MULTIPLE DEGREE OF FREEDOM COMPLIANT MECHANISM
BACKGROUND OF THE INVENTION
(1) Field of the Invention
The present invention relates generally to precision alignment machines and mechanisms. More particularly, this invention relates to a compliant -mechanism enabling relative movement between a stage portion and a support structure to be controlled with a relatively high degree of accuracy and precision with up to six degrees of freedom.
(2) Background Information
There is a growing need for fine motion control and positioning at meso, micro, and nano scales. Examples include active alignment of components in fiber optics packages, x-y stages with nanometer level resolution, and machine elements for meso- and micro-scale machinery.
Culpepper, in U.S. Patent Application Serial No. 10/005,562, filed 11/8/2001, entitled "Apparatus and Method for Accurate, Precise, and Adjustable Kinematic Coupling", (the '562 patent'), discloses an adjustable kinematic coupling in which one or more of the kinematic elements (e.g., balls and grooves) may be rotated about or translated along an axis thereby effecting a relative movement between two components. The coupling is well suited for applications where alignment with nanometer/microradian accuracy and precision (i.e., repeatability) and/or where controlled adjustment of the relative position of the coupled components is required.
An alternate approach to fabricating machines requiring fine motion control and positioning has employed the use of compliant mechanisms, and in particular monolithic compliant mechanisms. These compliant mechanisms, however, have typically been planar in nature, having the ability to control at most two translational degrees of freedom and one rotational degree of freedom (i.e., x, y, and θz). Examples include a rotational flexure stage for positioning a wafer relative to a microlithography projector disclosed by Barsky in U.S. Patent 5,083,757, entitled "Rotational Flexure Stage"; a precision in plane (i.e., x, y, and θz) stage for optical components disclosed by Hale in
"Principles and Techniques for Designing Precision Machines", Ph.D. Thesis, M.I.T., Cambridge, MA, 1999, p. 184; and a flexure-hinge guided motion nano-positioner disclosed by Elmustafa, et al. in "Flexural-hinge Guided Motion Nano-positioner Stage for Precision Machining: Finite Element Simulations," Precision Engineering, 2001, vol. 25, pp. 77-81.
Next generation applications (e.g., fiber optic alignment) will likely require compliant mechanisms capable of providing high resolution (i.e., nanometer/microradian) position control with six degrees of freedom (i.e., x, y, z, θx, θy, and θz). Therefore there exists a need for new and improved flexures and/or compliant mechanisms that may be suitable for next generation applications.
SUMMARY OF THE INVENTION
One aspect of the present invention includes a compliant mechanism. The compliant mechanism includes a stage and a support both coupled to a plurality of flexure hinges, and at least one tab coupled to at least one of the flexure hinges. The tab is sized and shaped so that displacement of the tab results in a displacement of the stage relative to the support in any one or more of six degrees of freedom.
In one variation of this aspect, the compliant mechanism is of monolithic construction and includes a stage coupled to three flexure hinges. Three tabs are each coupled to a mutually distinct one of the three flexure hinges. A plurality of support beams are coupled to the three tabs, and at least one support member is coupled to the support beams. The support member includes at least one mount for fastening the compliant mechanism to another structure. The three tabs form lever arms and are coupled to the support beams at three fulcrum points. Displacement of any one of the three tabs generates a displacement of the stage relative to the support member(s) and enables the relative position between the stage and the support member(s) to be adjusted in any one or more of six degrees of freedom.
In another aspect, this invention includes an apparatus of a substantially monolithic construction including first and second reference frames, and at least one flexure hinge coupled therebetween. An actuator is coupled to the flexure hinge(s). Movement of the actuator generates displacement of the first reference frame relative to the second reference frame in any one or more of six degrees of freedom.
In still another aspect, this invention includes a method of aligning a first component and a second component to one another. The method includes using a compliant mechanism including a stage coupled to a plurality of flexure hinges, at least one tab coupled to one of the flexure hinges, and at least one support coupled to the tab(s). The method further includes fastening the first component to the stage, fastening the second component to the support, and displacing the tab(s) to effect a change in position of the first component relative to the second component in any one or more of six degrees of freedom.
In a further aspect, this invention includes a method of fabricating a compliant mechanism. The method includes providing a substantially planar work piece, forming a stage in the work piece, the stage being coupled to a plurality of flexure hinges, and forming at least one tab in the work piece, the tab(s) being coupled to one of the flexure hinges. The method further includes forming at least one support in the work piece, the support being coupled to the tab(s). The tab is sized and shaped so that displacement of the tab(s) results in a displacement of the stage relative to the support member(s)in any one or more of six degrees of freedom.
BRIEF DESCRIPTION OF THE DRAWINGS
The aforementioned invention will now be described with reference to the accompanying drawings in which:
Fig. 1 is a perspective view of one embodiment of an adjustable kinematic coupling of the present invention with kinematic elements disengaged;
Fig. 2 is an exploded view of the embodiment of Fig. 1;
Fig. 3 is a perspective view of an axi-symmetric convex kinematic element of Fig. 1;
Fig. 4 is a top view of the convex kinematic element of Fig. 3;
Fig. 5 is a section view illustrating the structural connections between the convex kinematic element of Fig. 3 and the components to which it is attached;
Fig. 6 is an elevational view of the coupling of Fig. 1 with the kinematic elements engaged;
Fig. 7 is a top plan view of the coupling of Fig. 1 with the kinematic elements engaged in a default position and with portions thereof removed for clarity;
Fig. 8 is a similar view to that of Fig. 7 after one of the convex kinematic elements has been rotated away from the default position;
Fig. 9 is a partially cross-sectional view of an alternate embodiment of an adjustable kinematic coupling of this invention;
• Fig. 10 is a perspective view of an alternate embodiment of a concave kinematic element of this invention;
Fig. 11 is a perspective view of an alternate embodiment of a convex kinematic element of this invention;
Fig. 12 is a top plan view of an another alternate embodiment of a convex kinematic element of this invention;
Fig. 13 is a perspective view of the convex kinematic element of Fig. 12;
Fig. 14 is a perspective view of a brake that may be utilized to lock the position of actuated shafts within the adjustable kinematic coupling of this invention; and
Fig. 15 is an elevational view of the coupling of Fig. 1 with kinematic elements engaged and showing a pin used to calibrate the kinematic joint.
Fig. 16 is a plan view of one embodiment of a compliant mechanism fabricated in accordance with the present invention;
Fig. 17 is a plan view of another embodiment of a compliant mechanism fabricated in accordance with the present invention; and
Fig. 18 is a plan view of still another embodiment of a compliant mechanism fabricated in accordance with the present invention.
Fig. 19a is a plan view of the compliant mechanism of Fig. 16;
Fig. 19b is a plan view of the compliant mechanism of Fig. 19a showing the effect of an exemplary actuation of one tab;
Fig. 20a is a plan view of the compliant mechanism of Fig. 16 showing the effect of an exemplary actuation of two tabs;
Fig. 20b is a plan view of the compliant mechanism of Fig. 16 showing the effect of an exemplary actuation of three tabs.
DETAILED DESCRIPTION
Referring to the accompanying figures, the present invention is directed to a kinematic coupling, referred to herein as an adjustable kinematic coupling, that may meet the stringent demands of next generation processes. The coupling is well suited
for applications where alignment with sub-micron accuracy and precision (i.e., repeatability) and/or where controlled adjustment of the relative position of the coupled components is required. Exemplary applications to which this invention may be well suited include, but are not limited to, precision automation, precision actuated motion stages, optical mounts and assemblies such as precision fiber optic alignment machines, semiconductor and microelectromechanical mask alignment, structures with integrated precision actuation methods, and other precision alignment devices.
In one embodiment, the present invention utilizes the concept of providing a space between an axis of rotation and an axis of symmetry of a convex member (e.g., a ball). As a result, precision alignment procedures are typically not required. Rather, the convex members may be provided with alignment cavities that allow .the interface to be coupled and calibrated for position control with nanometer level errors. In one aspect, this invention includes a method of attaining precision alignment using a kinematic interface. Elements of the kinematic interface are attached to one or more components so that they may be translated along and/or rotate about an axis (referred to herein as an axis of rotation) that is static with respect to the component to which the elements are attached. In a generally desirable embodiment, the kinematic elements are coupled to the components in a manner that results in the axis of rotation being parallel with the mating direction of the coupling and spaced from an axis of symmetry of the kinematic element. Rotation of the kinematic element about and/or translation along the axis of rotation, tends to result in relative movement between two reference frames, one fixed to a first component, the other fixed to a second component. As used herein, the term 'mating direction' refers to a direction of movement of one or both of the components that serves to engage the kinematic elements of these components with one another. For example, in the embodiments shown, the mating direction is along the 'z' axis, such as shown in Fig. 6.
The present invention is generally advantageous in that it provides an improved, relatively low cost kinematic coupling and method which enables repeatable location of two or more components, surfaces, assemblies, and the like, which overcomes at least one of the above-described limitations of prior couplings. Another advantage of this invention is that it may provide a novel coupling in which the relative position of the coupled components may be adjusted in any one or more of the six degrees of freedom by controlled actuation of kinematic elements. Yet another advantage of this mvention
is that it may provide for an adjustable kinematic coupling with a sufficiently high displacement ratio (ratio of actuator input motion to coupling (i.e., output) motion) to produce small output movements (on the order of tens of nanometers or arc seconds) with inputs that may be several orders of magnitude larger. Still another advantage of this invention is that it may provide for a modular adjustable kinematic interface in which the resolution (as well as other aspects) of the motion control and displacement ratio may be varied via interchangeable kinematic elements with different geometries.
The present invention is further advantageous in that it may provide for nonlinear motion control with a linear actuator input, by varying the geometric size, location, and/or orientation of the kinematic elements. Alternatively, the size, location, and/or orientation of the kinematic elements may also advantageously be configured to provide for linear motion control with non-linear actuator inputs. Yet a further advantage of this invention is that it provides an adjustable kinematic interface that tends to minimize the risk of damage to the concave kinematic elements by providing two possible input states to achieve a desired position adjustment, thereby serving as a backup should the geometry required by the first state be damaged.
This invention tends to be further advantageous in that accurate machining of the kinematic elements (e.g., to within micron levels) is not necessary as errors in their shape, orientation, and form may be mapped and compensated for. In addition, accuracy of size and shape of the various features (e.g., of the kinematic elements) of the present invention, and of the actuation method used to impart motion between the coupled components, is relatively less critical owing to the ratio of actuator motion to coupling motion that is a result of the coupling's geometry. This displacement ratio makes motion control on the order of tenths of microns possible even when the coupling components' feature sizes, orientation, and locations vary from nominal by tens of microns.
These and other advantages of this invention will become evident in light of the following discussion of various embodiments thereof.
Referring now to Fig. 1, one embodiment of an adjustable kinematic coupling (shown decoupled) of the present invention is shown. The coupling includes three concave elements (e.g. three spaced grooves) 8a, 8b, and 8c attached to or machined into the inner surface 3 of one (e.g., a lower) component 4, and three corresponding convex elements 6a, 6b, and 6c attached to another (e.g., an upper) component 2 by
shafts 22a, 22b, and 22c (Fig. 2). The convex elements 6a, 6b, and 6c may or may not be spherical, but nevertheless may be referred to hereinafter as "balls", which is in keeping with the tradition of those skilled in the art. The mating of the coupling shown in Fig. 1 is generally accomplished by bringing each ball 6a, 6b, and 6c into contact with a corresponding groove 8a, 8b, and 8c, which typically results in surfaces 1 and 3 being substantially parallel and separated by a finite gap 33 (Fig. 6). The resulting mate forms six points of contact 9a, 9b, 9c, 9d, 9e, and 9f between balls 6a, 6b, 6c and corresponding grooves 8a, 8b, and 8c (Figs. 6 and 7). The constraints imposed by points of contact 9a, 9b, 9c, 9d, 9e, and 9f provide deterministic constraints of the six degrees of freedom between the components 2, 4.
Referring now to Figs. 2, 5, 7 and 8, a generally desirable embodiment of this invention is illustrated. Balls 6a, 6b, and 6c are rigidly attached to shafts 22a, 22b, and 22c using holes 5a, 5b, and 5c. The shafts 22a, 22b, and 22c are assembled to the balls 6a, 6b, and 6c such that the axes 10a, 10b, and 10c (Fig. 4) of shafts 22a, 22b, and 22c are offset (i.e., spaced) from the axes of symmetry 12a, 12b, and 12c, of balls 6a, 6b, and 6c. The shafts 22a, 22b, and 22c are constrained radially by bearings 20a, 20b, and 20c that allow the shafts 22a, 22b, and 22c to rotate about, and translate along, their axes 10a, 10b, and 10c. These bearings 20a, 20b, and 20c may operate on fluid, pneumatic, magnetic, contact, or any design in which rotation and translation in directions that are orthogonal to the axes of bores 11a, lib, and lie are minimized. For clarity, in many instances herein (e.g., in Figs. 3-6 and 9-15) elements operatively associated with only one (e.g., 6a) of the three balls 6a, 6b, 6c may be shown and/or described, with the understanding that such depiction/discussion similarly applies to the remaining balls (e.g., 6b, 6c) and their associated elements. For example, bores lib and lie (associated with balls 6b and 6c, respectively) are not shown in the Figs., with the understanding that they are substantially similar to bore 11a (associated with bore 6a). Should bearings 20a, 20b, and 20c rely on physical contact with the shafts 22a, 22b, and 22c, the bearings 20a, 20b, and 20c may be suitably sized and shaped to result in a press fit between the bearings 20a, 20b, 20c and shafts 22a, 22b, 22c, to thereby increase the stiffness of the coupling.
Bearings 20a, 20b, and 20c may be constrained within bores 11a, lib, and lie using a relatively stiff interface, such as may be provided by press fitting or potting in place. In some applications, snap rings 15a, 15b, 15c, 24a, 24b, and 24c may be placed
in axi-symmetric grooves 17a, 17b, 17c, 17d, 17e and 17f to constrain the axial movement of bearings 20a, 20b, and 20c relative to bores 11a, lib, and lie. The bearings 20a, 20b, and 20c may be sealed from outside contaminates using seals 14a, 14b, 14c, 25a, 25b, and 25c.
Brakes 26a, 26b, and 26c are typically included to selectively prevent and/or resist rotary and linear translation of shafts 22a, 22b, and 22c. In the embodiment shown, the brakes provide a constant resistance to both rotary and linear movement, so that once adjusted, the components 2, 4 tend to remain in their desired positions. Alternatively, actuatable brakes (not shown) such as electrically or hydraulically actuatable brakes similar (but reduced in size) to those used in automotive applications, may be used. In the embodiment shown, the brakes 26a, 26b, and 26c are typically placed coaxially with bores 13a, 13b, and 13c (Fig. 5) and are attached thereto in a manner that does not over constrain the bearings 20a, 20b, and 20c. This may be accomplished, for example, by use of flexure elements 74a, 74b, 74c, 74d, 74e, and 74f (Fig. 14) that permit some radial motion, but provide a high stiffness constraint of motion in other directions. Flexure elements or other compliant elements that are not integral to the braking mechanism may also suffice. As also shown, brakes 26a, 26b, and 26c may use friction pads 76a, 76b, and 76c (Fig. 14) or other similar elements commonly used in rotary and axial clutch applications to constrain shafts 22a, 22b, and 22c.
To impart rotary and linear translation of shafts 22a, 22b, and 22c, couplings 28a, 28b, and 28c are provided to couple shafts 22a, 22b, and 22c to motion input from standard rotary and/or linear actuation mechanisms that attach to the upper component 2 via three or more sets of attachment points 18a, 18b, and 18c.
Components 2 and 4 are engaged in their deterministic position by engaging balls 6a, 6b, and 6c with corresponding grooves 8a, 8b, and 8c. Those skilled in the art will readily understand that after the initial mating engagement, a nesting force may be applied in a direction substantially parallel to the direction of mating to maintain contact between the balls 6a, 6b, and 6c and grooves 8a, 8b, and 8c. The actuation of one or more of shafts 22a, 22b, and 22c in rotation about their respective axes 10a, 10b, and 10c generates relative motion in the x, y, and θz directions between components 2 and 4. Such relative motion is effected due to the offset between the axes of the shafts and the axes of symmetry of the balls, as will be described in greater detail hereinbelow.
Similarly, actuation of one or more of shafts 22a, 22b, and 22c in translation along their respective axes 10a, 10b, and 10c generates relative motion in the θx , θy, and z directions between components 2 and 4.
Referring now to Figs. 7 and 8, more detail regarding the adjustable nature of the coupling is provided. For clarity, component 2 has been removed to show balls 6a, 6b, and 6c. Fig. 7 shows balls 6a, 6b, and 6c mated into corresponding grooves 8a, 8b, and 8c. Planes containing the pairs of axes 10a and 12a, 10b and 12b, and 10c and 12c, and which are peφendicular to surface 3, are referred to herein as rotation planes. As shown in Fig. 7, these rotation planes are coincident with the respective planes of symmetry 32a, 32b, and 32c of the corresponding grooves 8a, 8b, and 8c. (Planes 32a, 32b, and 32c are viewed on edge and thus appear as lines in Fig. 7.) This position, in which the planes of symmetry and rotation planes are coincident, is referred to as the "default position" of the coupling.
As also shown, a coordinate system (e.g., Cartesian Coordinates) may be attached to each component 2, 4. In the default position, the coordinate systems are coincident with each other and, for example, with their origins located at the intersection of planes 32a, 32b, 32c, and surface 3. Movement of component 2 relative to component 4, results in the two coordinate systems being offset (i.e., so that they are no longer coincident), as they may be considered to be "rigidly" attached to their respective components. This is illustrated in Fig. 8 in which the coordinate system of component 2 has been translated and rotated with respect to the coordinate system of component 4 due to actuation of ball 6a. The coordinate system attached to component 4 is shown as x-y and the coordinate system attached to component 2 is shown as x'-y'.
Turning now to Fig. 8, the relative movement of the components 2, 4 is described in greater detail. The coupling shown is substantially identical to that of Fig. 7, with the exception that ball 6a has been rotated about axis 10a through an angle ΘE- The rotation plane (which as described above includes axes 10a and 12a), is no longer coincident with plane 32a. Axis 10a is therefore offset some perpendicular distance from plane 32a. During actuation, balls 6a, 6b and 6c may slide in grooves 8a, 8b and 8c to maintain geometric compatibility. The result is a displacement of the coordinate systems in the x, y, and θz directions. Moreover, although in example shown only element 6a has been rotated relative to component 2, elements 6b and 6c may also be
rotated separately or jointly to affect the relative positions of components 2 and 4 in a controlled (x, y, and θz.) and mathematically predictable manner.
Referring again to Figs. 2, 5, 7, and 8, actuation of shafts 22a, 22b, and 22c along their axes 10a, 10b, and 10c serves to alter the distance between and the angular orientation between planes containing surfaces 1 and 3. Such actuation of the shafts 22a, 22b, and 22c thus serves to displace the coordinate systems in the θx, θy, and z directions relative to one another. Shafts 22a, 22b, and 22c may also be used to translate components 2 and 4 relative to each other so as to bring surfaces 1 and 3 into surface-to- surface contact, e.g., to form a hermetic seal. This combination of linear and rotary actuation of shafts 22a, 22b, and 22c thus enables relative movement in all six degrees of freedom, i.e., in the x, y, z, θx, θy, and θz directions.
The deterministic mate between component 4 and balls 6a, 6b, and 6c in conjunction with the defined relationship between elements 6a, 6b, and 6c and upper component 2 allows one to mathematically model the geometric relationship between the upper and lower components 2 and 4. With respect to motion in the x, y, and θz (shown as ΘE in Fig. 8) directions, the mathematic solution provides two possible inputs to achieve a desired position. For example, when ΘE is zero, the coupling is in its default position and the coordinate systems for components 2 and 4 are coincident, as discussed hereinabove. Similarly, when ΘE is 180 degrees, the coupling may also be in its default position. The skilled artisan will readily recognize that the rotary actuation of elements 6b and or 6c alone or in combination with one another and/or with element 6a achieves similar results. However, in particular embodiments, ΘE may be intentionally limited to motion between -90 degrees and 90 degrees (i.e., within a range of 0 +/- 90 degrees) to establish a one to one relationship between a combination of rotations of shafts 22a, 22b, and 22c and a relative position between components 2 and 4. Advantageously, in the event part of the surfaces of grooves 8a, 8b, and 8c become damaged, balls 6a, 6b, and 6c may be restricted to rotate within a range 180 degrees offset from the foregoing range, i.e., the balls may be rotated within a range of 180 +/- 90 degrees. In this manner, a different section of grooves 8a, 8b, and 8c may be used to achieve the same position control. This feature reduces the risk of inoperability due to equipment crashes, surface damage from dings, and may effectively double the useful life of the coupling with respect to wear from general use.
One advantage of various embodiments of the present invention is that they provide a coupling having a variable displacement ratio. As set forth hereinabove, the displacement ratio is defined as the ratio of actuator input motion to coupling (output) motion.
For the embodiment shown in Figs. 2, 7 and 8, the displacement ratio is related to the magnitude of the offset between the axis of rotation 10a, 10b, and 10c and the axis of symmetry 12a, 12b, and 12c, and the angle, ΘE to which the balls 6a, 6b, and 6c are rotated relative to the default position. In the default position, the coupling has a relatively low displacement ratio, which enables components 2 and 4 to be moved relatively large distances relative to one another for a given input motion. For example, for an embodiment in which the offset is about 125 microns, rotating ball 6a through an angle ΘE of about 0 to about 5 degrees results in a relative x translation of the coordinate system x'-y' of about 7 microns.
Conversely, at an angle ΘE of about 90 degrees the coupling has a relatively high displacement ratio, enabling relatively small movements of components 2 and 4 relative to one another. For example, in the above-described embodiment having an offset of about 125 microns, rotating ball 6a through an angle ΘE of about 89 to about 90 degrees results in an x translation of coordinate system x'-y' of about 15 nanometers. This feature of variable displacement ratio thus tends to be highly advantageous in that it enables both relatively fine and relatively coarse motion control in a single coupling (e.g., from on the order of tenths of microns or less, to tens of microns or more). The range of actuation along axes 10a, 10b, and 10c depends in part on the gaps 30a, 30b, and 30c between the component 2 and the upper surfaces 19a, 19b, and 19c of the balls 6a, 6b, and 6c. The range of rotational actuation of the shafts (i.e., about axes 10a, 10b, and 10c) may be dependent primarily on the method of actuation. For example, the shafts may be actuated by rotary actuators (e.g., conventional servo motors), and/or linear actuators (which may include rotary actuators, as described hereinbelow), which may be controlled by a conventional microcontroller or microprocessor, such as shown and described with respect to Fig. 9 hereinbelow. The default position may be calibrated by actively measuring the relative position of components 2 and 4 or by forming detents (such as with spring-loaded balls or pins 90 (Fig. 15)) configured for engagement with both alignment cavities 7a, 7b, and 7c and with component 4 (e.g., with grooves 8a, 8b,
and 8c), when the balls are disposed in a predetermined position. For example, in the embodiment shown, in which the plane of symmetry of alignment cavities 7a, 7b, and 7c is coincident with the corresponding planes defined by axes 10a and 12a, 10b and 12b, and 10c and 12c, the detent pins 90 will engage the alignment cavities 7a, 7b, 7c, when the planes of symmetry of each of the alignment cavities are coincident with the planes of symmetry of each of the grooves 8a, 8b, and 8c, respectively, to define the default position. In the example shown in Fig. 15, the pin 90 is spring-loaded (biased) along the plane of symmetry of groove 8a, towards ball 6a. This spring-loading may be accomplished in any convenient manner, such as, for example, by a coil spring (not shown) disposed within the groove 8a. In this manner, a nose portion of pin 90 suitably sized and shaped, may be biased into mating engagement with alignment cavity 7a when the ball is moved into the default position. (Similar alignment cavities are shown as 49 and 56 in Figs. 11-13, discussed hereinbelow.)
Another design consideration is the friction between balls 6a, 6b, and 6c and grooves 8a, 8b, and 8c. Friction forces between these elements may result in poor precision since they may prevent balls 6a, 6b, and 6c from settling into their lowest energy state within grooves 8a, 8b, and 8c. Friction may also contribute to sliding wear and fretting at the point contacts. These problems tend to be common to any kinematic coupling and may be addressed by using friction-reducing methods such as lubrication or low friction coatings. It may also be desirable to use mutually distinct materials for the balls 6a, 6b, and 6c and the grooves 8a, 8b, and 8c to prevent cohesion between micron and/or nanometer scale surface asperities. It is typically further desirable that the materials used for the balls 6a, 6b, and 6c and grooves 8a, 8b, and 8c possess relatively high surface energy, which tends to make the detachment) of asperities from the surfaces of the kinematic elements more difficult. This may be effected by use of wear resistant coatings and hardened materials. Note that the detachment of particles from the surface of the elements is preferably avoided as the detached material may re- weld to the surface or combine with other particles to form relatively large particles. Avoiding particle formation, in particular large particles, is desired since their presence between the kinematic elements tends to cause location errors between the coupled components. In some cases, long term wear errors may be mapped and calibrated out of the coupling, however particles between the kinematic elements may move into and out
of the contact zone at random and are desirably minimized using the methods described above as well as by other common methods known to those skilled in the art.
In alternative embodiments, the inventive coupling may also be used, as stated hereinabove, in the precision alignment of product components, of parts to machine tool fixtures, of machine tool fixtures to machines, of casting mold portions, and the like. In some applications, balls 6a, 6b, and 6c may be spaced to form triangles that are not substantially equilateral. This may be beneficial in that it renders motion control in a predetermined direction more or less sensitive to actuator input. Further, the need to use non-equilateral geometry may arise in applications in which the structure of the mating components does not permit equilateral spacing.
Other embodiments may be actuated by more than three shafts. For example, Fig. 9 illustrates an alternate embodiment, including another shaft (also referred to herein as a threaded shaft) 54 that controls axial movement of the shaft 42 with respect to an upper component 52. The shaft 42 may be rotated by a knob 34 that is rotationally isolated from the threaded shaft 54, yet preloaded by a force source 36 to eliminate slop between the shaft 42 and the top component 52. This force source may include a compression spring, magnets, air bladder, piezoelectric actuator, magneto-strictive actuator, or other force sources known to those skilled in the art. The threaded shaft 54 is decoupled from the shaft 42 by a low friction interface 38 and by force source 36 and therefore tends to not transmit torque to the shaft 42. Threaded shaft 54 may be operated manually, or alternatively, may be operated automatically, such as by a servo motor 80 engaged therewith, and which may be controlled by a controller (e.g., a programmed microprocessor) 84, as shown in phantom. The combination of threaded shaft 54, servo motor 80, and controller 84 thus form an automatic linear actuator. Knob 34 may be similarly actuated by a servo motor 82 (also shown in phantom) and controller 84. The skilled artisan will recognize that any suitable actuator or combination of actuators, manually or automatically activated, may be used to move shafts 42 along their axes, without departing from the spirit and scope of the present invention.
The shaft 42 is radially constrained by a bearing set 40. The shaft is offset from the center 50 of the spherical kinematic element 48 while it is housed in hole 47 (Fig. 10). The top component 52 and bottom component 46 are mated using the kinematic interface, with point contacts 56a and 56b. Alignment cavity 49 (Fig. 11) may be used
to calibrate the coupling in the manner described hereinabove with respect to alignment cavities 7a, 7b, and 7c.
Moreover, in addition to utilizing physical contact to calibrate the coupling as described hereinabove, non-contact means of sensing the position and capturing data may be used to perform calibration (e.g., to identify the default position). Examples of such non-contact means include conventional vision systems, laser systems, and field effect proximity sensors.
Referring now to Fig. 10, in another embodiment, one may use parts 58 that replace the monolithic groove pattern 8a, 8b, and 8c (Fig. 1) described hereinabove. A part 58 including a groove 55 is equipped with a hole 60 that is offset from the groove's 55 plane of symmetry. The groove 55 may be attached to one of the coupled components in a manner similar to the attachment scheme for balls 6a, 6b, and 6c (Figs. 1 and 2) described hereinabove. This may enable position control by actuation of die grooves rather than of the balls. In this manner, the actuatable parts 58 may engage a set of balls that are rigidly disposed on a component. Alternatively, for some applications it may be desirable to enable both the balls and grooves to be actuatable.
Referring to Figs. 12 and 13, another embodiment of a convex kinematic element (element 60) is illustrated. In contrast to a previously described convex element (ball 6a in Fig. 4), element 60 is not axi-symmetric, but rather has planes of symmetry through axis 64. Element 60 retains the feature of a hole 68 that is offset from an axis 64 that is formed by the intersection of one plane of symmetry and contact points 62a and 62b. The contact points 62a and 62b are shown for the default position. Rotation of element 60 about axis 66 may result in relative motion of the components in the x, y, and θz directions as described hereinabove, and, since element 60 is not axi-symmetric, may also result in relative motion of the components in the θx, θy, and z directions. This change in geometry may lead to coupling between the six degrees of freedom. The artisan of ordinary skill will readily recognize that convex elements of many shapes, ranging from asymmetric to axi-symmetric, may be utilized to affect relative motion of the components depending on the demands of a particular application. The movement provided by an element 60 of a particular geometry may be mathematically modeled, and/or may be mapped as described hereinabove.
There are many variable surface profiles for grooves and balls that can be used to alter the repeatability, resolution, displacement ratio, and position control capabilities of the coupling.
Other embodiments of the invention may use features of different geometry and location for calibrating the coupling. As with the embodiments described hereinabove, these features may appear on the kinematic elements or other parts of the coupling.
Other variations, modifications, and other implementations of what is described herein will also occur to those of ordinary skill in the art without departing from the spirit and the scope of the invention as claimed. Accordingly, the invention is to be defined not just by the preceding illustrative description, but instead by the spirit and scope of the following claims. Referring to the accompanying figures, an alternate embodiment of the present invention is directed to a flexure based mechanism, also referred to as a compliant mechanism or a planar compliant mechanism that may meet the stringent demands of next generation processes. The compliant mechanism is well suited for applications where alignment with nanometer, micron or millimeter scale accuracy and precision (i.e., repeatability) and/or where controlled adjustment of one component (e.g., an optical fiber) relative to another component (e.g., an optical lens) is required. Exemplary applications to which this invention may be well suited include, but are not limited to, precision automation, precision actuated motion stages, optical mounts, and assemblies such as precision fiber optic alignment machines, semiconductor and microelectromechanical mask alignment, structures with integrated precision actuation methods, micro testing and measurement devices, and other precision alignment devices.
In one embodiment, the present invention includes a planar compliant mechanism having an inner stage coupled to one or more relatively simple compliant elements, such as flexure hinges and tabs, that may utilize both elastic and plastic material deformation. Actuation of one or more of the tabs tends to result in relative movement between two reference frames, one fixed to the inner stage, the other fixed to a support member. In general, in plane -actuation of the tabs tends to result in in plane motion while out of plane actuation tends to result in out of plane motion of the inner stage relative to the support member (in plane motion typically refers to motion in the x, y, and θz directions, while out of plane motion typically refers to motion in the θx, θy, and z directions). In a generally desirable embodiment, an inner stage is coupled to
three flexure hinges, which are further coupled to three tabs. Although each flexure hinge is shown being axially aligned with corresponding tabs, the tabs may be offset or otherwise disposed off-axis relative to their corresponding flexure hinge, without departing from the spirit and scope of the present invention. Actuation of any one or more of the three tabs may provide for precision alignment in up to six degrees of freedom.
The present invention may be advantageous in that it provides for an improved compliant mechanism and method which enables accurate and repeatable location of two or more components, surfaces, assemblies, and the like, which overcomes at least one of the above-described limitations of prior alignment mechanisms. Another potential advantage of this invention is that it enables the relative position of the coupled components to be adjusted in any one or more of the six degrees of freedom by controlled actuation of one or more tabs. Yet another potential advantage of this invention is that the relative positions of the coupled components may be repeatedly adjusted. Still another potential advantage of this invention is that it may utilize both elastic and plastic deformation of the compliant mechanism's structure. A further advantage of this invention is that it may provide for an adjustable compliant mechanism with a sufficiently high displacement ratio (ratio of actuator input motion to output motion) to produce relatively small output movements (potentially sub- nanometer) with inputs that may be at least one order of magnitude larger.
This invention tends to be further advantageous in that it may be adapted for use with micro, meso, and macro scale applications. Furthermore, the tabs may be actuated using substantially any known, or yet to be developed, means. For example, comb drives may be utilized for MEMs applications, piezo electric actuators for meso scale machines, and conventional mechanical actuators for meso-macro scale applications.
Embodiments of this invention may be still further advantageous in that they are of a monolithic construction, which tends to reduce and/or eliminate friction induced hysteresis and wear that results from the repeated rubbing of components against one another, such as in a conventional contact based or segmented mechanism. These and other advantages of this invention will become evident in light of the following discussion of various embodiments thereof.
Referring now to Fig. 16, one embodiment of a planar compliant mechanism 100 of the present invention is illustrated. The compliant mechanism 100 is typically
(although not necessarily) of monolithic construction (i.e., made from a single piece of material). The compliant mechanism 100 includes a stage 110, coupled to a plurality (e.g., three) of flexure hinges 120a, 120b, and 120c, which may be uniformly spaced thereabout (i.e., in the form of a substantially equilateral triangle). Inner stage 110 may optionally include a center aperture 112 through which a chuck or some other component may extend. Stage 110 may further optionally include other features, such as but not limited to, holes, slots, grooves, and the like for mounting one or more components thereto.
Flexure hinges 120a, 120b, and 120c are coupled to inner stage 110 at hinge points 122a, 122b, and 122c, respectively. At least one of the flexure hinges 120a, 120b, and 120c is further coupled to a tab 130a, 130b, and 130c. In a desirable embodiment, each of three flexure hinges 120a, 120b, and 120c is coupled to a mutually distinct tab 130a, 130b, and 130c. Tabs 130a, 130b, and 130c typically extend radially outward along radial axes 132a, 132b, and 132c that typically pass through the center point 114 of the inner stage 110. Axes 132a, 132b, and 132c are fixed relative to support member(s) 150, with their position(s) being unchanged by actuation of the tabs 130a, 130b, and 130c. The tabs 130a, 130b, and 130c effectively function as lever arms and are coupled to at least one support beam 140 at at least one of fulcrum points 134a, 134b, and 134c about which they pivot. The term "fulcrum" is used herein in a manner consistent with the conventional dictionary definition, i.e., a pivot point about which a lever arm operates (Academic Press Dictionary of Science and Technology, 1992). Actuation of tabs 130a, 130b, and 130c is discussed in more detail hereinbelow.
The support beams 140 may be configured in substantially any manner. For example, in compliant mechanism 100 the support beams are configured in a substantially equilateral triangular pattern rotated about 180 degrees (i.e., half a turn) out of phase with the triangular inner stage 110. Support beams 140 may further be coupled to at least one support member 150, for example at the corners 142 of the triangular pattern of support beams 140. Support member(s) 150 are typically adapted to provide for mounting the compliant mechanism 100 to another structure. For example, compliant mechanism 100 includes a support member 150 in the form of an outer support ring (i.e., a circular portion that encloses and supports the other portions of the compliant mechanism) that includes a plurality of holes 152 configured for fastening (e.g., screwing) the compliant mechanism to another fixture. As described in more
detail hereinbelow, actuation of tabs 130a, 130b, and 130c causes relative movement of the inner stage 110 with respect to support member(s) 150.
Referring now to Figs. 17 and 18 two alternate embodiments 100' and 100" of the compliant mechanism of this invention are illustrated. Compliant mechanisms 100' and 100" are similar to that of compliant mechanism 100 in that each includes an inner stage 110', 110" coupled to a plurality of flexure hinges 120a', 120b', 120c', and 120a", 120b", 120c", respectively, which are further coupled to one or more tabs 130a', 130b', 130c', and 130a", 130b", 130c", respectively, which are still further coupled to support beams 140', and 140", respectively, at fulcrum points 134a, 134b, and 134c.
Compliant mechanism 100' includes a substantially circular inner stage 110', as opposed to the substantially triangular inner stage 110 of compliant mechanism 100. Further, support beams 140' are configured in a substantially circular pattern about the circular inner stage 110'. Compliant mechanism 100' further differs from compliant mechanism 100 in that it includes a plurality (e.g., three) of support members 150' coupled to support beams 140'. Each of the support members 150' is typically in the form of a disk, having a central hole 152' for fastening the compliant mechanism 100' to another structure. Support members 150' are typically spaced to form a triangular (e.g., a substantially equilateral triangular) pattern in order to provide balanced support and stiffness.
Compliant mechanism 100" is further similar to that of compliant mechanism 100 in that it includes a substantially equilateral triangular inner stage 110 oriented about 180 degrees (i.e., half a turn) out of phase with a substantially triangular configuration of support beams 140". Support beams 140" are typically coupled to three support members 150" at the corners of the triangular configuration thereof. Support members 150" typically include a plurality of holes 152" for fastening the compliant mechanism 100" to another structure. Support members 150" further include constraining compliant mechanisms 156, which allow for a relatively high degree of relative movement between the inner stage 110" and support members 150" (as compared to compliant mechanisms 100 and 100') without requiring plastic deformation of the compliant mechanism's components. The thickness of support beams 140" may optionally be reduced proximate the tabs 130a", 130b", 130c" and support members 150", to reduce their stiffness and thus facilitate their plastic deformation.
The tabs may further optionally include other features, such as but not limited to, holes, slots, grooves, protrusions, detents, and the like for interfacing with an actuator. For example, the protrusions 163 in compliant mechanism 100" provide a feature against which an actuator may press. Rounded features may be used in some applications to ensure that the actuator engages the tab 130a", 130b", 130c" at a predetermined distance from the fulcrum point 134a, 134b, 134c, and is properly oriented to operate in a predetermined direction.
Referring now to Figs. 19a and 19b, the relative movement of inner stage 110 with respect to support member 150 is discussed in more detail. The artisan of ordinary skill will readily recognize that while the following discussion pertains to compliant mechanism 100 in particular, the same general principles apply to compliant mechanisms 100' and 100" regarding movement of inner stages 110' and 110" with respect to support members 150' and 150", respectively. The compliant mechanism paradigm described herein effectively utilizes the concept of offsetting (i.e., moving) at least one of the corners of a triangle (e.g., hinge points 122a, 122b, and 122c) from the axes 132a, 132b, and 132c. This may be thought of, and treated mathematically, as being analogous to offsetting the shaft axis of rotation from the plane of symmetry of the groove of one embodiment of the adjustable kinematic coupling disclosed in the above referenced '562 patent application. The compliant mechanism 100 shown in Figs. 19a and 19b is substantially identical to that of Fig. 16 with the exception that in Fig. 19b, tab 130a has been actuated to the left (i.e., the negative x direction) a distance Δx. The hinge point 122a, originally at position Ml (Fig. 19a), is moved a peφendicular distance x from the axis 132a to position Al (Fig. 19b). The two non- actuated tabs (tabs 130b and 130c) are constrained by their respective support beams 140 to move parallel to axes 132b and 130c, respectively. To maintain geometric congruence after actuation of tab 130a, tabs 130b and 130c displace along axes 132b and 132c, respectively, resulting in the movement of hinge points 122b and 122c from points M2 and M3 to A2 and A3, respectively. The exemplary actuation shown in Figs. 19a and 19b results in a displacement of the inner stage 110, with respect to the support member 150, in the x and θz directions.
Moreover, although in the example shown only element 130a has been actuated, tabs 130b and 130c may also be actuated separately or jointly to affect the relative positions of the inner stage 110 and support member 150 in a controlled and
mathematically predictable manner. For example, as shown in Fig. 20a, actuation of tabs 130b and 130c with an equal magnitude (shown as Δ') in opposite directions about axis 114 results in a displacement of the inner stage 110 with respect to the support member 150 in the y direction. In an alternate example, shown in Fig. 20b, actuation of tabs 130a, 130b, and 130c with an equal magnitude (shown as Δ") in the same direction about axis 114 results in a displacement of the inner stage 110 with respect to the support member 150 in the negative θz direction.
The examples shown in Figs, 19a, 19b, 20a, and 20b illustrate actuation of tabs 130a, 130b, and 130c effecting in-plane (i.e., x, y, and θz) relative movement between the inner stage 110 and support member 150. However, tabs 130a, 130b, and 130c may also be actuated in an out-of-plane (i.e., z) direction. Such actuation serves to effect out of plane (i.e., θx, θy, and z) relative movement between the inner stage 110 and support member 150. For example, actuation of tabs 130a, 130b, and 130c with an equal magnitude in the negative z direction (i.e., into the page in Fig. 20a or 20b) typically results in a displacement of the inner stage 100 with respect to the support member 150 in the positive z direction (i.e., out of the page). However, depending on the geometry and orientation of the tabs, support beams, and hinges, this motion may be in the positive or negative direction. Further, a combination of in plane and out of plane actuation of the tabs 130a, 130b, and 130c enables controlled relative movement in all six degrees of freedom, i.e., in the x, y, z, θx, θy, and θz directions.
One advantage of various embodiments (e.g., compliant mechanisms 100, 100', and 100") of the present invention is that they may be fabricated to include a relatively wide range of predetermined displacement ratios. As set forth hereinabove, the displacement ratio is defined as the ratio of actuator input motion to the relative motion between the inner stage 110 and the support member 150. For example, for applications requiring relatively small-scale (e.g., nanometer and sub-nanometer range) accuracy and precision, a compliant mechanism having a relatively large displacement ratio, enabling relatively small relative movements, may be desirable. A compliant mechanism having a relatively large displacement ratio may be fabricated by increasing the length 138 (Fig. 19a) of the tab relative to the distance between the hinge and flexure points (e.g., the distance between hinge point 122a and fulcrum point 134a in Fig. 17a) and/or the compliance of the tab. Alternatively, for applications in which a wider range of relative
motion is required (e.g., on the order of millimeters or more), a compliant mechanism having a relatively small displacement ratio may be desirable. A compliant mechanism having a relatively small displacement ratio may be fabricated by decreasing the length of the tab 138 relative to the distance between the hinge and flexure points. For typical applications it may be desirable for the compliant mechanisms of this invention to include a displacement ratio in the range from about 0.1 to about 1000. For some particular applications it may be desirable for the compliant mechanisms of this invention to include a displacement ratio in a range of about 2 to about 20.
In alternative embodiments, the inventive compliant mechanism may be used, as stated hereinabove, in the precision alignment of a relatively wide range of product components, fixtures and the like. In some applications, the flexure hinges (e.g., flexure hinges 120a, 120b, and 120c in Fig. 16) may be spaced to form triangles that are not substantially equilateral. This may be beneficial in that it renders motion control in a predetermined direction more or less sensitive to actuator input. Further, the need to use non-equilateral geometry may arise in applications in which the structure of various components does not permit equilateral spacing.
The compliant mechanisms of this invention may be fabricated from substantially any material. Prototypes have been fabricated using metallic materials, such as aluminum alloy 6061, using an abrasive water jet cutting tool. Metallic compliant mechanisms may be advantageous in that they may provide for both elastic and plastic deformation of the mechanism's components. It is further envisioned that the compliant mechanisms of this invention may be fabricated from other materials, such as silicon or doped silicon wafers, using a technique such as deep reactive ion etching (DRIE). DRIE fabrication may be advantageous in that relatively small compliant mechanisms may be formed (e.g., having a characteristic diameter as small or even smaller than 2500 microns).
The artisan of ordinary skill will readily recognize that there are many variable shapes and configurations for the various portions (i.e., the inner stage, the hinge portion, the tabs, the support beams, and the support members) of the compliant mechanism of this invention that may be used to alter the repeatability, resolution, displacement ratio, and position control capabilities of the compliant mechanism.
The modifications to the various aspects of the present invention described hereinabove are merely exemplary. Other variations, modifications, and other
implementations of what is described herein will also occur to those of ordinary skill in the art without departing from the spirit and the scope of the invention as claimed. Accordingly, the invention is to be defined not just by the preceding illustrative description, but instead by the spirit and scope of the following claims.