Beam-based nonlinear spring
09599180 ยท 2017-03-21
Assignee
Inventors
- Jocelyn Maxine Kluger (South Glastonbury, CT, US)
- Alexander Henry Slocum (Bow, NH, US)
- Themistoklis Panagiotis Sapsis (Cambridge, MA, US)
Cpc classification
H02K7/1876
ELECTRICITY
F16F15/035
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
G01B7/16
PHYSICS
F03G1/02
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F03G1/028
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F16F7/104
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F16F1/22
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F03G1/00
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
G01B11/16
PHYSICS
F03G1/10
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
International classification
F16F1/22
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
G01B11/16
PHYSICS
H02K7/18
ELECTRICITY
G01B7/16
PHYSICS
G01L1/04
PHYSICS
F03G1/00
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F03G1/02
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F03G1/10
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
F03G7/08
MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
Abstract
Nonlinear spring. In one embodiment, the spring includes two opposed curved surfaces curving away from one another. A flexible cantilever member is disposed between the two opposed curved surfaces and a mass is attached to a free end of the cantilever member wherein the flexible cantilever member wraps around one of the curved surfaces as the cantilever member deflects to form a nonlinear spring. Energy harvesting devices and a load cell are also disclosed.
Claims
1. A nonlinear load sensitive structure comprising: a. a first rigid structure with at least one curved contact surface; b. a second rigid structure with at least one curved contact surface; and c. at least one curved flexible member disposed between said first and second rigid structures, one end of said at least one curved flexible member attached to said first rigid structure and other end of said at least one curved flexible member attached to said second rigid structure, such that when said first rigid structure is displaced toward said second rigid structure, said at least one curved flexible member bends and makes increasing contact with the contact surfaces of the rigid structures to provide an increasing force as a function of displacement of said rigid structures toward each other.
2. The nonlinear load sensitive structure of claim 1 further comprising an optical or eddy-current sensor configured to measure the displacement between said first and second rigid structures.
3. The nonlinear load sensitive structure of claim 1 further comprising one or more strain gages attached to said at least one of said flexible structures to measure the force required to cause displacement between said first and second rigid structures.
4. A nonlinear load sensitive structure comprising: a. a first rigid structure with at least one concave contact surface and one convex curved contact surface facing each other; b. a second rigid structure with at least one concave contact surface and one convex curved contact surface facing each other; and c. at least one curved flexible member disposed between said first and second rigid structures' contact surfaces, one end of said a least one curved flexible structure attached to said first rigid structure and other end of said at least one curved flexible structure attached to said second rigid structure, such that when said first rigid structure is displaced towards said second rigid structure, said at least one curved flexible member bends and makes increasing contact with the concave or convex contact surfaces of the rigid structures to provide an increasing force as a function of displacement of said rigid structures towards or away from each other, respectively.
5. The nonlinear load sensitive structure of claim 4 further comprising an optical or eddy-current sensor configured to measure the displacement between said first and second rigid structures.
6. The nonlinear load sensitive structure of claim 4 further comprising one or more strain gages attached to said at least one flexible member to measure the force required to cause displacement between first and second rigid structures.
7. The nonlinear load sensitive structure of claim 4 wherein each rigid structure has two each of said concave and convex surfaces and two said flexible members wherein the surfaces and flexible members are approximately circular arcs.
8. The nonlinear load sensitive structure of claim 4 configured as an energy harvesting device where a magnet is attached to said first rigid structure and a wire coil is in proximity to the magnet and attached to said second rigid structure.
9. A nonlinear load cell comprising: a first member further including a first symmetrical member with a left and a right curved surface; a second member displaced parallel to the first member and including a second symmetrical member with a left and a right curved surface, the first and second members having first and second anchor points to receive tensile or compressive forces; and a total of four flexible structures including a left and a right flexible structure projecting out from center regions of both said first and second members, respectively; and left and right connection structures connecting left flexible structure ends to each other and right flexible structure ends to each other, respectively.
10. The nonlinear load cell of claim 6 further comprising a motion sensor between at least one of the connection structure and one of said first or second members.
11. The nonlinear load cell of claim 9 further comprising an optical or electromagnetic linear motion sensor between the first and second members.
Description
BRIEF DESCRIPTION OF THE DRAWING
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
DESCRIPTION OF THE PREFERRED EMBODIMENT
(19)
(20)
(21) The present invention can be made to be any size and out of a large range of materials. Dimension limitations and applied force limitations are related to the stress in the deflected beams 3a and rotational springs 20 (
(22) Other embodiments may have different features. Some of these features may be teeth cut along the curves 2a and 2b, as shown on curve 2n in
(23) Further, the concept of a stiffening member can be extended from a one-dimensional beam 3a wrapping around a one-dimensional curved surface 2a or 2b to a two-dimensional or three-dimensional flexure. For example, the two-dimensional flexure may be a conical coil spring or a plate. The three-dimensional flexure may be a shell, for example. For a two-dimensional flexure, the surfaces 2a and 2b may be two-dimensional shapes where the curvature changes as the radius from the origin increases, for example. For a three-dimensional flexure, the surface may be a three-dimensional sphere or ellipsoid, for example.
(24)
(25)
(26)
(27) The present invention may be used as a spring component in other oscillators and systems as well. Other systems may use any number and configuration of the present invention. In energy applications, the present invention may be used with transducers such as electromagnetic systems, piezoelectric systems, and electrostatic systems among others. Electromagnetic system configurations, for example, may use the masses 4a and 4b as magnets or coils. The piezoelectric system, for example, may use cantilever 3a as the piezoelectric element.
(28) In the embodiment shown in
(29) Surfaces 1e and 1f may have holes 16e-16m cut into the roots of surface curves 2e-2m. Holes 16e-16m may be necessary to satisfy manufacturing practices that may not be able to cut a point at the intersections of 3e with 1e, 3g with 1e, 3f with 1f, and 3h with 1f. When surfaces 1e 375 and 1f have holes 16e-16m, insert 17 may be made to fit into all or some of holes 16e-16m. The presence of insert 17 extends the length of curves 2e-2m.
(30) The top surface 1e be may be connected to the object of interest while the bottom surface may be connected to the tabletop. The displacement measurements of this load cell embodiment could be measured by an optical sensor or eddy-current sensor that compares the displacement of the top surface 1e to the bottom surface 1f. The force acting on the load cell could also be determined by measuring the strain on a strain gage 20a located on the flexible member 3 or 20 (see
(31) For example, if an optical sensor can detect changes as small as 0.1 m, then to achieve 1% accuracy in the force measurement requires a change in displacement per force: dy/dF110.sup.7 m/0.01F. For F=0.01 N [1 gram], it is desirable, then, to have a stiffness of K=dF/dy1000 N/m. For F=1,000 N [100 Kg], it is desirable to have K1e8 N/m.
(32)
(33)
(34)
(35) The concept of stiffening members 3e, 3f, 3g and 3h in load cells 24 and 26, as shown
(36) The present invention can be made to be any size and out of a large range of materials. Dimension limitations and applied force limitations are related to the stress in the deflected beams 3 and rotational springs 20.
(37) Here we briefly summarize the theory for the force, deflection, and stress relationships for designing the vibrating spring and load cells. Designing the spring or load cell maximum stress to remain below a certain value increases its fatigue lifetime. Further details and equation derivations can be found in the journal article J. M. Kluger et al, Robust Energy Harvesting from Walking Vibrations by Means of Nonlinear Cantilever Beams, Journal of Sound and Vibrations (2014).
(38) As shown in
(39)
(40) where D is the gap between the surface end and undeflected cantilever, and n is an arbitrary power greater than 2 (a requirement for essential nonlinearity), z is a spatial coordinate measured from the cantilever/surface root, and L.sub.Surf is the surface length. The theory derived below should apply to any surface with a monotonically increasing curvature,
(41)
When a sufficiently larger force F is applied to the beam tip, the cantilever begins to wrap around the surface. The contact point z.sub.c is the axial coordinate where the cantilever stops wrapping around the surface and becomes a free beam of length L.sub.Free. To the left of the contact point, we assume that the beam is tangent to (equal to) the surface shape given by eq. (1). For the free beam segment to the right of the contact point, the boundary conditions on the beam are
(42)
(43) where w is the beam deflection along its free length, F is the force applied to the mass, L.sub.Free is the cantilever segment to the right of the contact point z.sub.c, EI is the cantilever rigidity, S is the surface shape defined in eq. (1), z.sub.c is the contact point between the cantilever and surface for the given force, and x is the spatial coordinate with its origin at z.sub.c. Based on Euler-Bernoulli beam theory and solving
(44)
the deflection along the free beam length, x, is
(45)
(46) Substituting x=L.sub.Free into eq. (3), the beam tip deflection due to the force F is
(47)
(48) We can slightly modify eq. (4) to describe the deflection of the end-mass center .sub.Mass by accounting for the beam tip angle:
(49)
(50) where L.sub.Mass is the length of the undeflected end mass in the z direction. In eq.s (4) and (5), we assume that L.sub.Mass is small and causes a negligible moment on the beam tip. Eq.s (4) and (5) and the following equations may straightforwardly be modified for larger L.sub.Mass and other beam loading conditions.
(51) The location of the contact point z.sub.c along the surface is the point at which the cantilever curvature equals the surface curvature (surface contact condition):
(52)
(53) This is the case because the free cantilever curvature decreases along its length (cantilever gets flatter), while the surface curvature is constant (n=2) or increases (n>2) along its length (surface gets rounder). z.sub.c is the point where the surface would no longer prevent the natural curvature of the free cantilever. Alternatively, at z.sub.c, the curvature at the root of a free cantilever of length L.sub.Free subject to tip force F equals the surface curvature to which it is tangent. The boundary condition defined by Eq. (6) is required for static equilibrium because no external moment is applied to the beam at the contact point.
(54) The free beam length is the full beam length minus the beam length in contact with the surface. Assuming a slender beam, the beam length in contact with the surface is approximately equal to the surface arc length from z=0 to z.sub.c. For small deflections, one can assume that L.sub.Free=L.sub.Cantz.sub.c.
(55) Further using the slender Euler-Bernoulli beam theory, the maximum stress magnitude, , in the beam cross-section is
(56)
(57) where E is the beam elastic modulus,
(58)
is half the beam height and
(59)
is the beam curvature. For the beam segment in contact with the surface,
(60)
can be found by using w(z)=s(z) and differentiating eq. (1). For the free beam segment,
(61)
can be found by differentiating eq. (3).
(62) As shown in
(63) Below, we describe the relationship of the load cell's applied force F, contact point x.sub.c, tip moment M.sub.Tip, and deflection . When F is applied to the load cell and the load cell deflects by , the complete load cell experiences the applied force 2F and deflection 2.
(64) The surface shape follows the curve
(65)
(66) where D is the gap between the surface end and undeflected cantilever, and n is an arbitrary power greater than or equal to 2, x is a spatial coordinate measured from the cantilever/surface root, and L.sub.Surf is the surface length. The theory derived below should apply to any surface with a monotonically increasing curvature,
(67)
(68) Referring to
M.sub.Internal,Cant=F(Lx)+M.sub.Tip,(9)
(69) where F is the applied force on the load cell, L is the length of the full cantilever, and M.sub.Tip is the moment applied at the junction of the cantilever and ring. The internal moment along the ring is
M.sub.Internal,ring=FR(1+sin )+M.sub.Tip,(10)
(70) where R is the radius of the ring and is the angle along the ring.
(71) Using Euler-Bernoulli beam theory and equating the angle of the cantilever tip to the angle of the ring at =3/2, the value of M.sub.Tip as a function of the applied force F and contact point x.sub.c is
(72)
(73) where L.sub.FreeLx.sub.c (using small beam deflection approximation) is the free cantilever segment length,
(74)
is the slope of the surface at the contact point found by differentiating eq. (8), and EI is the cantilever and ring rigidity.
(75) At the contact point, the cantilever curvature must be continuous because there is not an applied external moment. To the left of the contact point, we assume that the cantilever segment in contact with the surface is tangent to the surface. Then, the contact point relates to the applied force F and tip moment M.sub.Tip by
(76)
(77) Eq.s (11) and (12) can be simultaneously solved to relate the applied force F, tip moment M.sub.Tip, and contact point x.sub.c.
(78) Having determined F, M.sub.Tip, and x.sub.c, the deflection of the load cell indicated in
=.sub.1+.sub.2+.sub.3+.sub.4.(13)
(79) The first component is the cantilever deflection at the contact point x.sub.c. This deflection component is the vertical location of the surface curve at x.sub.c:
.sub.1=S(x.sub.c)(14)
(80) The second component is the deflection of the free cantilever segment due to the cantilever's slope at the contact point. Since the beam is tangent to the surface at the contact point, its slope equals the surface slope. The free length of the beam rotates by this slope (i.e. small angle) about the contact point, which results in the deflection:
(81)
(82) where L.sub.Free=Lx.sub.c is the length of the free cantilever segment, assuming small deflection and small surface curves, S(x.sub.c). The third deflection component is due to the free cantilever segment bending. Using Euler-Bernoulli beam theory, integrating the moment-curvature relation given by eq. (9), and using boundary conditions that the deflection and slope due to bending equal zero at the free cantilever segment root (the contact point, x.sub.c), this deflection component is:
(83)
(84) The fourth deflection component is due to the ring bending. When an infinitesimal segment of the ring, l=Rd, bends, it rotates the segments of the ring on either side of it by an angle =l with respect to each other, where is the change in the curvature of the beam at the infinitesimal segment due to bending
(85)
Based on geometry and the small angle approximation, the vertical tip deflection due to this change in angle is the horizontal distance between the infinitesimal segment and the tip, X=R(1+sin ), multiplied by the change in angle, . Integrating this infinitesimal deflection along the curved beam results in the total deflection of the curved beam due to bending:
(86)
(87) The load cell stiffness is K=dF/d. The entire load cell stiffness is also K=dF/d.
(88) Again using Euler-Bernoulli beam theory, the normal stress in the cantilever segment in contact with the surface can be found by
(89)
(90) where E is the cantilever elastic modulus, h is the beam height, and
(91)
(assuming the cantilever segment in contact with the surface is tangent to the surface). The normal stress in the free cantilever segment and ring can be found by
(92)
(93) where M.sub.Internal is the internal moment in the free cantilever segment (eq. (9)) or in the ring (eq. (10)), his the cross-sectional height, and I is the cross-sectional moment of inertia.
(94) As shown in
(95)
(96) where we assume that the ring arc length equals R.sub.0.sub.R. The flexible curved beams have mean radii R.sub.0. The curved beams have cross-sectional height h which may vary along the angle . We may choose to keep h constant along .sub.R or vary the height along .sub.R according to
(97)
(98) where q is an arbitrary power. Eq. (21) is valid from 0.sub.R/2 and then symmetrical in the other load cell quadrants. The outer rigid surface has shape S.sub.out(.sub.s) and the inner rigid surface has shape S.sub.In(.sub.s, In), defined in polar coordinates. The outer and inner surfaces have curvatures .sub.S and .sub.S,In, respectively. The rigid surfaces have monotonically increasing curvatures,
(99)
(100) The load cell may deflect up to a total distance .sub.max, after which overstops prevent further deflection from additional force. The load cell may be fabricated with gaps if the fabrication technique cannot allow the curved beam root to meet the rigid surface at a point.
(101) Below, we derive the theory for the bottom right quadrant of the circular load cell in compression, shown in
(102) The equation for the internal moment in the z direction along the curved beam as a function of the angle with respect to the vertical, .sub.R, is
(103)
(104) where M.sub.D is the moment in the z-direction acting at the top of the load cell.
(105) Next, we determine the relationship of the applied force P/2, moment M.sub.D, and contact point .sub.Rc by simultaneously solving two equations. First, the rotation of the ring at point D with respect to point B must be 0 due to symmetry. That is,
(106)
(107) .sub.c is the change in angle of the ring at the contact point:
(108)
(109) where .sub.Rc is the angle of the contact point on the undeflected ring (before the force is applied) and
(110)
is the angle of the surface at the contact point, .sub.Sc (one way to find the surface angle with respect to the horizontal,
(111)
is to convert S(.sub.Sc) from polar to Cartesian coordinates).
(112) For the ring, the internal energy from .sub.R=0 to the [unknown] contact point .sub.R=.sub.Rc, depends on how much the ring curvature changes to match the surface curvature to which it is tangent. This internal energy due to bending is
(113)
(114) where E is the curved beam elastic modulus, and I is the cross section moment of inertia (which may be a variable function along .sub.Rc, i.e. if the cross-section height is defined by eq. (21)). The surface curvature .sub.S(.sub.S) can be converted to a function of .sub.R using eq. (20).
(115) From the [unknown] contact point .sub.R=.sub.Rc to .sub.R=/2, the internal energy depends on the internal moment in the ring that causes bending. This internal energy component is
(116)
(117) where the internal moment M is defined in eq. (22). The total internal energy in the ring is
U=U.sub.1+U.sub.2.(27)
(118) Next, we minimize the internal energy U with respect to the contact point .sub.R. That is, we solve
(119)
(120) To find the relationship of the applied force P/2, moment M.sub.D, and contact point .sub.Rc, we may simultaneously solve eq.s (23) and (28) for a fixed force and geometric parameters.
(121) To find the deflection of the load cell, we rearrange Castiglaino's first theorem into
(122)
(123) where the internal energy is a function of the dummy variable, the applied force, F.
(124) Again, the stiffness of the load cell is K=dP/d.
(125) Finally, the equations for stress in the load cell are similar to those of the vibrating spring and straight beam load cell. For the curved beam segment in contact with the rigid surface, the normal stress is
(126)
(127) where =R.sub.0.sub.S(.sub.S) is the required change in the beam curvature for it to be tangent to the surface. The normal stress in the free segment of the curved beam is
(128)
(129) where M is a function of .sub.R defined in eq. (22) and h may be the function of .sub.R defined in eq. (21)
(130) A fuller mathematical analysis underpinning the present invention may be found in the provisional application referred to earlier and in J. M. Kluger et al, Robust Energy Harvesting from Walking Vibrations by Means of Nonlinear Cantilever Beams, Journal of Sound and Vibrations (2014). The contents of this reference is incorporated herein by reference in its entirety. The numbers in square brackets refer to the references listed herein.
REFERENCES
(131) 1. P. Mitcheson, T. Green, E. Yeatman, A. Holmes, Architectures for vibration-driven micropower generators, J. Microelectromechanical Systems, 13 (2004) (3) pp. 429-440. 2. A. Hajati, S. Bathurst, H. Lee, S. Kim, Design and fabrication of a nonlinear resonator for ultra wide-bandwidth energy harvesting applications, in: Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMs), 2011, pp. 130-1304. 3. O. Yaglioglu, Modeling and design considerations for a micro-hydraulic piezoelectric power generator, Master's thesis, Massachusetts Institute of Technology (2002). 4. O. Gendelman, T. Sapsis, A. Vakakis, L. Bergman, Enhanced passive targeted energy transfer in strongly nonlinear mechanical oscillators, J. of Sound and Vibration 330 (2011) pp. 1-8. 5. D. Quinn, O. Gendelman, G. Kerschen, T. Sapsis, L. Bergman, A. Vakakis, Efficiency of targeted energy transfers incoupled nonlinear oscillators associated with 1:1 resonance captures: Part i, J. Sound and Vibration (2008) pp. 1228-1248. 6. A. Vakakis, O. Gendelman, L. Bergman, D. McFarland, G. Kerschen, Y. S. Lee, Nonlinear Targeted Energy Transfer in Mechanical and Structural Systems, Springer, 2009. 7. D. McFarland, L. Bergman, A. Vakakis, Experimental study of non-linear energy pumping occurring at a single fast frequency, International Journal of Non-Linear Mechanics 40 (2005) pp. 891-899. 8. X. Tang, L. Zuo, Simulation and experiment validation of simultaneous vibration control and energy harvesting from buildings using tuned mass dampers, in: Proceedings of the American Control Conference, 2011, pp. 3134-3139. 9. F. Cottone, H. Vocca, L. Gammaitoni, Nonlinear energy harvesting, Physical Review Letters (2009) 102 (8) pp. 080601. 10. J. Paradiso, T. Starner, Energy scavenging for mobile and wireless electronics, IEEE 685 Pervasive Computing 4 (1) (2005) pp. 18-27. 11. S. Stanton, C. McGehee, B. Mann, Nonlinear dynamics for broadband energy harvesting: Investigation of a bistable piezoelectric inertial generator, Physica D 239 (2010) pp. 640-653. 12. L. Manevitch, A. Musienko, C. Lamarque, New analytical approach to energy pumping problem in strongly nonhomogeneous 2d of systems, Meccanica (2007) 77-83. 13. J. Sanders, R. Miller, D. Berglund, S. Zachariah, A modular six-directional force sensor for prosthetic assessment: A technical note, J. Rehabilitation Research and Development, (1997) 34 (2) pp. 195-202. 14. J. Mokhbery, Advances in load cell technology for medical applications, Medical Device and Diagnostic Industry newsletter. Accessed online September, 2014. 15. Y.-S. Change, T.-C. Lin, An optimal g-shaped load cell for two-range loading, Engineering in Agriculture, Environment, and Food, (2013) 6(4) pp. 172-176. 16. G. Mantriota, A. Messina, Theoretical and experimental study of the performance of at suction cups in the presence of tangential loads, Mechanism and machine theory, (2011) 46 (5) pp. 607-611. 17. A. Cordero, G. Carbone, M. Ceccarelli, J. Echavarri, J. Munoz, Experimental tests in human-robot collision evaluation and characterization of a new safety index for robot operation, Mechanism and machine theory, (2014) 80, pp. 185-199. 18. J. Smith, Electronic Scale Basics, Key Markets Publishing, 1991. 19. O. E. Limited, An introduction to load cells, history, theory & operating principles (2014). URL http://www.omega.co.uk/prodinfo/load-cells.html 20. Acuity, Principles of measurement used by laser sensors (2014). URL http://www.acuitylaser.com/support/measurement-principles. 21. Lion, Understanding sensor resolution specifications and effects on performance (2014). URL http://www.lionprecision.com/tech-library/technotes/article-0010-sensor-resolution.html 22. Microtrack 3, Tech. rep., MTI Instruments (2014). 23. A. Storace, P. Sette, Leaf spring weighing scale, U.S. Pat. No. 4,037,675. 24. S. Suzuki, Y. Nishiyama, T. Kitagawa, Multi-range load cell weighing scale, U.S. Pat. No. 4,711,314.