LIGHTWEIGHT HIGH POWER ELECTROMOTIVE DEVICE
FIELD OF THE INVENTION
This invention relates to high power-to-weight electromotive device capable of use as a motor, alter¬ nator or generator.
BACKGROUND OF THE INVENTION
Electromotive devices are known for use both in transforming electrical energy into mechanical power and transforming mechanical power into electrical energy. In both cases, the energy or power producing capability results due to relative movement between a magnetic field and electrically conductive elements.
Light weight motor, alternator and generator devic¬ es are well known and some are capable of operation at high speeds. However many such devices are not capa- z le of producing high power at high speeds. For exam¬ ple, high power density devices of 0.6 horsepower per pound of weight are known for intermittent operation, but such devices are incapable of continuous operation at high power densities in excess of 1.0 horsepower per pound.
Prior electromotive devices have riot been capable of simultaneous high speed and high torque operations nor have they provided efficiency of operation.
Known electromotive devices which include a stator and rotor arrangement can include magnetic elements on the rotor (for example, ≤ee U.S. Patent No. 3,663,850; 3,858,071; or 4,451,749) or on the stator (U.S. Patent No. 3,102,964; 3,312,846; 3,602,749y 3,729,642 or 4,114,057). Further more, double sets "of polar pieces can be utilized, as in U.S. Patent No. 4,517,484.
In addition, a shell rotor has been suggested in U.S. Patents No. 295,368; 3,845,338 and 4,398,167, with a double shell rotor arrangement suggested in U.S. Patent No. 3,134,037. Bundles of wires have been used in place of a single conductor in the armature assemblies of motors for high voltage and high current usage and/or to reduce current flow loss due to skin effect, and heat¬ ing due to eddy currents, see U.S. Patents No. 497,001; 1,227,185; 3,014,139; 3,128,402; 3,538,364 or 4,321,494, or British Patent No. 9,557. The plural wires are used with solid or laminated cores, see U.S. Patent No. 3,014,139 or 3,128,402; or British Patent No. 9,557. Some prior electromotive devices, such as U.S.
Patent No. 3,275,863, have a power-to-weight ratio of up to one horsepower per pound and U.S. Patent No. 4,128,364 teaches using a gas, liquid, or a mixture thereof to cool a motor to increase its power handling capability.
Many of the preceding difficulties in achieving a high power-to-weight ratio electromotive device have been addressed by a dispersed conductor electromagnet¬ ic device which is the subject of a co-pending U.S. Patent Application by the inventor of the present invention titled "Lightweight High Power Electromagnet¬ ic Transducer". The co-pending design utilizes a straight-sided armature bar of powdered iron which allows full exposure of the copper to the magnetic field. In addition, the powdered iron does not have the flux-carrying ability that the silicon iron does. To minimize the eddy current effect, it utilizes extremely fine wire. The armature bars are fabricated from powdered iron to further insure the 3-d disper- sion necessary to reduce/minimize back electromo¬ tive-force (back EMF) .
Unfortunately, this approach is inefficient in terms of power-in versus power-out due to the resis¬ tance characteristic of fine wire. This characteris¬ tic causes significant energy loss in the form of heat at higher operating levels, which translates into lost power and efficiency. In addition, the straight bars do not lend themselves to standard production automat¬ ic winding techniques as the coils would slip outward from in between the bars. The power loss due to fine wire resistance is compensated for by increasing the amount of permanent magnet material beyond the saturation level of the iron bars. Aside from the costs of additional materi¬ al, the bulk of this additional flux goes into the copper in the form of eddy current loss and is dis¬ persed, leaving very little gain in power for the additional material investment. While the preceding and other various arrangements have been used to at¬ tempt to achieve a high power-to-weight ratio electro- motive device, they have not been* completely success¬ ful. In particular, the prior art does not teach the necessity to disperse the conductors to enable high speed operation. This is due, at least in part, to a widely accepted theory that the magnetic field is relatively small in the non-magnetic winding conduc¬ tors. With conductors built according to conventional teachings, it has been found that torque, at constant current, decreases with increasing speed. This result is contrary to the conventional expectation that torque will remain high as speed increases.
OBJECTIVES OF THE INVENTION
It is a primary objective of this invention to provide an electromotive device which achieves a high power-to-weight ratio by dispersing the electromotive windings to minimize eddy currents within the coils.
It is a further objective of this invention to provide an electromotive device which achieves a high power-to-weight ratio by dispersing the electromagnet¬ ic field core pieces to minimize eddy currents. It is a still further objective of this invention to provide an electromotive device which achieves a high power-to-weight ratio by dispersing the electromo¬ tive windings to minimize eddy currents within the coils the electromagnetic field core pieces to mini- mize eddy currents generally.
It is another objective of this invention to pro¬ vide an electromotive device which achieves a high power-to-weight ratio by shielding the electromotive windings with field core piece extensions to minimize eddy currents within the coils.
It is a primary object of this invention to pro¬ vide an improved electromotive device.
It is another object of this invention to provide an improved electromotive device that is lightweight and provides high power.
It is still another object of this invention to provide an improved electromotive device that operates at high efficiency.
It is still another object of this invention to provide an improved electromotive device having high power density per unit weight.
It is still another object of this invention to provide an improved electromotive device having a high power-to-weight ratio. It is still another object of this invention to provide an improved electromotive device capable of use as a highly efficient motor, alternator or genera¬ tor.
-5- It is still another abject of this invention to provide an improved electromotive device that is capa¬ ble of continuous operation at high power densities in excess of one horsepower per pound. It is still another object of this invention to provide an improved electromotive device having an armature assembly with dispersed conductors, different sections of which have flux carrying elements posi¬ tioned therebetween with the conductors and flux carry- ing elements being formed and positioned in a manner so as to create low opposing induced currents.
It is still another object of this invention to provide an improved electromotive device having an optimum thickness armature assembly which represents a balance among the effects of heat transfer to the cooling medium, heat production from resistance heat¬ ing and other sources, and torque production.
The foregoing and other objectives will become apparent to one skilled in the art as the description proceeds. This invention resides in the novel con¬ struction, combination, and arrangement of parts sub¬ stantially as described and illustrated in the specifi¬ cation and drawings of this patent, and more particu¬ larly defined by the appended claims, it being under- stood that changes in the precise embodiments of the herein disclosed invention are meant to be included as within the scope of the claims. SUMMARY OF THE INVENTION
This invention provides an improved electromotive device with a high power density per unit weight ef¬ fected by utilization of an armature assembly having a large diameter-thin crosέ section speculation ratio. This results in low opposing induced currents, as well as low eddy currents, to enable operation of the elec-
tromotive device at high efficiency with high torque being maintainable during high speed operation.
When the armature moves relative to a magnetic flux producing assembly, eddy currents are established in the electrically conductive portions of the arma¬ ture and these currents lead to heating and skin ef¬ fects (collectively known as eddy current losses) . However, these currents also produce another effect not recognized by the prior art. They are opposing induced currents which alter the magnetic flux pattern and act to reduce the torque with speed increase. This power conversion reduction with speed increase is minimized in this invention by dispersing the conduc¬ tors forming the windings.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings illustrate complete embodiments of the invention according to the best mode so far devised for the practical application of the principles of this invention in which:
Figure 1 is a side sectional view of a rotary implementation of the electromotive device of this invention. Figure 2 is a sectional view taken through lines 2-2 of Figure l.
Figure 3 is a partial isometric view of an arma¬ ture showing the arrangement of the dispersed conduc¬ tors and flux carrying elements of the electromotive device shown in Figures 1 and 2.
Figure 4 is a diagram illustrating a typical ar¬ rangement of a two layer winding formed by the dis¬ persed conductors and illustrating the flux carrying elements positioned between turns of the windings.
Figure 5 is a single lamination stamping of con¬ trolled grain ferrous metal which, when laminated in mass, form "I*' beam stator bars.
Figure 6 is a single lamination stamping of con- trolled grain ferrous metal which, when laminated in mass, form a plurality of modified "I" beam stator bars.
Figure 7 is a partial isometric view of an arma¬ ture showing the arrangement of the dispersed conduc- tors and flux carrying elements of the electromotive device shown in Figure 1.
Figure 8 is a partially cutaway view similar to that of Figure 2 but illustrating an alternate embodi¬ ment of the electromotive device of this invention. Figure 9 is a partially cutaway view similar to that of Figure 2 but illustrating another alternate embodiment of the electromotive device of this inven¬ tion.
Figure 11 is a partial cutaway view similar to that of Figure 2 but illustrating still another alter¬ nate embodiment of the electromotive device of this invention.
Figure 11 is a side sectional view of an alternate embodiment of the electromotive device as shown in Figure 1, and illustrates the inductor fixed to the shaft as may be convenient to a brush commutated sys¬ tem.
Figure 12 is an exploded isometric view of still another alternate embodiment of the electromotive device of this invention, and illustrates a flat lin¬ ear implementation thereof.
Figure 13 is a graph illustrating the relationship between torque and speed for a conventional electromo¬ tive device b and for the electromotive device of this invention a.
Figure 14 is a graph illustrating tested eddy current, hysteresis and windage losses at different speeds of one example of the electromotive device of this invention.
DESCRIPTION OF THE INVENTION
This invention provides a high power density (1 to 5 horsepower per pound) electromotive device incorpo- rating a large diameter-thin cross section speculation ratio. This is advantageous because for a given num¬ ber of magnets or poles...the larger the diameter, the larger (circumferences) each can be. As diameter decreases, the circumferential size of each magnet decreases until it is virtually not seen or interacted with. Conversely, given a fixed size magnetic pole, as diameter increases more magnetic poles can be uti¬ lized resulting in working the copper-iron-magnets more times per revolution (producing more power) . Therefore within limits, a reduction in diameter induc¬ es loss of power and efficiency per unit mass. In addition, through basic physics, torque is directly proportional to the effective radius of the acting force (T = R x F) . Actually as you double the radius, you double the torque arm and double the amount of material producing torque, so power and torque go up as the square of the radius.
In a typical electric motor, torque falls off rapidly with increasing speed. This is primarily due to "opposing induced currents" or "eddy current loss¬ es" in the copper conductors and armature bars. The losses associated with the windings or copper are caused by; cross-leakage between bars (made worse by radially long bars) , direct exposure of the copper to the magnetic field, and over saturation of the ar a-
ture bars due to an excess amount of permanent magnet material. These losses are minimized by this inven¬ tion.
Losses associated with bar-to-bar cross leakage are reduced by designing the electromotive device of this invention so that it incorporates radially short armature bars.
Losses induced by the copper being directly ex¬ posed to the magnetic field is solved by the present invention by an I-shaped armature bar acting as a shield to the magnetic field.
Finally, losses caused by over saturation of the armature bars is solved by reducing the amount of permanent magnet material such that the bars just "approach" saturation. Thi£ is accomplished empirical¬ ly as explained later.
These three factors allow for πftich heavier wire to be utilized without fear Of eddy current losses (cross-sectional area of wire approximately eight times that of prior designs) . The heavier gage wire provides two significant functions; it significantly reduces resistance heating due to its cross-sectional area increase and it allows more conductor (copper) per available space. These two'functions enable in- creased efficiency and increased power output respec¬ tively. In addition, the armature bars themselves are constructed as a lamination of several individual thin stampings, each insulated one from the other. The insulation is a silicon oxide by-product produced during the annealing process. Because sheet metal stampings are utilized, the material grain direction can be and is controlled in the radial direction there¬ by insuring a maximum flux carrying capability (see Fig. 6) .
The controllable dispersion characteristics achieved via thin lamination grain control provides much better performance than the powdered iron 3-d dispersion solid bars in the co-pending application previously referenced. The I-shaped lamination assem¬ bly (armature bar) lends itself to conventional auto¬ matic winding techniques when a special holding fix¬ ture is utilized.
The losses associated with the windings or copper caused by over saturation of inductor bars due to an excess of permanent magnet material is addressed in the present invention by designing the proper amount of permanent magnet material to "approach" the satura¬ tion level of the armature bars. This is accomplished by empirical methodology to optimize the combination of copper, iron, and permanent magnet materials to achieve optimum power density and optimum efficiency through "saturation approaching". Saturation or over-saturation is not necessary and is a serious detriment to good performance. In the empirical meth¬ od, a very sensitive dynamometer is used to measure and plot losses as a function of field. When copper eddy currents were gone, the flux was not reduced anymore. Based on the developed data, the flux con- ducting bars are fabricated from a metal alloy having an iron content which creates a flux carrying ability approximately equal to the flux saturation point as determined by the electrical properties of the design. The preferred embodiments of the invention use a hollow cross-sectional arrangement which lends itself to multiple concentric elements or multiple-motors- within-a-motor. These could be operated concurrently to maximize power density per available space, or individually in a staged manner (like shifting gears in a transmission) .
The cross-sectional arrangement features two radi¬ ating and confecting surfaces for rejecting heat from the armature (conventional designs have one) . Thus the motor can be driven at higher power levels for longer durations without overheating.
The invention can be used as brush-commutated motor or brushless, both in radial and linear configu¬ rations. It can be used as a DC generator or an AC alternator. The ultimate use depending on whether an electrical signal is conveyed to the armature to cre¬ ate a force, causing movement of the magnetic flux producing structure relative to the armature, or wheth¬ er the magnetic flux producing structure is moved relative to the armature. An exemplary embodiment of the electromotive de¬ vice is illustrated in Figure 1. This embodiment includes an outer cylindrical housing 43 which is completed by front 45 and rear end plates 46 secured at opposite ends of the cylindrical housing. A shaft 51 includes a central portion 52 extending through the cylindrical 'housing. The shaft is mounted in the end plates 45 and 46, respectively, by means of bearings 57 and 58 so that the central portion of the shaft is coaxially positioned with respect to the cylindrical housing. The reduced diameter rear por¬ tion 60 of the shaft is mounted in bearing 58 and the front portion 62 of the shaft extends through the front bearing 57 and end plate 45.
The end plates, 45 and/or 46 may include air in- take and exhaust apertures 66 and 67. These apertures allow cooling air to flow through the housing. In addition, an aperture 68 is positioned to allow arma¬ ture conductor connections through end plate 46. In some environments, the device cannot operate in a gas (air) medium so liquid coolant, such as oil, is used.
In such cases, the housing is sealed to retain the liquid.
The rotor 70 has a double shell configuration provided by inner and outer spaced cylinders 72 and 73 which extend normally from the ring connection portion 75. The inner cylinder 72 is secured to the shaft center section 62 by a pair of hubs 54 and 55 to hold the double shell coaxially inside the cylindrical housing 43. Figure 2 is a portion of a cross-sectional view taken along lines 2-2 of Figure 1. It more clearly illustrates that the inner cylinder 72 of rotor 70 includes a magnetic flux return path in the form of a shell, 32, which is preferably a lamination of rings of silicon iron or some other magnetically permeable, low hysteresis loss magnetic material supported by the cylindrical section 72 extending from the hubs 54 and 55. The cylinders 72 and 73 and connecting ring 75 are formed of any suitable material, including iron. The outer cylinder 73 comprises a magnetic flux return path, 33, which may be solid iron or some other permeable, low hysteresis loss magnetic material and a plurality of magnetic elements 30 are mounted on the inner surface of return path 33. In the exemplary embodiment, the magnets 30 are permanent magnets pref¬ erably formed of neodymium boron ferrite (NdFeB) , but they may be formed of barium ferrite ceramic (BaFe Ceramic) , smarmy cobalt (SmCo) , or the like. Perma¬ nent magnets are used in the illustrated exemplary embodiment but they could be replaced with electromag¬ nets.
Returning to Figure 1, the stator inductor 82 is fixed with respect to housing 43. It is mounted on the rear end plate 46 so that the rotor 70 rotates around the common axis of the stator 82 and the hous¬ ing 43. The stator 82 is a stationary cylindrical
- - -13- shell encompassed by the. inner and outer cylinders 72 and 73 of the rotor.
The stator 82 includes electrical conductors 84 of Figures 2 and 3 which are randomly dispersed be- tween stator bars 86. Dispersedv^conductors 84 are preferably a bundle of relatively large (for an elec¬ tromotive device) diameter ..insulated copper wires wound into a linking patfeern with the opposite ends of the wire bundles connected to connectors 89 which extend through aperture 68 in end plate 46 of Figure 1. The use of dispersed, large diameter windings en¬ able the resultant electromotive device to achieve a high power-to-weight ratio because (1) the dispersed windings minimize eddy currents within the coils and (2) the large diameter wire reduces the number of field generating elements? for a given power factor which also reduces eddy currents within the coils.
Conductors 84 are formed into a bundle throughout the armature with each tfljrn of the wire windings hav- ing a flux carrying element or stator bar 86 therebe¬ tween. A typical winding is schematically illustrated in Figure 4. *•
The flux carrying elements, stator bars 86, are preferably a lamination of a plurality of silicon iron sheets. Figure 5 illustrates the configuration of a single layer or sheet of a laminated stator bar. The extensions 34 at the four corners give the bar an "I" beam cross sectional configuration and provide in¬ creased surface area for cooling as well as flux shielding for the windings. These two advantages over the prior art are further features which enable the resultant electromotive device to achieve a high pow¬ er-to-weight ratio. Shielding the electromotive wind¬ ings from the magnetic fields .within the motor minimiz- es eddy currents within the coils. This and the
increased cooling heat exchange surface allows higher current flow which increases field strength without increasing eddy currents in the windings.
The use of stampings such as illustrated in Figure 5 allow the grain direction within the metal forming the bar to be controlled. Thus a bar may be produced with a grain direction as illustrated in Figure 5 wherein the grain direction is parallel to the primary flux path through the stator bar. This reduces heat generation because of the reduced level of resistance to magnetic flux. A random grain pattern provides maximum resistance which leads to maximum heat genera¬ tion and a uniform grain pattern reduces resistance and its resultant heat. A grain pattern following the direction of flux minimizes resistance and heating. Thus a controlled grain inductor bar construction allows higher flux densities without increased heat¬ ing. This increases the efficiency of the device and aids in reaching the stated objectives of the inven- tion.
Figure 6 illustrates an alternate shape for each layer of the laminated induction core or stator bar. In this embodiment, all of the bars share a common central section which simplifies stamping, laminating and assembly.
When used as a motor at constant current, the torque output of this invention can be maintained nearly constant even with increases in rotor speed, as illustrated in Figure 13 by line a. This is unlike prior art devices wherein torque drops off rapidly with increased speed, as indicated in Figure 13 by line b. The combination of high torque and high speed, made possible in the electromotive device of this invention, results in a high power-to-weight ratio.
The stator 82 (formed by the dispersed conductors 84 and flux carrying members 86) is closely spaced with respect to magnets 80 positioned about the inner surface of the cylindrical flux return path 33, and also closely spaced with respect to the laminated cylindrical flux return path 32, see Figures 2 and 7. As previously explained and illustrated, cylindrical sections 72 and 73 provide support for the inner and outer magnetic flux return paths. Typical flux paths have been illustrated in Figure 2. As shown, these flux paths are loops, each of which penetrates the inductor or stator, twice passing through the flux carrying members 86. The flux carrying members are dimensioned to create a thick induction to maintain a high flux density which is essential to high torque. Thus, as illustrated in Figure 7, the dimension of the flux conducting bars 82 along the axis parallel to the primary flux path through the bars is short relative to the longitudinal axis of the bars which parallels the major axis of the electric windings 84 disposed about the bars for generating an electromagnetic field.
As indicated in Figure 8, the electromotive device may be configured with magnets 80 on the outer surface of the inner cylindrical section 72 rather than on the inner surface of the outer cylindrical section 73. In Figure 9, the electromotive device is configured with the magnets 80 on both inner and outer sections 72 and 73. In Figure 10, two cylindrical stators 82 encompass both sides of the magnets 80. In addition, while not specifically shown, it is also to be realized that the electromotive device could be configured by placing additional layers of stator-rotor elements radially inwardly and/or outwardly of that shown in the fig¬ ures.
The electromotive device of this invention thus includes a magnetic flux producing assembly (having at least one pair of poles which can be embodied by using permanent magnets or electromagnets) , and an inductor assembly (which intercepts the magnetic flux produced by the magnetic flux producing assembly and has an alternating structure of conductive windings and flux carrying elements. A winding can be used as the prin¬ cipal component of the inductor with the winding con- sisting of bundles of separate dispersed conductors.
The use of dispersed conductors of large diameter wire permit high speed rotation of the rotor when used in conjunction with winding flux shielding, flux carrying elements. In the case of conductors of large cross section or conductive flux carrying elements of large cross section, as used at least in some prior known devices, as the frequency of the magnetic field reversal in¬ creases, the magnitude of the induced currents in the bars increases, and the induced currents react with the magnetic field to create a resisting torque which opposes the increase of rotational sped. Thus, known shell type devices are inherently limited to low speed by the reaction torque, and cannot be rotated at high speed and are therefore not suitable, for example, for use as traction motors in most practical applications. However, by shielding the windings from the generated magnetic flux and isolating the flux created within the windings, induced currents are limited and the forgoing impediments to high-speed/high-torque operation are eliminated.
When used as a motor, a means to displace (i.e., rotate) the magnetic field relative to the armature at high speed must be provided so that electric power can be converted into mechanical power in a manner similar
to that used by known motors. This can be accom¬ plished by connecting connectors 89 of the armature 82 in Figure 1 to a current source.
When used as an alternator or generator, an actua- tor rotates shaft 51 which rotates rotor 70 to induce a voltage on conductors B.4 and thereby generate an electrical current flow from conductors 84 to a load via connector 89
While not specifically shown, it is to be under- stood that the inductor includes necessary electric commutation devices, including those devices wherein commutation is performed electronically (as in a brush- less DC motor, for example) , as well as those devices which employ rectifiers instead of commutation (as is often used in power generating applications) . A hall device, 21 of Figure 7, may be used in conjunction with a magnetic ring 22 to sense inductor bar or pole piece passing to produce the required timing data.
Figure 11 illustrates an embodiment of the electro- motive device of this invention in which the inductor 82 becomes an armature. It is connected to shaft 52 by mounting disk 101, and inner and outer cylinders 72 and 73 are fixed to the housing 43. In this embodi¬ ment, the inductor becomes the rotor with electric power being communicated to it by means of brushes or slip rings 102 (with brushes being utilized in the case of a DC machine, and slip rings being utilized in the case of an AC machine) . The embodiment shown in Figure 11 is preferred for some applications, particu- larly in the case of a DC commutated machine.
This invention has a significant advantage over a conventional motor by utilization of a minimum amount of iron which undergoes flux reversal. That is, only the iron in the flux carrying elements in the armature is subject to the reversing flux as each pole is
passed, and thus low hysteresis losses are experi¬ enced. In addition, the effects of flux leakage are reduced so that all of the armature windings experi¬ ence the total flux change and thus are equally useful at producing torque.
This invention has significant heat transfer advan¬ tages through the use of "I" beam shaped stator bars, see Figure 5. They make it possible to provide cool¬ ing to both the inner and outer surfaces of the induc- tor. For this reason, the superior high power to weight ratio is further enhanced.
By the principles of heat transfer, heat buildup in an inductor, with constant surfaces temperature and uniform internal heating per unit volume, depends on the square of its thickness. For example, compare an "I" beam armature 0.25 inches thick (as is possible in this invention) to a solid rotor, five inches in diame¬ ter (as is common in known devices) . The heat buildup in such known devices is 400 times as great as experi- enced by this invention.
The electromotive device of this invention can be produced in several topological variations of the basic design. In addition to the rotating cylindrical shell configuration, by changing the orientation of the magnets and the windings, the motor can be made to produce a linear motion. Other variations (not shown) include pancake and conical configurations.
Figure 12 illustrates a linear reciprocating imple¬ mentation of the electromotive device of this inven- tion in which the structure is flat. As shown, mag¬ nets 113 are mounted on flat lower return plate 114. Inductor 115 is provided with dispersed conductors 116 and flux carrying elements 117 in the same manner as described hereinabove with respect to the other embodi- ments illustrated except that the inductor is essen¬ tially flat rather than cylindrical. An upper return
plate 118 is also provided, and inductor 115 is mov¬ able linearly with respect to, and between, lower and upper plates 114 and 118 by means of rollers 120 mount¬ ed on the edges of upper plate 118 and rollers 121 mounted in roller mounting boxes 111 (carried by lower plate 114) .
While preferred embodiments of this invention have been illustrated and described, variations and modifi¬ cations may be apparent to those skilled in the art. Therefore, I do not wish to be limited thereto and ask that the scope and breadth of this invention be deter¬ mined from the claims which follow rather than the above description. What I claim is: