LIMITED-ANGLE ACTUATOR FOR ELECTROMECHANICAL ENGINE VALVE ACTUATION

FIELD OF THE INVENTION [0001] This invention relates to actuators and more particularly to electromagnetic rotary actuators.

BACKGROUND OF THE INVENTION

[0002] United States patent no. 6,755,166 issued to Chang et al. (hereinafter "the '166 patent") describes an electromechanical engine valve drive which uses a motor, instead of a conventional camshaft, to open and close an engine valve independently of the engine. By eliminating the mechanical link between the valve and the crankshaft, the invention is able to achieve variable valve timing, which offers benefits such as higher fuel efficiency and lower emissions.

[003] The system described in the '166 patent works with a shear force actuator. There are several requirements imposed on this actuator. An actuator for an electromagnetic engine valve needs to have an inertia which is low enough to allow the valve to achieve transition fast enough for high engine speed. For instance, the transition needs to be finished as fast as 3.5 millisecond (ms) or less at 6000 rpm engine speed. The actuator needs to have physical dimensions small enough to allow as many as are required to fit on the head of an automotive engine.

[004] For example, in a four-cylinder engine with two intake valves and two exhaust valves for each cylinder, 16 actuators will be needed for independent actuation of each valve. The actuator also needs to deliver enough torque to do work against friction force and gas pressure while keeping system loss, especially ohmic loss, low enough so that it can be cooled and so that the overall power requirements of the electromagnetic valve drive system are acceptably small.

[005] This set of requirements is challenging and commercially available electric motors have not proven to be satisfactory in all respects, which is a significant disadvantage to the system described in the '166 patent.

SUMMARY OF THE INVENTION ^'

[0006] Accordingly, in view of the above problems with prior art actuators, one object of this invention is to provide a limited-angle actuator which is capable of fast valve actuation with a high torque/inertia ratio, while keeping power consumption (mainly armature ohmic loss) acceptably low and actuator size small enough to fit into the limited space over an engine head.

[007] Thus, in accordance with the present invention, an actuator includes (a) a iron having a gap provided therein, (b) a plurality of permanent magnets and (c) an armature formed by an array of current conductors.

[008] With this particular arrangement, an electromagnetic rotary actuator having a limited angular range is provided. In one embodiment, the gap in the iron is provided along an axis of mirror symmetry for a field structure (sometimes shown and described hereinbelow as a so-called "vertical" gap). The iron also includes arcuate gaps in which the armature moves. In one embodiment, the actuator is adapted for use as a valve actuator for an electro-mechanical engine. By providing an actuator having a limited angular range, an actuator having all of the necessary attributes including but not limited to being able to deliver enough torque to do work against friction force and gas pressure while keeping system loss, especially ohmic loss, low enough so that it can be cooled and so that the overall power requirements of the electromagnetic valve drive system are acceptably small is provided. These attributes are a result of the innovative structural design of the actuator as well as the material choice. The actuator of the present invention thus enhances the feasibility of the whole electromechanical valve actuation concept.

[009] By appropriate selection of iron shape, magnet shape and position and armature shape and position, the actuator is provided having a torque/inertia ratio and a torque/resistance ratio which is higher than prior art actuators. Also, by taking advantage of the innovative structure design of flux circuit and armature and smart material choices, the actuator of the present invention is provided having favorable system package compared with that of prior art approaches.

[0010] In the present invention, the only moving element besides the actuator shaft is a one-phase hollow-shaped winding, which is centered on, and rotates relative to, the axis of the actuator. All of the iron and permanent magnets, which form the magnetic circuit, are stationary. This type of winding-only rotor eliminates inertia from the permanent magnets and the iron. Instead of rotating repeatedly in one direction, the armature swings back and forth symmetrically within the limited angular range. This feature allows unique iron and winding structure designs to be used which results in reduced actuator size, winding inertia, and winding resistance. Instead of round- shaped copper conductor used in most armature design, other shapes may also be used. For example, conductors having rectangular, trapezoidal, oval or round cross-sectional shapes conductors may be used to form the armature. In one embodiment, aluminum conductor is preferably used in the present invention to keep both inertia and resistance low. Permanent magnets with high energy-product, such as Neodymium-lron-Boron, may be used to excite high flux fields and therefore help to generate a high torque output.

[0011] In summary, there are several innovative features in the present invention. First, the magnet-excited ferromagnetic structure and the armature are designed to reduce actuator size and rotor inertia while keeping torque output high enough to drive valves open and closed. Second, in a preferred embodiment, the armature is constructed from individual conductors selected to have a shape (e.g. a rectangular or trapezoidal shape) which proves very effective to reduce winding resistance and therefore ohmic loss while keeping

rotor inertia low enough to achieve fast valve transitions. Third, to take advantage of the rapid development of high quality permanent magnets, NdFeB having high remanent flux density, high coercive force, high energy- product, high operation temperature and low cost may be used to ensure that an actuator having a high torque output and a compact package is provided. Thus, the actuator of the present invention allows great compromises to be made between torque output, rotor inertia, ohmic loss, actuator size, and cost, which make it a very practical candidate for engine valve actuation.

[0012] From the discussion above, it is apparent that with innovative designs of the magnet-excited ferromagnetic structure and the armature structure, with the choices of the high quality permanent magnet NdFeB and the shaped aluminum conductor, the actuator described herein includes all the necessary attributes for engine valve actuation. Those attributes include high torque/inertia ratio to accomplish fast valve transitions, high torque/resistance ratio to minimize ohmic loss of armature, and appropriate actuator size and cost to make the whole system acceptable for practical engine package.

BRIEF DESCRIPTION OF THE DRAWINGS [0013] The foregoing features of this invention, as well as the invention itself, may be more fully understood from the following description of the drawings in which:

[0014] FIG. 1 is a cross-sectional view of an actuator;

[0015] FIGs. 2 and 2A are isometric views of an armature;

[0016] FIG. 2B is a side view of the armature of FIGs.2 and 2A; [0017] FIGs. 3-3D are a series of exploded cross-sectional views illustrating a series of different techniques for utilizing an armature binding member;

[0018] FIG. 4 is an isometric view of an alternate embodiment of an armature;

[0019] FIG. 4A is an isometric view of an alternate embodiment of an armature;

[0020] FIG. 4B is an enlarged, isometric cross-sectional view of a portion of a winding on the armature of FIG. 4 and with a portion of an armature binding member removed to reveal a plurality of conductive members;

[0021] FIG. 5 is a cross-sectional view of a portion of an actuator having two permanent magnets placed ninety (90) degrees apart with respect to an actuator axis;

[0022] FIG. 6 is a cross-sectional view of a portion of an actuator having two permanent magnets placed parallel to each other with respect to an actuator axis;

[0023] FIG. 7 is an isometric view of an actuator having four permanent magnets disposed to generate an axial flux field;

[0024] FIG. 7A is a cross-sectional view of the actuator shown in FIG. 6;

[0025] FIG. 8 is an isometric view of an actuator having six permanent magnets which provide a combination of an axial flux field and a radial flux field;

[0026] FIG. 8A is an isometric view of the actuator of FIG. 8 without an iron;

[0027] FIG. 8B is an isometric view of the actuator of FIG. 7 without the iron and without the armature;

[0028] FIG. 8C is a horizontal cross sectional view of the actuator of FIG. 8 taken along lines 8C - 8C in FIG. 8;

[0029] FIG. 8D is a vertical cross sectional view of the actuator of FIG. 8 taken along lines 8D - 8D in FIG. 8;

[0030] FIG. 9 is an actuator embodiment which includes permanent magnets having a polygonal shape; and

[0031] FIG. 10 is an actuator embodiment having an electrically excited flux field (the excitation windings are shown in top and bottom positions).

[0032] FIG. 11 is another actuator embodiment having an electrically excited flux field (the excitation windings are shown in top and bottom positions).

[0033] FIG. 12 is an isometric view of a mold;

[0034] FIG. 12A is an end view of a mold;

[0035] FIG. 13 is an isometric view of fixture having windings thereon;

[0036] FIG. 13A is isometric views of a mold using the process of fabricating an armature;

[0037] FIGS. 14-16 are a series of views illustrating removable of an armature from a mold;

[0038] FIGS. 17, 19, and 20 are isometric views of an ion yoke;

[0039] FIG. 18 is an end view of an ion yoke;

[0040] FIGS. 20-27 are a series of isometric views illustrating the assembly of an actuator; and

[0041] FIG. 28 is an isometric view of an actuator.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0042] Referring now to FIGs. 1 and 1A in which like elements are provided having like reference designations, an actuator 10 includes a magnetic circuit (or "iron") 12 having a shape selected to form therein a central opening 13 having a vertically oriented gap 14 emanating therefrom. The iron 12 shape is also selected to form a pair of arcuate gaps 15 into which are disposed respective ones of a pair of permanent magnets 16a, 16b having north and south poles N, S as shown.

[0043] The actuator also includes an armature 18 (Fig. 2) having a shaft 20 and a winding 22. The winding 22 is comprised of straight sections 22a, 22b (also sometimes referred to herein as "active" sections) and end turn sections 22c, 22d. The armature 18 is arranged in the iron 12 such that the armature shaft 20 is disposed in the central opening 13 and straight portions 22a, 22b of the armature winding 22, are disposed in the arcuate openings 15 of the iron 12. It should be appreciated that for clarity in the drawing figure, the windings 22a, 22b shown in Fig. 1 are depicted as a unitary piece while in Fig. 2 the winding 22 is illustrated as being comprised from a plurality of turns of a conductive member (e.g. conductive member 22). Thus, the windings 22 are also sometimes referred to herein as "conductive members" or "armature conductors." The outside diameter of the armature shaft 20 is selected such that the shaft 20 can freely rotate relative to a surface of the iron 12 which defines the central opening 13. Similarly, the armature windings 22a, 22b are disposed in the arcuate openings 15 such that they can freely move within the arcuate openings 15 of the iron 12. Thus, a first surface of each armature winding portion 22a, 22b is spaced from a surface of the respective magnets 22a, 22b by a distance (or gap) designated by reference numeral 19a in FIG. 1. Similarly, a second surface of each armature winding portion 22a, 22b is spaced from and from surfaces of the iron 12 which define the arcuate openings 15 by a distance (or gap) designated by reference numeral 19b in FIG. Ideally, the gaps 19a, 19b are substantially the same size, however, this need not be so. In one embodiment, the thickness of each armature winding portion 22a, 22b is about 1.5 mm and the air gaps 19b are about 0.2 mm (i.e.

the distance between respective surfaces of the armature windings 22a, 22b and the iron).

[0044] As alluded to above, in one embodiment, the armature 18 (Fig. 2) is formed from one or more conductive members 22 which are electrical current conductors. Thus, the winding 22 may be fabricated from a single conductive member which is "wrapped" or "wound" into a desired shape with a desired number of turns as will be described hereinbelow. Or alternatively, the winding may be provided from a plurality of individual conductive members or alternatively still the winding may be provided in a desired shape from a unitary piece of conductive material.

[0045] It should be appreciated that the gap 14 shown in the exemplary embodiment of FIG. 1 is referred to herein as vertical only because an axis of mirror symmetry designated A-A in FIG. 1 for the field structure is pictured as vertical in FIG. 1. Thus, if the image of FIG. 1 were simply rotated ninety degrees, the gap 14 would be horizontal in FIG. 1. It should also be appreciated that the armature 18 moves within the arcuate gaps 15 provided within the iron 12.

[0046] The armature is disposed on a base 30 which is shown in phantom in FIG. 1 since it is not properly a part of the actuator 10.

[0047] Conventional motor designs typically include two separate parts of back iron with the shape of hollow cylinders. One is at the inner side of permanent magnets and the other is at the outer side of armature winding.

[0048] In contrast, the actuator of FIG. 1 is provided having two parts of iron which can be combined into one whole part due to the feature that the armature windings 18a, 18b swings back and forth in a limited angular range within the arcuate gaps 15. For the same reason, and as shown in FIG. 1 , both the outer and inner part of the iron 12 are provided having a partially round shape (e.g. arc-shaped at top and bottom 12a, 12b as shown in FIG. 1)

while the left and right side of the outer part of the iron 12c, 12d can be cut-off as much as a reasonable flux path permits (i.e. it is desirable to maintain enough thickness so that none of the iron is saturated). The shape is selected to allow the desired movement of the armature 18 while at the same time reducing actuator dimension in the direction along the axis labeled B-B in Fig. 1. In other words, given the same motor size limit, an actuator provided in accordance with the principles and concepts described herein results in an increase the motor/armature dimension in the direction A-A and provides an accordingly higher torque output than a conventional actuator given the same current input.

[0049] It should also be appreciated that an actuator provided in accordance with the principles and concepts described herein need not be round on all sides at the left and right sides. This makes it easy - compared with prior art approaches - to respect a desired valve center-to-center dimension. The gap 14 in the inner part of the iron serves to increase the reluctance of the circuit to flux excited by armature current which helps decrease winding inductance and armature reaction.

[0050] In the embodiment of FIG. 1 , the pair of permanent magnets 16a, 16b are disposed to generate a radial flux, which is symmetric from left to right as indicated by the dashed lines 20 in FIG. 1. The permanent magnets 16a, 16b can be sintered to arc shape, magnetized along the axis of mirror symmetry of the actuator, and placed as shown in FIG. 1 , where opposite poles are facing each other and attached to an inner part of the iron 12. It should, of course, be appreciated that other magnet shapes having different magnetization directions can also be used.

[0051] The availability and affordability of the high quality permanent magnet Neodymium-lron-Boron (NdFeB), which has a high energy product, high remanent flux density and high coercive force, make it a preferred choice in order to excite high flux field and therefore to generate high torque.

[0052] Referring now to FIG. 1A, since the actuator 10 is symmetric from left to right with respect to centerline A-A and from top to bottom with respect to centerline B-B, it is possible to define all angle-related parameters of each part in a quarter portion of the whole cross section (it should be noted that to improve clarity in drawing FIG. 1A, cross-hatching has not been used to indicate that the elements of the armature are shown in cross-section). Therefore, suppose the conductor windings 22a, 22b of the armature 18 range from -a to +α (FIG. 1) and the whole winding 22 swings from -θ to +6> (FIG. 1A), then the permanent magnet 16a, 16b should cover at least from - (a + θ) to +(a + θ) while the arc-shaped part of the iron 12 should offer a reasonably larger angular range than that of the permanent magnets 16a, 16b to ensure proper main flux path and satisfactory flux density in the air gap. The reasonably larger angular range should be selected so that much of the magnet flux, even that linking the arcuate extreme portions of the magnets 16a, 16b, crosses the main air gap and thus links the armature winding 22 when it is within angular range of these portions of the magnet, rather than closing on paths which travel circumferentially to portions of the iron 12 which are not radially outside the armature. These constraints in angular range for this design can be summarized in Equation (1) as below,

a + θ ≤ β < γ

(1)

in which: a is the half angular range of armature; β is the half angular range of permanent magnet; γ is the half angular range of iron; and θ is the half angular range of armature rotation.

[0053] In the present structure, the armature 18 is the only element in the actuator 10 that is capable of movement due to the existence of two air gaps 19a, 19b presenting at the inside and the outside of the armature 18

respectively. The air gaps 19a, 19b allow the armature to move without contacting the stationary surfaces of the iron 12 and the permanent magnets 16a, 16b. Torque is generated when current flows in the conductors 22. To make torque, current must flow in opposite directions in the conductors in the top and bottom air gaps.

[0054] Referring now to FIGs. 2 - 2B in which like elements are provided having like reference designations throughout the several view, a rotor 17 includes armature 18, the shaft 20 and the end clamps 24. The armature 18 used in the actuator 10 of FIG. 1 includes the winding 22 having straight sections 22a, 22b and end turn sections 22c, 22d. The winding is provided from a single conductive member which is wrapped or wound into the desired shape around the shaft 20. The embodiment of FIGs. 2-2B corresponds to a four-turn winding with connecting parts between the winding 22 and the shaft 20. Those of ordinary skill in the art will appreciate how to select the particular number of turns to use in any particular application.

[0055] As will be described in more detail below in conjunction with FIGs. 3- 3D, to make an insulated and rigid armature an armature binding member is used. There are several possible techniques and materials for providing an armature binding member ad several exemplary materials and techniques are described below in conjunction with FIGs. 3-3D.

[0056] The armature 18 includes a pair of end clamps 24 disposed at opposite ends of the shaft 20. Each end clamp 24 includes an outer end clamp 24a and an inner end clamp 24b secured together using any known fastening technique. In this particular example, the inner and outer end clamps 24a, 24b are secured using screws (the screws have been omitted for clarity but threaded screw holes 25 are visible in FIGs. 2 and 2A. The end clamps 24 fasten and secure the winding end turns 22c, 22d at each end. Collars 26a, 26b are disposed on opposing ends of the shaft 20. The combination of the armature, shaft and end clamps forms a rotor 17.

[0057] Referring now to FIGs. 3 - 3B, a plurality of conductive members 30a - 3Od (also referred to as armature conductors) have an armature binding member 32 disposed thereover. In the embodiments of FIGs. 3 - 3B the armature binding member 32 is provided as an epoxy pre-preg tape (fiber- glass woven insulation tape with B-stage epoxy) disposed over the conductive members. This may be done as part of the process of making the winding. The winding and armature binding member combination are heated (e.g. in an oven to cure the epoxy) resulting in a rigid and strong armature.

[0058] The embodiments of FIGs. 3C-3D, on the other hand, utilize a regular fiber-glass woven tape 34 wrapped around each conductor 30a-30d prior to winding the conductor. Once the conductor is wound into a desired shape with a desired number of turns, an extra epoxy resin is added to the whole winding (i.e. after it is wound) to secure the structure. Again, the winding is heated (e.g. in an oven to cure the epoxy) resulting in a rigid and strong armature.

[0059] Referring now to FIG. 4, an armature 42 includes a shaft 40 and a winding 42 having active or straight portions 42a, 42b and end turn portions 42c, 42d.. An armature binding member 44 is disposed over the winding 42 to bond together the conductive members. The armature binding member may be of one of the types described above or may be provided from any polymer capable of forming the structural element of the armature. Each axial path down the air gap (i.e. the active portion of the winding 42) is connected to the next axial path by a conductor lying in a plane normal to the axis of the armature (i.e. the end turn). The structure of the end turns 42c, 42d is apparent from the figure. In particular, the end turn 42d is wound to provide an opening 41 through which the shaft 40 passes. At the other end, he actuator shaft 40 is attached to the conductor/polymer structure of the armature winding at the end turn 42c.

[0060] It should be appreciated that the armature 18 described in conjunction with FIGs. 2-2B includes a shaft which passes through both end-turn sections

of the armature winding. In the actuator embodiment of FIG. 4, however, the shaft passes through only the armature windings at only one end while the other end of the shaft does not pass through the armature winding but instead attaches to an inside surface of the armature.

[0061] Referring now to FIG. 4A, an armature 38' similar to that the armature 38 in FIG. 4 is shown, but a shaft 40' passes through both end-turn sections 42c', 42d' of the armature winding 42'.

[0062] Referring now to FIG. 4B, an enlarged isometric view of a portion of the armature 42 in FIG. 4 is shown as if cut from the portion of the armature 42 which passes axially through the arc-shaped gap in the iron/permanent magnet structure. A portion of the armature binding member 44 has been removed to reveal a plurality of conductive members 46a-46r, generally denoted 46. In this particular embodiment 18 the winding is comprised of eighteen conductive members. Those of ordinary skill in the art will appreciate how to select the particular number of turns to use in any particular application. Thus, depending upon the a particular application, the winding may be provided from fewer or more than eighteen conductive members.

[0063] In one embodiment, the material from which the armature binding member is formed fills the gaps between the conductors 46 in addition to providing a surface coating. In embodiments in which the armature binding member is provided as a polymer, the polymer may be reinforced. Any of several common reinforcing additives may be used, including particulate matter, chopped fiber and/or continuous fiber.

[0064] As mentioned above, in some embodiments, the electrically conductive members are provided as an array of turns of wire. In this particular embodiment, the turns of wire are provided as aluminum turns of wire. Other conductive materials, may of course, also be used. In this particular embodiment, the turns of aluminum wire are each provided having a rectangular cross-sectional shape. The rectangular cross section of the wire

allows a high packing factor, while the polymer allows a low-mass-density structure. Conductors having other cross-sectional shapes may also be used. Fort example the conductors may be provided having cross-sectional shapes including but not limited to trapezoidal, arc, oval or even round cross-sectional shapes. Some factors to consider in selecting the shape of the conductors from which the armature is provided include but are not limited to mechanical strength (i.e. shear force), packing factor, cost of manufacturing the conductors and assembling the armature. Regardless of the shapes, any gaps between the conductors are preferably filled (e.g. with the a polymer of the type described above including the use of additives which may include reinforcing additives, particulate matter, chopped fiber and/or continuous fiber).

[0065] It should be appreciated that the preferred conductor in the vast majority of conventional electromechanical devices is copper.

[0066] In one embodiment, however, instead of copper, aluminum is chosen for the armature. This is due, at least in part, to the need to put an upper limit on the inertia of the moving element so that fast enough transitions at high engine speed can be achieved. On the other hand, it is also desirable to have the electrical resistance of the winding as low as possible to assure ohmic loss in an acceptable range. Therefore, one design goal is low electrical resistance, combined with a limit on the inertia of the moving element.

[0067] The formula for electrical resistance is R = p _e * I /A. Here, R is resistance, p _e is electrical resistivity and I and A represent the length and cross-section area of the conductor path, respectively. In the actuator, the length of conductor is principally determined by the geometric parameters of the magnetic circuit, which are independent of conductor material. The electrical resistivity is directly determined by conductor material.

[0068] In one embodiment, the allowable cross-sectional area A, for the conductor, is an indirect function of the choice of material. For the armature

construction proposed here, the inertia of the moving element is very substantially determined by the mass of the conductor. The mass of conductor is proportional to mass density and to cross sectional area. Thus, if one selects the mass of conductor to be the same value, independent of conductor material choice (to achieve equivalent inertia), the allowable cross- section area is inversely proportional to the conductor mass density.

[0069] In one embodiment, the conductivity of aluminum conductor chosen (ASM grade 1350) is 0.61 times that of copper, so the electrical resistivity is 1.64 times as great. On the other hand, the mass density ratio is 3.2, with copper being heavier. So the net electrical resistance R _a ι/R _C u = 1.64/3.2 = 0.51. In other words, when the volume of conductor is constrained to limit inertia of the moving winding, the electrical resistance achievable with aluminum conductor is just about half that achievable with copper, which results in about half winding loss accordingly if all other parameters are the same.

[0070] It should be appreciated that in the embodiment shown in FIG. 1 , the two permanent magnets 16a, 16b are disposed 180 degrees apart with respect to the actuator axis (i.e., the angle σ between a centerline A-O of the permanent magnet 16a and a centerline O-A of the permanent magnet 16a is 180 degrees in this design). It should be understood, however, that other embodiments originate from this same concept, with different arrangements of permanent magnets, iron and coil. Several such exemplary embodiments are shown and described in conjunction with FIGs. 5-10.

[0071] Referring now to FIG. 5, a portion of an actuator 50 includes an iron 52 which defines an opening 54 into which are disposed two permanent magnets 56a, 56b. The two permanent magnets are placed 90 degrees apart with respect to the actuator axis, i.e., σ = 90 in this case. An armature is disposed in the iron 52 such that armature windings 58a, 58b are disposed in the opening 54 in accordance with the principles discussed above in

conjunction with FIGs. 1 and 1A (e.g. the armature is selected having shapes and dimensions such that the armature is free to move within the iron).

[0072] Referring now to FIG. 6, a portion of an actuator 60 includes an iron 62 which defines openings 64a, 64b into which are disposed two permanent magnets 66a, 66b. The two permanent magnets 66a, 66b are parallel to each other, in other words, σ = 0 in this case. An armature is disposed in the iron 62 such that armature windings 68a, 68b are disposed in the opening 64 in accordance with the principles discussed above in conjunction with FIGs. 1 and 1A (e.g. the armature is selected having shapes and dimensions such that the armature is free to move within the iron).

[0073] Referring now to FIGs. 7 and 7A in which like elements are provided having like reference designations, an actuator 70 comprises an iron 72 and four permanent magnets 74a - 74d disposed to generate an axial flux field (rather than a radial flux field as shown in the embodiment of FIG. 1). An armature 76 is disposed such that the tops of the armature 76a, 76b move relative the iron 72.

[0074] Referring now to FIGs. 8 - 8D in which like elements are provided having like reference designations throughout the several views, an actuator 80 is comprises an iron 82 having an armature 84 and permanent magnets 86 disposed therein. The armature 84 includes windings 88 and a shaft 90. The actuator comprises six permanent magnets 86a-86f disposed to generate a combination of axial flux field and radial flux field. It should be appreciated that magnets 86a and 86e generate radial flux fields while magnets 86b-86e generate axial flux fields. It should also be appreciated that the embodiment illustrated in FIGs. 8-8B is a combination of the embodiments described above in conjunction with in FIGs. 1 and 7.

[0075] Referring now to FIG. 9, another embodiment of an actuator 92 includes a iron 94 having a central opening 96 and a vertically oriented gap 98. The iron also defines a pair of polygonal-shaped opening openings into

which permanent magnets 102 and an armature windings 104 are disposed. The iron 94 and armature 104 may be formed using the techniques described herein.

[0076] In this actuator, the permanent magnets 102 are sintered or otherwise provided such that only one surface has an arc-shape. The opening 100 in the iron 94 is appropriately selected to have a size and shape which accommodates the magnets 102 as well the armature 104. The thickness of the magnets 102 generates a flux density which is larger compared with the flux density generated by a magnet of the same material but having an arc- shape. Thus, this embodiment provides relatively high torque output, however the torque coefficient may be variable due to the change in thickness of the magnet with respect to angle.

[0077] Referring now to FIG. 10, an actuator 110 having an electrically excited flux field is shown. The excitation windings are shown in top and bottom positions. The actuator is provided from an iron 112 which defines openings 114 in which armature windings 116 are disposed. The iron 112 and armature 114 may be formed using the techniques described herein.

[0078] Referring now to FIG. 11 , an alternate embodiment of an actuator 120 having an electrically excited flux field includes an iron 122 which defines openings 124 in which armature windings 126 are disposed. The excitation windings are shown in top and bottom positions. The iron 122 and armature 124 may be formed using the techniques described herein.

[0079] Exemplary processes for fabricating and assembling a limited-angle actuator are described below in conjunction with FIGs. 12-28. The steps in the processes may be performed in an order other than that described below. Also fabrication of certain piece parts is not described as such fabrication is within the level of skill of one of ordinary skill in the art.

[0080] The below description includes a process for fabricating a hollow- shaped winding and a technique for connecting an armature with a shaft to form a rotating armature (also sometimes referred to as a rotor) as well as a description of one technique which may be used to assemble a plurality of parts to form an actuator.

[0081] It should be appreciated that in the below description it is assumed that the thickness and the angle range of an armature to be fabricated are already decided. A decision must be made as to how many turns to use in the winding. The decision and factors to consider in selecting the particular number of winding turns for a specific application are well known to those of ordinary skill in the art. For example, in one application for a 42 volt (V) voltage bus, it may be desirable to use a 20-turn winding to provided a desired output. However, in a 12 V system (e.g. the voltage level commonly used in electrical systems in automotive engines), a 4-turn winding for the actuator may be desirable. Thus, the details of a particular armature and/or actuator will change depending upon the particular application.

[0082] In one exemplary actuator embodiment intended for operation with a 12V system to be described below in conjunction with FIGs. 13-28, a conductive member used to provide a winding is provided from an aluminum conductor having a rectangular cross-sectional shape with 1.3 mm of height and 2.3 mm of width. As mentioned above and as will be described further below, the conductive members have an armature binding member disposed thereover and in one embodiment, the armature binding member is provided from a dielectric material having a thickness of about 0.1 mm. The above dimensions were selected to satisfy a desired 1.5 mm thickness of the armature winding.

[0083] In one embodiment, an air gap of about 0.2 mm exists between each side of an armature winding and an iron surface. Thus, it is important for the winding thickness to be held to a desired thickness within an appropriate tolerance range. The design also assumes the active flux is available within

an angle range of 90 degrees while the armature will have an angle range of 45 degrees. For desired operation, the armature rotates within an angle range of 40 degrees, which leaves five (5) degrees as a tolerance in the theta domain. Therefore, it is also important that the angle range of the winding is accurate. Thus, it is desirable to fabricate the armature and actuator in such a way that the components (e.g. iron, windings, etc ..) have accurate shapes in every dimension.

[0084] To solve the problem of fabricating an armature within desired mechanical tolerance levels, a mold for use in fabricating an armature is used. Before describing how the mold is used in the fabrication of an armature, the piece parts which comprise the mold are first described in conjunction with FIGs. 12 and 12A. It should be appreciated that the parts which comprise the mold may be provided using conventional manufacturing techniques.

[0085] Referring now to FIGs. 12 and 12A, a mold 130 comprises seven parts 132-144. Parts 132-136 can be secured together to form a cylinder 141 having a diameter which is the same as that of the inner diameter of the armature. For reasons which will become apparent from the following description, the cylinder 141 is sometimes referred to herein as an armature winding fixture 141. Parts 138, 140 are arc-shaped portions of a hollow cylinder, having the same inner and outer diameters of the armature. The parts 138, 140 are preferably provided as matched pieces having substantially identical dimensions. In the exemplary embodiment of FIG. 12, the parts 132- 140 are secured via screws 146, but any fastener or fastening technique which allows the parts 132-140 to be removably secured together can be used.

[0086] Parts 142, 144 again are two identical arc-shaped portions of a hollow cylinder, which share the same outer diameter of the armature and which are adapted to be clamped or otherwise secured over the outer surfaces of parts 138, 140. As will become evident from the description provided hereinbelow, two spaced regions 148, 150 between side edges of the parts 138, 140 are

60150 used to accurately define the shape of the active portions of the armature winding.

[0087] Referring now to FIGs. 13 - 16, the process for forming an armature in accordance with the principles described herein will be described.

[0088] Turning first to FIG. 13, after cleaning the mold parts 132-140, the parts are secured together (e.g. via screws 146) to form a fixture 141 having spaced regions 148, 150 (spaced regions 148, 150 being more clearly visible in FIG. 12A) between edges of parts 138, 140. A caliper or other measuring device can be used to ensure the concave mold sections are even and symmetric before tightening screws 146. Clamp portions 143a (only one clamp portion 143a being visible in FIG. 13) are attached to each side of the fixture 141. A shaft (not shown) may be disposed through the fixture 141 to ensure good clearance before tightening all the screws. Once the fixture 141 is properly formed, the fixture is ready to accept conductive members 170 which will eventually form the armature windings 170.

[0089] Prior to wrapping the conductive member 170 around the fixture 141 , the conductive member 170 (which will form the conductive part of the winding) is first prepared. In one embodiment, an aluminum conductor having a length no less than 720 mm (the aluminum grade is 1050A as rolled with dimensions of 1.3 mm(+/- 0.03..) x 2.3 mm (+/- 0.05 mm) is used. This product is commercially available from AirCraftMaterialsUK.com Ltd, United Kingdom. Other conductive materials, may of course also be used and those of ordinary skill in the art will appreciate how to select an appropriate conductor for a particular application.

[0090] An armature binding member (not shown in FIG. 13) is prepare by cutting sections of a fiberglass sleeve. In one embodiment, seven (7) sections of fiberglass sleeve of 30 mm length each are cut and the sections will be disposed over the aluminum conductor. The sleeve is a product which is commercially available from Varglas Litewall from Varflex Corp, Rome, NY.

[0091] Also, two pieces of pre-preg tape (E-761 with 120-Eglass reinforced) are cut to a length which is a bit greater than 60 mm and a width no less than 35 mm. The pre-preg tape is a product which is commercially available from Park Electrochemical Corp, Waterbury, CT.

[0092] An amount of epoxy (W19) and catalyst (catalyst 11) (both products from Emerson & Cumming, Billerica, MA) are mixed together in a container.

[0093] With all of the above materials preferably being prepared in advance, the winding process begins by disposing one piece of pre-preg tape on each surface of the recessed portions 148, 150 of the fixture 141 (i.e. two pieces of pre-preg tape are used with one piece being disposed in one the recesses 148, 150). One long end of each piece of pre-preg tape touches the mold boundary. With the pre-preg tape in place, one end of the conductive member is disposed on the surface of the recessed section of the fixture 141 over the tape and the conductive member is wound around the fixture 141 staying within the recesses 148, 150. During the winding process, the sleeve sections are properly positioned on the portions of the conductive member which form the end turns. Thus, during the process of winding the conductive member, the pre-preg tape covers the conductive member throughout the winding process to form an insulating and supporting structure.

[0094] As discussed above in conjunction with FIGs. 3-3D, a plurality of different patterns can be used when applying the pre-preg tape to the conductive members. It should also be appreciated that patterns not specifically shown in any of FIGs. 3-3D may also be used. For example the pre-preg tape could be wound around the entire length of the conductive member (e.g. in a spiral pattern) prior the winding process. In this way the, armature binding member could be applied to the entire length of the conductive member before the conductive member is wound or otherwise formed into the desired shape. This approach would also eliminate the need to use sleeves over the end-turn portions of the winding. Still even other

approaches, may of course, be used. Furthermore, it may be desirable to provide the conductive members with an integral insulative cover or surface thereby obviating the need to wrap the conductive members in an insulative cover. For example the aluminum members may be provided having an aluminum oxide surface. Or another alternative may be to apply a coating (e.g. a liquid or semi-liquid coating) to the conductive members (e.g. by a dipping or spraying process) to form an insulative cover over the conductive members.

[0095] Regardless of the specific techniques used, once the conductive member 170 is wrapped around the fixture 141 (as shown in FIG. 13) the remaining portions of the mold are secured to the fixture 141 as will be described below in conjunction with FIG. 13A.

[0096] As mentioned above, to achieve a properly insulated and strong enough armature structure, an armature binding material is used. In one embodiment, armature binding material is provided as an epoxy pre-preg fiberglass tape, epoxy resin material and fiberglass woven sleeve. The active (or straight) turns (e.g. winding portions 22a, 22b in FIG. 2) and end turns (e.g. winding portions 22c, 22d in FIG. 2) of the armature are treated differently. For active turns, an epoxy pre-preg fiberglass tape is used. The epoxy pre-preg fiberglass tape offers insulation and mechanical connection between each turn. It should be appreciated that the material is an electrical insulator.

[0097] As described and illustrated above in conjunction with FIGs. 3-3D, there are several woven patterns suitable for this application. The first three patterns fall into a similar category in the sense that it is necessary to finish the arrangement during (i.e. as part of) the process of forming the winding.

[0098] The patterns shown in FIGs. 3C and 3D, on the other hand, can be achieved before the winding process takes place. Dimensional constraints of the armature and / or actuator and thickness of available pre-preg tape, not all

patterns shown in FIGs. 3-3D may satisfy desired dimension requirements of all applications. Among the patterns shown in FIGs. 3-3C, the first pattern (shown in FIG. 3) offers the most mechanical force to hold the turns together. Therefore, in one embodiment the pattern shown in FIG. 3 was used. In another embodiment, the pattern shown in FIG. 3D was used.

[0099] The decision as to which particular pattern to use in a particular application (e.g. which of the patterns shown in one of FIGs. 3-3D or some other pattern not explicitly shown herein) is based upon a variety of factors including but not limited to the available size, cost, tolerances of the materials, the size and design parameters of the armature and / or the actuator, and the mechanical strength of the resulting structure - e.g. as determined using a pull test.

[00100] As for end turns for the winding, they are pushed or otherwise placed into a slot provided in the clamp portions 143a after the winding process and while making sure that enough space remains between the end-turns so that a shaft can go through them (e.g. as shown in FIGs. 2 and 4). If insulation tape and the pattern of FIG. 3A is used, then it is not necessary to deal with end turns separately since the whole conductor 170 is wrapped with pre-preg tape before the winding process begins. However, if insulation is used with the pattern of FIG. 3, this requires dealing with the pre-preg tape during the winding process. In this case, it is necessary to address the end turns as discussed above (or in some other way) so that they can be moved to the desired position after winding. As discussed above, in one embodiment a fiberglass woven sleeve is chosen to cover the end-turns. The sleeve is placed over the end turns and enough epoxy resin is dropped onto sleeve to ensure strong connection between each of the turns and between end turns and clamps.

[00101] Referring now to FIG. 13A, soon after finishing the winding, the two outside pieces of the mold 142, 144 are disposed over the windings and are coupled (e.g. via a clamp such as a hose clamp) to the fixture 141 so that the

entire mold is assembled with winding inside. The force of the parts 142, 144 over the winding portions disposed in the recesses 148, 150 ensures that the active winding sections are provided having desired shapes and mechanical tolerances.

[00102] Once the active (or straight) portions 170 of the winding (which are in the spaces 148, 150) are tightly clamped within the mold pieces (i.e. by applying pieces 142, 144 as shown in FIG. 13A) it is possible to work on end turns 172 without affecting the shape and position of the effective turns. To achieve this, the structure is first clamped (e.g. using a stationary clamp) together and a tool (e.g. a screw driver or the like) is used to move or otherwise align each end turn 172 into the desired position while being careful not to destroy the sleeves around the end turn conductors 172. The desired positions are the positions that enables the shaft 164 to easily pass through the end turns 172. After this treatment, the structure can be removed from the clamp and an appropriate amount of the mixture of epoxy and catalyst is applied onto each of the end turns 172 (i.e. at each end of the armature straight sections 170). Next, as shown in FIG. 13A the front end clamp portions 143b are applied and screws 161 are used to forcefully secure and make a strong connection between the front and back end clamp portions 143a, 143b with the end-turns 172 disposed therebetween. Again, the shaft 164 is used to make sure there is enough clearance around the end turns 172. The front end clamps 143b are coupled to a shaft 164 via spring pins 166. The armature is now ready to be put through cure process.

[00103] The pre-preg material may be cured using a cure process (step 4) which includes three stages: (a) increase the temperature from a room ambient temperature to a desired peak temperature; (b) hold the peak temperature for a certain period of time; and (c) decrease the temperature back to the room ambient temperature. In this approach, it is not necessary to cure the material under pressure (e.g. using an autoclave) although such a pressure-temperature approach may also be used.

[00104] Referring now to FIG. 14, after cure process, the armature is taken off the mold. This is accomplished by first loosening any clamps (e.g. hose clamps) or other devices or structures securing the parts 142, 144 to the fixture (FIG. 14 shows the structure after the mold parts 142, 144 have been removed). Then the side screws 146 which hold together parts 132-140 are loosened which allows parts 138,140 to be removed. Next parts 132, 136 are removed leaving only part 134 behind as shown in FIG. 15.

[00105] Referring now to FIG. 15, to remove part 134, the part may be rotated by about ninety (90) degrees. This may be accomplished, for example, by securing the structure of FIG. 15 and then using a tool to rotate the part 134.

[00106] The structure of FIG. 15 may be secured, for example, by clamping the side surfaces of the armature end clamps 160 at both ends and then using a tool (e.g. a screw driver, wrench or other tool including a specially designed tool) to rotate the middle piece 134 by about 90 degrees with respect to a central longitudinal axis of the part 134. Once the part 134 is rotated, the shaft can be removed from the armature structure and the mold part 134 can be removed. This leaves the armature structure 180 without a shaft, as shown in FIG. 16. It should be appreciate that the armature structure 180 is the same as the armature structure 18 of FIGs. 2-2B with the exception that the armature 180 does not have the shaft disposed therein. For later convenience in assembling, the shaft will not be connected to the armature via spring pins until the actuator is fully assembled.

[00107] Referring now to FIGs. 17 - 20 in which like elements are provided having like reference designations throughout the several views, a magnetic circuit or iron 12 (also sometimes referred to as an iron yoke 12) comprises five pieces 200, 202a, 202b, 204a, 204b,

[00108] It should be appreciated that in an alternate embodiment, the iron yoke may be provided as a unitary piece out of a piece of mild steel (10xx series). In a preferred embodiment, however, for ease of installation of the

armature and permanent magnets into the air space within the iron, it is desirable to provide the iron from multiple pieces. It should be appreciated that there are many different approaches which can be used to complete the assembly depending, in significant part, upon the procedure to be used to assemble the whole actuator. In this exemplary embodiment, the yoke is built in three layers and from five pieces 200, 202a, 202b, 204a, 204b, as shown in FIG. 17, with screws from the side to secure the pieces together (for clarity, the screws have been omitted from the figures). The three layers from outside to inside will be referred to herein as core 1 (i.e. part 200) , core 2 (i.e. parts 202a, 202b), and core 3 (i.e. parts 204a, 204b) respectively. For this design, in order to make core 1 a continuous part with a full inner hollow cylinder, the width is provided as 33.1 mm. This is done just for convenience when assembling the prototype actuator. For magnetic flux, a width of 28 mm would also be sufficient for this application.

[00109] Referring now to FIGs. 19 and 20, in order to achieve minimum friction from actuator, two bearings 220 are disposed in core 3 at both ends to support a shaft 222. To avoid the complicated and bulky bearing housing outside of the actuator, a recess is formed in core 3 and the bearings are press fit into the inner layer of the iron yoke. Hence, core 3 is adapted to have a properly sized opening 224 (as shown in FIG. 19) at both ends to accept the bearings 220, No reduction of torque output is expected due to the modification of core 3. A stainless steel bearing of the type provided by Mcmaster (57155K358), has an ID of 5/16 inch, an OD of ^λ A inch, and a width of 5/32 inch may be used.

[00110] Referring now to FIGs. 21 - 28 in which like elements of FIGs. 1-2B are provided having like reference designations throughout the several views the process for forming an armature in accordance with the principles described herein are described.

[00111] Turning first to FIG. 21 , assembly of all the parts to obtain a functional actuator begins by attaching the two pieces of core 3 and the two pairs of

magnet spaces together with a fastener. In the exemplary embodiment of FIG. 21 , flat-head screws (for clarity, the screws are not shown in the figures) are disposed through openings 210 and secured via threaded openings provided in the pieces 204a, 204b. Two slim spacers 250 (one spacer being clearly shown in FIG. 21) with a thickness of 1 mm and proper height and length will be inserted into the air gaps between the two pieces of core 3 before assembling. The spacers may be provided from plastic or any other material which does not interfere or detract from the desired operation of the actuator.

[00112] In order to achieve minimum friction from actuator, the two bearings 220 are disposed in both ends to support the shaft.

[00113] Magnet spacers 252 are secured to core 3. In one embodiment, the magnet spacers are provided as aluminum magnet spacer which secure permanent magnets at desired positions while not attracting iron pieces to unwanted positions during the assembly process. The spacers 252 are fixed to the inner layer of iron yoke (e.g. via screws, as shown in FIG. 21).

[00114] Referring now to FIG. 22, a pair of permanent magnets 16a, 16b are disposed in the positions defined by the magnet spacers 252. In one exemplary embodiment, the magnets are provided as high temperature grade NdFeB permanent magnets purchased from NingBo Zhaobao Magnet Co, LTD, China. The size of the magnets are designed as a arc section of a hollow cylinder, which has an ID of 16.4 mm, an OD of 25.4 mm, hence a radial thickness of 4.5 mm, and a arc range of 90 degrees. It should be appreciated that for other applications, magnets having different size, shapes and magnetic characteristics than those described herein may be preferred. Those of ordinary skill in the art will appreciate how to select a particular magnet having a particular size, shape and magnetic characteristic for a particular application.

[00115] The magnet spacers prevent the magnets from moving linearly along axial direction or rotating with respect to the axis during and after it is assembled.

[00116] FIGs. 23 and 24 illustrate the first two steps of assembling the armature.

[00117] Referring now to FIGs. 23 and 24, the armature 18 (without the shaft inserted) is disposed over the top of the permanent magnet -core 3 assembly and the orientation of the armature will allow it to accommodate the assembly. It should be appreciated that at this moment the shaft is not fixed to the front end clamps for convenience to fit armature into the air space within iron yoke. Thus, the armature 18 is disposed over the central core-permanent magnetic assembly.

[00118] Referring now to FIG. 25, a shaft is installed into the assembly of FIG. 24. Since the armature is the only moving element of the actuator, it is preferred to use a shaft with rather big diameter to increase the ruggedness of the armature-shaft structure. For our design, the diameter of the rod is 5/16 inch, very close to that of the first motor and much larger than that of the second motor. We choose to make our shaft from a stainless steel rod with tight tolerance and precision ground finish from Mcmaster (8934K23).

[00119] As shown in FIG. 25, the shaft is installed through the end clamps of the armature and the hole within core 3. Now the armature is rotated with respect to the shaft 90 degrees so the armature will arrive the desired position where the active turns are located right on top of the permanent magnets, as shown in FIG. 25. FIGs. 25 and 26 illustrate the last two steps of assembling the armature.

[00120] Referring now to FIG. 27, after fitting in the armature into the assembled structure (i.e. the structure of FIG. 26), the second layer of iron yoke may be coupled to the structure by putting the two pieces of core 2 into

the positions radially defined by the magnet spacers and the two pieces of core 3.

[00121] Referring now to FIG. 28, the outside layer of iron yoke is disposed over the structure of FIG. 27 and secured to the other portions of the iron to form the iron 12 and thus provide an assembled actuator 10 as shown in FIG. 31. In one exemplary embodiment, the outside layer of iron yoke (i.e. core 1) is secured to the other portions of the iron yoke (i.e. core 2 and core 3) via screws (e.g. flat head screws). Other fastening techniques, may of course, also be used.

[00122] Having described preferred embodiments of the invention it will now become apparent to those of ordinary skill in the art that other embodiments incorporating these concepts may be used. Accordingly, it is submitted that that the invention should not be limited to the described embodiments but rather should be limited only by the spirit and scope of the appended claims.