is just a starter motor.

So, (F) is something other than just a starter motor. By visual appearance alone, anyone with any experience with cars would identify it as a typical 1970s continuously-running automotive alternator, rather than a starter motor. Could it provide all the electrical energy needed to keep the main motor running in stand-alone mode? A magnetic repulsion motor like this could well require only a small amount of such "self-generated" electrical energy to "top-up" minor losses, and hence to keep running continuously. Of course (F) could be made to act briefly, initially, as a starter motor. (As a general rule, almost any conventional electrical generator can be fairly easily made to reverse its direction of energy conversion between electrical and mechanical, and hence operate as an electric motor, if required).

Power, weight and efficiency

Although the Popular Science article noted that "no performance details have been revealed," power (45 hp) and weight (155 lb) were given. The Kure Tekko motor therefore had a better power to weight ratio than other generally available conventional electric motors of the time. That alone should have caused further interest and investigation, but what in fact ensued was — silence.

The main thrust of the article is that the Kure Tekko motor was designed and built to be more efficient than conventional motors, and hence it seems surprising at first sight that the most important performance detail of all, i.e. an efficiency figure, was not given. However, if an efficiency of over 100% had indeed been achieved, the omission of such a controversial result would become more understandable.

Capscrews into electromagnet cores

In the illustration it appears that the electromagnet cores are fastened to their steel backing plate with capscrews. At first sight, this method of fastening would seem impractical with laminated cores; but lamination would normally be required, to minimise eddy current losses in the cores. 

Perhaps an experiment with solid cores is being shown here? Still, laminated cores could be successfully drilled and tapped for capscrews, if the laminations were first glued together, e.g. with a strong epoxy glue allowed to set under compression. Sometimes such "unusual" techniques are used in prototypes — we see that the motor design allows for easy substitution of electromagnets. This would be as expected for a prototype stated to be in an "early development stage," to permit experimentation with different electromagnet cores, windings and pole pieces.

Metal alloy castings

The article states that "most of the engine castings, including the rotor, are made of light alloy." This would almost certainly be an aluminium alloy, a good electrical conductor, which would therefore have eddy currents flowing in it during operation. Did the designers do a full assessment of expected eddy current losses? I have reason to believe they might have done better to have made these components from non-conducting material such as ceramic, or perhaps a suitable plastic.

Tight clearances

To obtain maximum benefit from the (non-linear) forces of magnetic repulsion which are much higher at small separation distances, the Kure Tekko motor has an extremely tight clearance, for a motor of its size, of only 0.1mm between the rotor magnets and the start of the spiral stator magnets.

Presumably such a tight clearance would not have been specified without good reason — the designers must have considered it essential. It would probably require, among other things, finish grinding and magnetizing of the rotor magnets after they had been assembled into the rotor — routine enough procedures for industry, but not for amateurs. The same clearance of 0.1mm would then also be expected between the rotor magnets and the electromagnet pole pieces. So, careful manufacture of the electromagnets would also have been required.

On the other hand, although the tight clearance gives a very high repulsive force, that force drops back very soon to lower values as the airgap increases, i.e. not much energy can be gained from exploiting even a very high force over a tiny distance.

Permanent magnet mismatch

There is an obvious mismatch between the "cobalt" rotor magnets B (assumed to be samarium-cobalt — alnico would be far too weak and easily demagnetized in a repulsion motor like this) and the plastic bonded ferrite stator magnets E. Was cost the over-riding factor here? Another consideration is that stator magnets made from electrically conducting material such as samarium-cobalt or neodymium-iron-boron would have high eddy-currents flowing in them during operation, unlike ferrite, a non-conductor.

I'll have more to say about this topic in future. Unlike conventional motors, where eddy currents almost always cause problems and should therefore be minimised, there are cases where they may actually be beneficial in genuine repulsion motors.

Ferrite stator magnets would give lower overall performance than samarium-cobalt or neodymium-iron-boron ones, but would also require less electrical energy to be delivered to the electromagnet (because it would require less repulsive force to expel the rotor magnets against the weaker ferrite stator magnets).  Although most of that electrical energy is returned to the source, minor losses still occur, such as Joule heating, which is proportional to the square of the electromagnet current.

Another very important question is — are the stator magnets really worth having at all? More on that next time.

Timing of electromagnet impulse

The article states that the electromagnet impulse "is precisely timed to start after the cobalt [rotor] magnet reaches top dead center..." So there is no doubt that a mechanical energy "bonus" is being obtained, by allowing the rotor magnet to be attracted-in to the unenergized electromagnet core, before the impulse is applied.

Consequently, the electromagnet now has "wrong-way" flux from the permanent magnet in its core, but the question can again be raised of exactly what energy penalty has to be paid to expel that flux, (over and above that required to energize the electromagnet anyway).

Request for update?

Bloomberg Business Profiles has Kure Tekkosho K.K. currently at:—

1-8-15, Hirotagaya, Kure, 737-0134, Japan 
Phone: 81-823713121 
Fax: 81-823740698

I suppose an attempt could be made to contact them for an update on what happened with their motor since the Popular Science article, but personally, as an ordinary member of the public, I gave up on such attempts long ago, having had zero response to every one I had ever made.

When is a "Repulsion Motor" NOT a Repulsion Motor?

According to its manufacturer at http://www.motcointernational.com/repulsion.htm
this is a present-day repulsion motor, rated 0.75 kW, 2880 rpm, 230V 50Hz single phase

The short answer to the question posed in the title above is "whenever it is the so-called 'repulsion motor' described by all orthodox sources".

I'll now go into some detail to establish that answer.

Definition

There is an electric motor design well-known to orthodox technology as a "repulsion motor". Searching on-line, or in electrical technology textbooks, we most often find definitions such as the following, from Alternating Current Machines, M. G. Say, 4th Edition, Pitman Publishing, p504:—

Repulsion motor

This is a form of series motor with the rotor energized inductively. The rotor winding is designed for a low working voltage, and its brushes are joined by a short-circuiting connector to provide a closed current path. The brush axis is displaced from quadrature with the stator d-axis so that the induced rotor current can develop interaction torque. When the motor spins the d- and q-axis fluxes have a phase displacement, developing a crude travelling-wave resultant of "elliptical" form. At a rotor speed corresponding to the synchronous, a nearly uniform travelling flux wave is attained, so that the commutation conditions are relatively good.

A better definition

This sort of definition is OK as far as it goes, but it gives no indication of why this machine is called a repulsion motor. Only much more rarely, in textbooks at least, do we find a comprehensive, honest definition and description such as the following, from Theory of Alternating Current Machinery, Alexander S. Langsdorf, Tata McGraw-Hill Publishing Co Ltd, 9th reprint 1985 (my emphasis at the beginning and end of the first paragraph):—

15-10. The Repulsion Motor. 

     The qualifying adjective employed to designate this type of motor is a hopeless misfit, but the term has been in use for such a long time that it is generally accepted. It appears to have been used for the first time in the specification of U.S. patent 363,185, issued to Elihu Thomson in 1887, to describe the machine shown in Fig. 15-18a, and in modified form in Fig. 15-18b. 

Fig. 15-18. Elementary repulsion motor of Elihu Thomson

The distinctive feature of this machine which differentiates it sharply from the types now in use is the open-circuited armature winding; the individual armature coils of Fig. 15-18b successively experience a torque when they are short-circuited by the pair of brushes displaced from the axis of the stator winding, the direction of the torque being in the direction of the brush displacement. It is perfectly clear that such an arrangement as Fig. 15-18a or b will not develop torque if the brush axis is in the plane of the stator coil, for in that case the active rotor coil constitutes merely the short-circuited secondary of a transformer linking with the stator flux; and if the brushes are displaced by 90º, there is likewise no torque since there is then no rotor current. It is only in intermediate positions of the brushes that torque is developed. But it is to be noted that if the brushes are in the plane of the stator coil, and if the plane of the active rotor coil is parallel to, but slightly displaced from, the stator coil, there will be a force of repulsion between the coils acting in the direction of the displacement; this is in accordance with the experimental fact (discovered by Ampère in 1820) that parallel conductors carrying currents in opposite directions repel each other, this phenomenon being utilized in the constant-current transformer described in Art. 2-27. It is probable that this particular behavior of parallel coils gave rise to the term "repulsion" in connection with the motor of Fig. 15-18, but the genuine repulsion between parallel coils has nothing whatever to do with the operation of the so-called repulsion motor.

Fig. 15-19. Connections of singly fed repulsion motor

     Whatever its origin may have been, the phrase "repulsion motor" is used to designate the arrangement shown diagrammatically in Fig. 15-19a. It consists of a closed-type commutated drum-armature winding entirely similar to that of a d-c machine, the brushes being short-circuited along an axis displaced by an angle α from the axis of the single-phase stator winding. The latter is disposed in slots around the inner periphery of a laminated ring similar to the stator frame of an induction motor.

     It is readily apparent that the single-phase alternating mmf supplied by the stator winding S of Fig. 15-19a can be resolved into two components, cophasal with respect to time, but displaced in space by 90º, so that one component is in line with the brush axis and the other is perpendicular thereto; or, what amounts to the same thing, the actual single stator winding may be regarded as equivalent to the two windings F and T of Fig. 15-19b, as has already been indicated in Chapter 5. Winding T then plays the part of the primary of a (series) transformer with the short-circuited armature winding acting as the secondary; and the reaction between the inductively supplied rotor current and the magnetic field due to F develops the torque. Since the current taken by the motor clearly varies with the load, and the flux due to winding F is proportional to the current (ignoring possible saturation), the motor is characterized by variable field flux and will therefore have the series characteristic of variable speed. One further fact evident from Fig.15-19b is that the magnetic field due to the armature mmf is largely neutralized by the mmf of winding T, which therefore serves as a compensating winding functioning by induction, instead of by conduction as in the series traction motor discussed in previous articles.

Further details

I have quoted only a small fraction of Langsdorf's discussion of the "repulsion motor." He has eight further articles on it, headed:—

15-11. Starting Conditions of the Repulsion Motor.
15-12. Phasor Diagram of Repulsion Motor, Running Conditions.
15-13. Analytical Relations in the Repulsion Motor.
15-14. The Circle Diagram and Performance Characteristics of the Repulsion Motor.
15-15. Commutation in Repulsion Motor.
15-16. The Compensated Repulsion Motor.
15-17. Phasor Diagram of the Compensated Repulsion Motor.
15-18. Doubly Fed Series and Repulsion Motors.

Langsdorf's articles on the so-called repulsion motor are by far the most thorough and useful I have found. His biographical notes are:—

Alexander S. Langsdorf, M.M.E., D.Sc.
Dean Emeritus, Schools of Engineering and Architecture, and Professor Emeritus of Electrical Engineering, Washington University
Fellow, American Institute of Electrical Engineers
Member, American Society of Mechanical Engineers

Orthodox repulsion motor: a "red herring"

I will have a lot more to say about genuine repulsion motors in future posts. Meanwhile, it is worth bearing in mind that the orthodox wrongly named repulsion motor is nothing more than a misleading "red herring," in comparison with those devices that do exploit genuine magnetic repulsion.