Electromechanically Actuated Valve

The invention relates to electromechanical actuators, of the type producing a substantially rectilineal force or displacement output, as for operation of a valve member in a pressure-fluid system.

Generally speaking, actuators of the character indicated have adopted the moving-iron principle, wherein output force or displacement is taken from an armature that is movably mounted in the gap or gaps of a fixed magnetic-core system. The mechanical force or displacement output are a function of input current to one or more coils which embrace one or more parts of the magnetic core. Because such a system must rely upon change of magnetic-flux density in the core in order to produce corresponding change in mechanical force output, hysteresis is a fact of life with which to contend, and if provision is made for suitable feedback control of the electrical-input signal, the hysteresis problem is not insurmountable for most applications. However, the inherent inability of such devices to produce a mechanical output which will accurately track input-current magnitude renders them inherently unsuited to use in so-called open-loop control systems.

I shall note the distinguishing difference between electromechanical power amplifiers and transducers. Electromechanical power amplifiers are exemplified by the above-mentioned moving-iron devices. When operated in closed-loop systems, such devices can be effective in spite of a degree of hysteresis, friction and non-linearity, as long as feedback summing with the control variable is handled by transducers of requisite precision; the precision of the transducers determines ultimate accuracy of control. On the other hand, devices suitable as precision transducers in closed-loop control or in open-loop control must be inherently accurate in order to achieve precision control.

To achieve such accuracy in electrical instrumentation, reliance has been placed upon the moving coil, to receive the input current. The coil is mounted for rotation about its central axis of symmetry, and is placed in the single air gap of a magnetic circuit; output force or displacement is the result of reaction between spaced parts of the coil and the magnetic flux in the gap.

To my best knowledge and belief, particularly in fluid-power applications, moving-iron type devices are always chosen, apparently for their high force, high response characteristics, achievable in a minimum of space. And I know of no such employment of the moving-coil principle.

It is, accordingly, an object of the invention to provide an improved precision transducer or actuator device of the character indicated.

Another object is to provide such a device which is essentially free of magnetic hysteresis.

A further object is to provide a device meeting the above objects and which is essentially friction-free and linear in its mechanical-output force or displacement, in response to and over a range of variation in the magnitude of input electrical current.

A specific object is to provide a precision device of the character indicated, lending itself to precision open-loop control situations, such as those necessarily involved in operation of an electro-hydraulic proportional relief valve which avoids use of electronic feedback transducers, i.e., wherein a pilot valve member is caused to produce in response to a given input direct-current signal, an equivalent mechanical force or displacement output, for pilot-pressure command to a main valve.

It is also a specific object to achieve the foregoing with relatively simple structure, lending itself to relatively low cost fabrication and ease of maintenance without sacrifice of inherent precision.

Another specific object is to meet the above objects with an hydraulic-valve construction wherein viscous damping of mechanically moving element utilizes the hydraulic fluid whose flow is controlled by the valve.

Other objects and various further features of novelty and invention will be pointed out or will occur to those skilled in the art from a reading of the following specification in conjunction with the accompanying drawings.

In said drawings, which show, for illustrative purposes only, preferred forms of the invention:

FIG. 1 is a partly broken-away view in perspective to show one embodiment of the invention;

FIG. 2 is a simplified diagram schematically indicating elements of FIG. 1;

FIG. 3 is a view similar to FIG. 1, to show another embodiment for which the diagram of FIG. 2 is also applicable;

FIGS. 4 to 8 are partly broken-away perspective views to illustrate further embodiments;

FIGS. 9 and 10 are sectional views of a specific hydraulic valve of the invention, FIG. 9 being a longitudinal section for the valve in unactuated condition, and FIG. 10 being a transverse section, taken at 10-10' in FIG. 9 for the valve in an actuated condition; and

FIG. 11 is a simplified, fragmentary, partly broken-away perspective view of part of the valve of FIGS. 9 and 10, to show viscous-damping structure.

Referring to FIG. 1, the invention is shown in application to an electromechanical actuator of transducer wherein current resulting from voltage applied to the flexible leads 10 of a coil 11 develops an actuating output force at 12 or 13, as desired along a coil-supporting beam 14, it being understood that the heavy arrows at 12-13 are schematic showings as to the single direction and relative magnitude of force output at the locations indicated. Beam 14 is of light-weight non-magnetic construction, as of aluminum or molded-plastic; it comprises elongate upper and lower arms 15-16, with connecting spacer elements 17-18 from which one of the force outputs 12-13 is selectively available. At its coil-supporting end, a spacer element 19 of beam 14 is shown notched, for bonded reception and mounting of upper and lower portions of the coil 11. At its other end, a frictionless suspension is achieved via a stiffly compliant flat blade 20, providing an upstanding effective axis of pivotal support with respect to fixed base structure 21. The turns of coil 11 may be generally rectangular; they importantly include spaced leg portions 22-23 which are substantially parallel, and they lie in a plane which is generally normal to the elongation direction of beam 14 and which is parallel to and substantially offset from the effective axis of pivotable support, afforded by the compliant means 20.

It is a feature of the invention that each of the spaced legs 22-23 of the single coil 11 is positioned in a different one of two oppositely polarized air gaps established by magnetized core means, shown in FIG. 1 as two like core members 24-25 which are fixedly mounted to non-magnetic supporting bracket structure 26. The core element 24 is generally U-shaped and is freely received in the beam opening between spacer elements 17-19. Element 24 terminates in flat pole faces 27-28 which lie in an upstanding plane that is substantially parallel to the at-rest plane of coil legs 22-23; these pole faces are rectangular and vertically elongate, in proximity to and overlapping register with coil legs 22-23. Element 24 will be understood to be polarized, as for example by using polarizing current to a coil linked to the magnetic core, but I prefer and show use of permanently polarized core material (e.g., Alnico 5), to establish North (N) polarization at 28 and South (S) polarization at 27. In like manner, the respective pole faces 27'-28' of core element 25 are positioned directly opposite the faces 27-28, and the polarization of element 25 is such as to present oppositely polarized poles at the air gaps to faces 27-28, as suggested by "N" and "S" notation applied to the ends of element 25.

Core elements 24-25 and their air gaps (at 27-28) establish a single polarized flux path loop wherein the polarization of flux traversing the gap at 27 is equal and opposite to that which traverses the gap at 28. And since the direction of a direct current flowing in leg 22 is always equal and opposite to that which flows in leg 23, the legs 22-23 will develop identical force reactions, in the same direction in the plane of legs 22-23, to thus produce a turning moment about the effective pivot axis established by the compliance 20. These force reactions are additive and at relatively large radius about the effective pivotal axis of compliance 20, so that force pick-off at 12 or 13 will represent amplification in accord with the applicable mechanical advantage.

FIG. 2 schematically illustrates the basic elements of FIG. 1 structure and affords a convenient means of identifying dimensional relationships. The electromagnetic coil-reaction force is identified F.sub.x and is at the extreme effective radius x determined by compliant suspension of the full length of beam 14. Output-force pick-off at 13 is identified F.sub.y and is at the lesser effective radius Y, thus reflecting an X:Y mechanical advantage at 13. By way of illustration, the pick-off means 13 is shown applied to the mount for a spherical valve member 30 poised at the open discharge end of a passage for continuous hydraulic flow in a pilot valve body 31, as suggested by arrow designation.

In further illustration, for the case of each coil leg 22-23 of 1-inch effective length, a coil 11 of 1000 turns, and for a maximum current (i) of 0.60 amperes in an air-gap flux density of 2000 Gauss, the leg force is 1.35 lbs., being directly calculated from the classic relation:

F = 1.140 .times. 10.sup.-6 B.sub.g.i.L.N

where:

B.sub.g = Air gap flux density (Gauss)

i = Current (amperes)

L = Effective length (inches)

N = Number of turns

This classic relation is normally used in conjunction with radius about a rotary axis to determine torque about that axis, and in such context each coil leg contributes its half of the force (in opposite directions about the axis); but since we are here concerned with a rectilineal output (provided by two coil-leg forces in the same direction), the force F.sub.x is twice that for each leg, being 1.35 lbs. for the stated conditions. And if the radius relation X:Y is 5:1, then F.sub.y is 6.75 lbs. This will be recognized as a highly acceptable maximum control force with which to operate the pilot valve element 30.

The modification of FIG. 3 differs from that of FIG. 1, in that dimensional requirements of the beam 14' for support of the moving coil 11 may be substantially reduced, through a different configuration of the magnetic-core means for establishment of the oppositely polarized air gaps. The compliant suspension remains as previously described, and so the same reference numbers are again used. In FIG. 3, the coil 11 is supported and retained by notches or grooves at the ends of the upper and lower beam members 15', with spacers 17'-18' reduced to accord with the vertical span of coil 11. The oppositely polarized gaps for coil legs 22-23 are provided by like suitably polarized core elements 33-34, each providing the opposed pole faces of a different one of the two gaps. The polarized relation again is designated by "N" and "S" notation, as appropriate, and is seen to conform to the same FIG. 2 schematic which serves FIG. 1. Each of the core elements 33-34 may be of single-piece construction, but dotted lines 33'-34' indicate the abutment plane in the event it should be more convenient to make each of the elements 33-34 in two halves, abutted at 33'-34'. Operation is in every respect as described for FIG. 1.

In the embodiment of FIG. 4, the moving coil 41 is carried at one end of a blade-like beam 42 that is suspended by a relatively stiffly compliant member 43 from fixed mounting means 44. The nature of the suspension is to establish an effective pivot axis, via the bendable mode of member 43, such axis being normal to the plane of the spaced parallel legs 45-46 of coil 41, and such axis being substantially offset from coil 41. The coil legs 45-46 are respectively positioned in oppositely polarized air gaps that are established by and between magnetic-core elements 47-48, in the manner discused in conjunction with elements 24-25 of FIG. 1. Beam 42 may be of suitable non-magnetic material, such as aluminum or plastic, with coil 41 affixed thereto or set therein, as desired. It is convenient to adhere coil leads 49 to one face of beam 42, thus allowing external electrical connection to be made to the device at a beam location of lesser-displacement magnitude. Except for the difference as to effective pivot-axis orientation, operation in FIG. 4 is as described for FIG. 1, the output-force pick-off from beam 42 being suggested at the heavy arrow labeled F.sub.y, based on coil-reaction force development suggested at F.sub.x.

The lower half of the embodiment of FIG. 5 will be recognized for its similarity to FIG. 4, and so corresponding reference numbers are used. The point of difference is that in FIG. 5, the beam 42' is much more elongate; its effectively pivotal, generally central suspension is provided by a single compliant blade 43 with fixed-end mounting 44', shown in ample lateral-clearance relation to the margins of a large central opening 50 in beam 42'. At its other end, beam 42' carries a second electrical input coil 41' which meets the general description of coil 41 and which has opposed parallel legs 46'-47' positioned within the oppositely polarized spaced air gaps of further fixedly mounted magnetic-core elements 47'-48'. It will be understood that upon application of different d-c input signals, at 49-49' respectively, each of the coils 41-41' will develop its own beam-rotating torque about the effective pivot axis established by compliant means 42'. If these input signals are in opposed sense, the output-force F.sub.y ' will reflect the difference between forces developed by the respective coils, and if these input signals are in the same sense, the force F.sub.y ' will represent the sum of the two coil-reaction forces, at a magnitude reflecting mechanical advantage attributable to the relative effective radii of F.sub.y ' pick-off and of the two coils.

Aside from the above-noted algebraic summing property of the FIG. 5 structure, still further advantages will be seen for the case in which all parts are symmetrically formed and balanced with respect to a central axis of the system, for example, as denoted by the phantom line 40. In such event, further summetry is available as to beam suspension by extending the compliant blade 43' so as to provide the same engagement of the fixed amount 44' to both the upper and lower halves of beam 42'. With the mechanical balance thus achievable by symmetry of the beam and its suspension, both symmetries being about the same axis 40, the system is inherently resistive to mechanical shock involving rectilineal bodily displacement or acceleration of the mount 44'.

The embodiments of FIGS. 6, 7 and 8 are all illustrative of the invention in the circumstance in which beam suspension can be dispensed with; for example, to reduce mass and thus inertia to a minimum, or to provide output-force pick-off in the plane of the spaced parallel driving legs of the coil. In all these embodiments, it is convenient to embed the wound coil in a suitable non-magnetic plastic, such as glass-fibre filled epoxy, to enable cured consolidation into a relatively rigid open-rectangular form. In FIG. 6, the moving coil, excited by leads 51, will be understood to be contained or "potted" into the open-rectangular form 52, wherein the spaced parallel driving legs 53-54 are movably positioned in spaced parallel and oppositely polarized air gaps, between magnetic-core elements 55-56 of the nature described for FIG. 1. Integrally formed with the potted-coil form 52 are oppositely projecting mounting arms, with integral mounting bosses 57-58, all in alignment with the axis of coil-force development, being in the plane of and perpendicular to legs 53-54. One or both of the bosses 57-58 may be formed to permanently retain a valve member 59, here shown spherical, as for a hydraulic-flow pilot-valve employment of the nature alluded to at 30-31 in FIG. 2. In order to eliminate friction or mechanical hysteresis in the suspension of coil means 52, I have schematically indicated a totally compliant suspension comprising three equally (angularly) spaced springs 60, connecting each boss 57 (58) to outer frame structure (not shown). With such suspension, it will be understood that valve-member displacements and forces are inherently self-centering on the axis of coil-force development.

FIG. 7 will be recognized for its similarity to FIG. 6 and so corresponding reference numbers are adopted, where applicable. In FIG. 7, however, the potted-coil form 52' is secured to or formed with a stem projection 61 extending in the line of coil-force development. Stem 61 may be part of valve mechanism that is otherwise not shown; for example a spool or other valve member may be formed at or mounted upon the projecting end of stem 61, so that stem 61 may be deemed to be part of valve-member structure. In the event that such structure is guided at its projecting end, there may or may not be need for a compliant support in the region of coil-force development; thus only one such suspension 60 is shown in FIG. 7.

FIG. 8 will also be recognized for its similarity to FIG. 6, and so the same reference numbers are again adopted, as applicable. The point of difference in FIG. 8 is that suspension of the potted-coil form 52" is by way of two spaced parallel frame-referenced compliant blade elements 63-64, shown anchored to the lower ends of the vertical coil-leg forms 53"-54". Force pick-off is schematically indicated at the heavy arrow F.sub.o.

FIGS. 9 and 10 illustrate a specific valve construction embodiment of the invention, particularly suited for employment as the first or signal-stage pilot of a multiple-stage electroproportional relief valve for a high-pressure hydraulic system, as disclosed in greater detail in my copending application, Ser. No. 721,936, filed Sept. 10, 1976.

The housing for the valve of FIGS. 9 and 10 comprises an elongate tubular body 70 having an integral offset portion 71 at one end and removably closed at its otherwise open other end by means of a flanged plug or plate 72, being sealed closed by O-ring means 73; matching local corner flanges or ears 74 on plug 73 and on the adjacent end of body 70 to enable bolted assembly, as at 74', to body 70. The offset portion 71 is of generally rectangular section and is removably closed by a rectangular access plate 75 sealed and secured as by gasket and bolt means, the latter being merely suggested at corner alignments 76.

Hydraulic-fluid flow enters the chamber cavity of the valve via the orifice of a seat fitting 78 secured and sealed to the central region of plate 72; the flow leaves the valve via an outlet passage 77 in plate 72. A ring-shaped base member 79 is mounted to the backside of plate 72 so as to surround the described intlet and outlet passages and to provide suitable mounting for plural angularly spaced coil springs 80 which constitute the force web by which a poppet-valve-member assembly 81 is suspended for coaction with the orifice passage of the seat fitting 78. This assembly 81 will be understood to comprise an enlarged holder for anchorage to the inner ends of springs 80 and positioning a poppet ball adjacent the orifice of seat fitting 78, in the general manner already discussed in connection with elements 30-30' of FIG. 2 and 57-59 of FIG. 6.

The valve-member assembly 81 receives actuating thrust, at substantial mechanical advantage, from a beam 82 which is electromagnetically positioned by force reaction between an excitation winding or coil 83 and a pair of like opposed polarized magnet elements 84-84' secured at 85 to a rearward region of the valve body 70. In the region of valve-member actuation, beam 82 is clamped by means 86 to one end of a compliant blade 87, and the other end of blade 87 is similarly clamped to a local mounting offset portion 88 of the base member 79. This compliant mounting will be understood to effectively establish a beam-pivot axis which is local to the short span between means 86 and the offset 88, said axis being normal to the plane of the paper in FIG. 9.

The mechanism shown for beam connection to the valve-member assembly 81 is seen in FIG. 9 to comprise a force-centering slide 89, guided in a longitudinal bore in the adjacent end of beam 82; a preloading spring 90 is captive in this bore and normally urges the centering slide 89 to the left, i.e., in the poppet-closing direction. The pointed end of slide 89 centers in the flared concave end of a thrust button 91 which transfers thrust to the valve-member assembly 81 via a relatively stiff coil spring 92, to reduce the chance of force misalignment as a function of beam position, a set screw 93 between ends of a groove in slide 89 retains the slide-to-beam assembly without interfering with slide displacement, for a range of possible slide positions in its guide bore.

The valve of FIGS. 9 to 11 is designed for fixed installation to a given machine, control center, or the like and the drawings depict the valve in its installed orientation. Thus, in FIG. 9, the elongation axis of body 70 is horizontal, and beam 82 articulates up and down about its horizontal axis, established by blade 87. To neutralize gravitational moments acting upon beam 82, upper and lower body plugs 94-95 mount suitable independently adjustable means for a "positive"-preload adjustment to a first spring 96 from below and for a "negative"-preload adjustment to a second spring 97 from above. The net of these opposing preload forces is effectively applied at the elongation axis of beam 82 and substantially normal thereto, as will be understood.

The fixed magnets 84-84' will be recognized from descriptions above as establishing upper and lower oppositely polarized gaps, each of which contains one of the vertically spaced parallel elongate legs of coil 83. For convenience, coil 83 is shown wound on a bobbin 98 which is assembled to beam 82 between upper and lower bifurcated end formations 82' thereof.

To complete the description, the valve-body offset 71 is seen to include a short ledge formation 99 to receive the downwardly bent end of a viscous-damper blade 100. Blade 100 is otherwise positioned at the notch of an upstanding bent wire 101 which is clamped at 102 to the upper bifurcation 82', at the outer end of beam 82. The damper blade 100 is of generally rectangular planiform, in relatively close clearance with the adjacent walls of the upstanding body offset 71, as perhaps best seen at 103 in FIG. 9 and at 104 in FIG. 10. Blade 100 is generally centrally apertured to receive wire 101, but the longitudinal position of this aperture is slightly offset from the horizontal center, in the direction away from the beam-pivot axis, so that there will always be a small gravitational moment in the counterclockwise sense of FIG. 9, causing blade 100 to rely upon ledge 99 for effectively hinged support.

In operation, the valve body of FIGS. 9 to 11 is flooded with the hydraulic fluid whose flow is controlled by poppet-member coaction with the orifice of fitting 78. The biasing springs 96-97 effectively neutralize gravitational moments on beam 82, so that force applied in the direction of closing the poppet valve is always in direct response to electrical signal magnitude to winding 83. The nature of force pick-off from beam 82 to the valve-member assembly 81 is such as to reduce friction to an insignificant level, and virtually hysteresis-free operation is achieved. In the presence of sudden large changes in excitation current supplied to winding 83, the blade 100 offers damped reaction to the change as it must be displaced through the hydraulic fluid, in the circumstance of its close clearance relation to adjacent body offset (71) walls.

The described embodiments of the invention will be seen to have achieved all stated objects, and the ability to perform has already been confirmed experimentally. In application of the stated illustrative coil size and polarized flux density, and at input current lends within the indicated maximum, a clearly linear relation has been observed between (a) current taken at ten increments up to 0.5 ampere and (b) hydraulic pressure behind the flow denited by an arrow in the valve structure of FIG. 2, i.e., using the compliant beam suspension and valve-member actuation described in connection with FIG. 2, the only additional further feature being that the valve-member mounting 30' of FIG. 2 was compliantly suspended by a three-spring network as described at 60 for one of the bosses 57-58 of FIG. 6. In these observations, the valve-seat diameter was 0.050 inch, meaning that the required transducer-output force F.sub.y was 1.96 lbs./1000 PSI. The observed linear relationship covered hydraulic pressures to just short of 1000 PSI, but there is absolutely no reason why such linearity cannot extend to a multiple of this pressure, using the described techniques. Importantly, the observed current:pressure relation was completely and reversibly retraceable, with no hysteresis; on the other hand, comparative observation of a stock moving-iron torque motor under comparable ranges of current and pressure revealed characteristic hysteresis, amounting for example to a 75 PSI difference between observed pressure for say a 0.3-ampere input signal, depending upon whether this signal magnitude was achieved under the condition of increasing-current values or under a decreasing-current condition.

Nor is the invention to be deemed to be limited to the described embodiments. For example, the designation 11 for the coil means will be understood also to cover a situation wherein more than one coil is wound to the same basic requirements described for coil 11, both such coils being on to the same form and effectively in registered side-by-side adjacency. One such coil may be used as an open circuit, i.e., with the described response to the input control signal, while the other such coil is connected for closed-circuit use, as to develop a desired rate damping of the compliantly suspended beam; thus connected, rate damping utilizes such induced-current force reaction as is generated by rapid cutting of the magnetic lines of force. Alternatively, by forming both coils in open circuit, there is inherent provision for each coil to serve a different input signal, so that the net force output (F.sub.y) will reflect instantaneous algebraic resolution of the two input signals, with inherent electrical isolation of the circuits relied upon to independently supply these input signals.

Still further, the single compliant blade by which the beam is suspended will be understood to schematically designate a simple (single-spring) or compound (multiple-spring) configuration, additional spring element or elements being compliantly yieldable consistent with the single axis about which the beam is effectively pivoted. Such compounding of compliant elements will be understood to enable a null-balance control adjustment of the beam position, for example, for a given single condition of input current and output force (or hydraulic pressure, as in FIG. 2); such compounding also enables effective selection of the physical magnitude of compliant whereby a particular beam-oscillation frequency characterizes the suspension, to best serve particular application requirements.

While the invention has been described in detail for the preferred forms shown, it will be understood that further modifications may be made within the scope of the invention.