Isotropic harmonic oscillator and associated time base without escapement or with simplified escapement
10365609 · 2019-07-30
Assignee
Inventors
Cpc classification
G04B15/14
PHYSICS
International classification
G04B15/14
PHYSICS
G04B23/00
PHYSICS
Abstract
A mechanical isotropic harmonic oscillator including a two translational degrees of freedom linkage supporting an orbiting mass with respect to a fixed base with springs having isotropic and linear restoring force properties.
Claims
1. A mechanical isotropic harmonic oscillator comprising: a fixed base; an intermediate block; a mass configured to oscillate; a first parallel spring stage connected between the mass and the intermediate block; and a second parallel spring stage connected between the intermediate block and the fixed base, wherein a direction of flexure of the first parallel stage is substantially perpendicular to a direction of flexure of the second parallel spring stage.
2. The oscillator as claimed in claim 1, wherein the first and the second parallel spring stage lie in a same plane.
3. A mechanical isotropic harmonic oscillator comprising: a fixed base; an intermediate block; a mass configured to oscillate; a first flexure means connected between the mass and the intermediate block; and a second flexure means connected between the intermediate block and the fixed base, wherein a direction of flexure of the first flexure means is substantially perpendicular to a direction of flexure of the second flexure means.
4. The oscillator as claimed in claim 3, wherein the first and the second flexure means lie in a same plane.
5. The oscillator as claimed in claim 1, wherein the first parallel spring stage includes a planar spring stage having two parallelly-arranged leaf springs.
6. The oscillator as claimed in claim 1, wherein the second parallel spring stage includes a planar spring stage having two parallelly-arranged leaf springs.
7. The oscillator as claimed in claim 1, wherein the first and the second parallel spring stage and the oscillating mass together form a two translational degree of freedom isotropic harmonic oscillator.
8. The oscillator as claimed in claim 1, wherein each one of the first and the second parallel spring stage form a one translational degree of freedom isotropic harmonic oscillator.
9. The oscillator as claimed in claim 1, further comprising: a rigid pin attached to the oscillating mass, configured to engage with a slot acting as a driving crank to maintain oscillation of the oscillating mass.
10. The oscillator as claimed in claim 1, further comprising: a second intermediate block; a second mass configured to oscillate; a third parallel spring stage connected between the second mass and the second intermediate block; and a fourth parallel spring stage connected between the second intermediate block and the fixed base, wherein a direction of flexure of the third parallel spring stage is substantially perpendicular to a direction of flexure of the fourth parallel spring stage, and wherein the direction of flexure of the third parallel spring stage is substantially perpendicular to the direction of flexure of the first parallel spring stage.
11. The oscillator as claimed in claim 10, wherein the first mass and the second mass are connected together.
12. A wristwatch including the oscillator as defined in claim 1.
13. The oscillator as claimed in claim 3, wherein the first flexure means includes two parallelly-arranged leaf springs.
14. The oscillator as claimed in claim 3, wherein the second flexure means includes two parallelly-arranged leaf springs.
15. The oscillator as claimed in claim 3, wherein the first and the second flexure means and the mass together form a two translational degree of freedom isotropic harmonic oscillator.
16. The oscillator as claimed in claim 3, wherein each one of the first and the second flexure means form a one translational degree of freedom isotropic harmonic oscillator.
17. The oscillator as claimed in claim 3, further comprising: a rigid pin attached to the mass, configured to engage with a slot acting as a driving crank to maintain oscillation of the mass.
18. The oscillator as claimed in claim 3, further comprising: a second intermediate block; a second mass configured to oscillate; a third flexure means connected between the second mass and the second intermediate block; and a fourth flexure means connected between the second intermediate block and the fixed base, wherein a direction of flexure of the third flexure means is substantially perpendicular to a direction of flexure of the fourth flexure means, and wherein the direction of flexure of the third flexure means is substantially perpendicular to the direction of flexure of the first flexure means.
19. The oscillator as claimed in claim 18, wherein the first mass and the second mass are connected together.
20. A wristwatch including the oscillator as defined in claim 3.
Description
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
(1) The present invention will be better understood from the following description and from the drawings which show
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
(38)
(39)
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)
(60)
(61)
(62)
(63)
DETAILED DESCRIPTION OF THE SEVERAL EMBODIMENTS
2 Conceptual Basis of the Invention
(64) 2.1 Newton's Isochronous Solar System
(65) As is well-known, in 1687 Isaac Newton published Principia Mathematica in which he proved Kepler's laws of planetary motion, in particular, the First Law which states that planets move in ellipses with the Sun at one focus and the Third Law which states that the square of the orbital period of a planet is proportional to the cube of the semi-major axis of its orbit, see reference [19].
(66) Less well-known is that in Book I, Proposition X, of the same work, he showed that if the inverse square law of attraction (see
(67) Newton's result for Hooke's Law is very easily verified: Consider a point mass moving in two dimensions subject to a central force
F(r)=kr
centered at the origin, where r is the position of the mass, then for an object of mass m, this has solution
(A.sub.1 sin(.sub.0t+.sub.1),A.sub.2 sin(.sub.0t+.sub.2)),
for constants A.sub.1, A.sub.2, .sub.1, .sub.2 depending on initial conditions and frequency
(68)
(69) This not only shows that orbits are elliptical, but that the period of motion depends only on the mass m and the rigidity k of the central force. This model therefore displays isochronism since the period
(70)
is independent of the position and momentum of the point mass (the analogue of Kepler's Third Law proved by Newton).
2.2 Implementation as a Time Base for a Timekeeper
(71) Isochronism means that this oscillator is a good candidate to be a time base for a timekeeper as a possible embodiment of the present invention.
(72) This has not been previously done or mentioned in the literature and the utilization of this oscillator as a time base is an embodiment of the present invention. This oscillator is also known as a harmonic isotropic oscillator where the term isotropic means same in all directions.
(73) Despite being known since 1687 and its theoretical simplicity, it would seem that the isotropic harmonic oscillator, or simply isotropic oscillator,, has never been previously used as a time base for a watch or clock, and this requires explanation.
(74) It would seem that the main reason is the fixation on constant speed mechanisms such as governors or speed regulators, and a limited view of the conical pendulum as a constant speed mechanism.
(75) For example, in his description of the conical pendulum which has the potential to approximate isochronism, Leopold Defossez states its application to measuring very small intervals of time, much smaller than its period, see reference [8, p. 534].
(76) H. Bouasse devotes a chapter of his book to the conical pendulum including its approximate isochronism, see reference [3, Chapitre VIII]. He devotes a section of this chapter on the utilization of the conical pendulum to measure fractions of seconds (he assumes a period of 2 seconds), stating that this method appears perfect. He then qualifies this by noting the difference between average precision and instantaneous precision and admits that the conical pendulum's rotation may not be constant over small intervals due to difficulties in adjusting the mechanism. Therefore, he considers variations within a period as defects of the conical pendulum which implies that he considers that it should, under perfect conditions, operate at constant speed.
(77) Similarly, in his discussion of continuous versus intermittent motion, Rupert Gould overlooks the isotropic oscillator and his only reference to a continuous motion timekeeper is the Villarceau regulator which he states: seems to have given good results. But it is not probable that was more accurate than an ordinary good-quality driving clock or chronograph, see reference [9, 20-21]. Gould's conclusion is validated by the Villarceau regulator data given by Breguet, see reference [4].
(78) From the theoretical standpoint, there is the very influential paper of James Clerk Maxwell On Governors, which is considered one of the inspirations for modern control theory, see reference [18].
(79) Moreover, isochronism requires a true oscillator which must preserve all speed variations. The reason is that the wave equation
(80)
preserves all initial conditions by propagating them. Thus, a true oscillator must keep a record of all its speed perturbation. For this reason, the invention described here allows maximum amplitude variation to the oscillator.
(81) This is exactly the opposite of a governor which must attenuate these perturbations. In principle, one could obtain isotropic oscillators by eliminating the damping mechanisms leading to speed regulation.
(82) The conclusion is that the isotropic oscillator has not been used as a time base because there seems to have been a conceptual block assimilating isotropic oscillators with governors, overlooking the simple remark that accurate timekeeping only requires a constant time over a single complete period and not over all smaller intervals.
(83) We maintain that this oscillator is completely different in theory and function from the conical pendulum and governors, see hereunder in the present description.
(84)
(85)
(86) 2.3 Rotational Versus Translational Orbiting Motion
(87) Two types of isotropic harmonic oscillators having unidirectional motion are possible. One is to take a linear spring with body at its extremity, and rotate the spring and body around a fixed center. This is illustrated in
(88) This leads to the body rotating around its center of mass with one full turn per revolution around the orbit as illustrated in
(89) This type of spring will be called a rotational isotropic oscillator and will be described in Section 4.1. In this case, the moment of inertia of the body affects the dynamics, as the body is rotating around itself.
(90) Another possible realization has the mass supported by a central isotropic spring, as described in Section 4.2. In this case, this leads to the body having no rotation around its center of mass, and we call this orbiting by translation. This is illustrated in
(91) In this case, the moment of inertia of the mass does not affect the dynamics.
(92) 2.4 Integration of the Isotropic Harmonic Oscillator in a Standard Mechanical Movement
(93) Our time base using an isotropic oscillator will regulate a mechanical timekeeper, and this can be implemented by simply replacing the balance wheel and spiral spring oscillator with the isotropic oscillator and the escapement with a crank fixed to the last wheel of the gear train. This is illustrated in
3 Theoretical Requirements of the Physical Realization
(94) In order to realize an isotropic harmonic oscillator, in accordance with the present invention, there requires a physical construction of the central restoring force. One first notes that the theory of a mass moving with respect to a central restoring force is such that the resulting motion lies in a plane. It follows that for practical reasons, the physical construction should realize planar isotropy. Therefore, the constructions and embodiments described here will mostly be of planar isotropy, but not limited to this embodiment, and there will also be an example of 3-dimensional isotropy.
(95) In order for the physical realization to produce isochronous orbits for a time base, the theoretical model of Section 2 above must be adhered to as closely as possible. The spring stiffness k is independent of direction and is a constant, that is, independent of radial displacement (linear spring). In theory, there is a point mass, which therefore has moment of inertia J=0 when not rotating. The reduced mass m is isotropic and also independent of displacement. The resulting mechanism should be insensitive to gravity and to linear and angular shocks. The conditions are therefore
(96) Isotropic k. Spring stiffness k isotropic (independent of direction).
(97) Radial k. Spring stiffness k independent of radial displacement (linear spring).
(98) Zero J. Mass m with moment of inertia J=0.
(99) Isotropic m. Reduced mass m isotropic (independent of direction).
(100) Radial m. Reduced mass m independent of radial displacement.
(101) Gravity. Insensitive to gravity.
(102) Linear shock. Insensitive to linear shock.
(103) Angular shock. Insensitive to angular shock.
4 Realization of the Isotropic Harmonic Oscillator
(104) Planar isotropy may be realized in two ways.
(105) 4.1 Rotating Springs Leading to a Rotational Isotropic Oscillator
(106) A.1. A rotating turntable 1 on which is fixed a spring 2 of rigidity k with the spring's neutral point at the center of rotation of the turntable, is illustrated in
(107) Note that gravity does not affect the spring when it is in the axial direction. However, these realizations have the disadvantage of having the spring and its support both rotating around their own axes, which introduces spurious moment of inertia terms which reduce the theoretical isochronism of the model. Indeed, considering the point mass of mass m and then including a isotropic support of moment of inertia I and constant total angular momentum L, then if friction is ignored, the equations of motion reduce to
(108)
(109) This equation can be solved explicitly in terms of Jacobi elliptic functions and the period expressed in terms of elliptic integrals of the first kind, see reference [17] for definitions and similar applications to mechanics. A numerical analysis of these solutions shows that the divergence from isochronism is significant unless the moment of inertia I is minimized.
(110) We now list which of the theoretical properties of Section 3 hold for these realizations. In particular, for the rotating cantilever spring.
(111) TABLE-US-00001 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes No Ye Ye One direction No No
4.2 Isotropic Springs with Orbits by Translation.
(112) The realizations which appear to be most suitable to preserve the theoretical characteristics of the harmonic oscillator are the ones in which the central force is realized by an isotropic spring, where the term isotropic is again used to mean same in all directions.
(113) A simple example is given in
(114) One can now show that this mechanism exhibits isotropy to first order, as illustrated in
F(dr)=(kdx,kdy)=kdr
and the central linear restoring force of Section 2 is verified. It follows that this mechanism is, up to first order, a realization of a central linear restoring force, as claimed.
(115) In these realizations, gravity affects the springs 11, 12 in all directions as it changes the effective spring constant. However, the springs 11, 12 does not rotate around its own axis, minimizing spurious moments of inertia, and the central force is directly realized by the spring itself. We now list which of the theoretical properties of Section 3 hold for these realizations (up to first order).
(116) TABLE-US-00002 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
(117) Many planar springs have been proposed and if some may be implicitly isotropic, none has been explicitly declared to be isotropic. In the literature, Simon Henein [see reference 14, p. 166, 168] has proposed two mechanisms which exhibit planar isotropy. But these examples, as well as the one just described above, do not exhibit sufficient isotropy to produce an accurate timebase for a timekeeper, as a possible embodiment of the invention described herein.
(118) An embodiment illustrated in
(119) Therefore, more precise isotropic springs have been developed. In particular, the precision has been greatly improved and this is the subject of several embodiments described in the present application.
(120) In these realizations, the spring does not rotate around its own axis, minimizing spurious moments of inertia, and the central force is directly realized by the spring itself. These have been named isotropic springs because their restoring force is the same in all directions.
(121) A basic example of an embodiment of the oscillator made of planar isotropic springs according to the present invention is illustrated in
5 Compensation Mechanisms
(122) In order to place the new oscillator in a portable timekeeper as an exemplary embodiment of the present invention, it is necessary to address forces that could influence the correct functioning of the oscillator. These include gravity and shocks.
(123) 5.1 Compensation for Gravity
(124) The first method to address the force of gravity is to make a planar isotropic spring which when in horizontal position with respect to gravity does not feel its effect.
(125)
(126) However, this is adequate only for a stationary clock/watch. For a portable timekeeper, compensation is required. This can be achieved by making a copy of the oscillator and connecting both copies through a ball or universal joint as in
(127) 5.2 Dynamical Balancing for Linear Acceleration
(128) Linear shocks are a form of linear acceleration, so include gravity as a special case. Thus, the mechanism of
(129) 5.3 Dynamical Balancing for Angular Acceleration
(130) Effects due to angular accelerations can be minimized by reducing the distance between the centers of gravity of the two masses as shown in
(131) Specifically,
(132)
(133)
(134)
6 Maintaining and Counting
(135) Oscillators lose energy due to friction, so there needs a method to maintain oscillator energy. There must also be a method for counting oscillations in order to display the time kept by the oscillator. In mechanical clocks and watches, this has been achieved by the escapement which is the interface between the oscillator and the rest of the timekeeper. The principle of an escapement is illustrated in
(136) In the case of the present invention, two main methods are proposed to achieve this: without an escapement and with a simplified escapement.
(137) 6.1 Mechanisms without Escapement
(138) In order to maintain energy to the isotropic harmonic oscillator, a torque or a force are applied, see
(139)
(140)
(141)
(142)
(143)
(144)
(145) Bracket 139 mounted on the orbiting mass holds the rigid pin 138 (illustrated in
(146) Each stage 131-134 may be for example made as illustrated in
(147)
(148)
(149)
(150)
(151)
(152) Typically, each stage 131-134 may be made in accordance with the embodiments described later in the present specification in reference to
(153) To construct the oscillator of
(154) Stages 133 and 134 are attached as stages 131-132 and placed in a mirror configuration over stages 131-132, stage 133 comprising as stages 131 and 132 springs 133a-133d and a mass 133e. The position of stage 133 rotated by 90 with respect to stage 132 as one can see in
(155) Then, as illustrated in
(156) As illustrated in
(157) Of course, the stages 131-134 of
(158) 6.2 Generalized Coordinate Isotropic Harmonic Oscillators
(159) The XY isotropic harmonic oscillators of the previous section can be generalized by replacing X translation and Y translation by other motions, in particular, rotation. When expressed as generalized coordinates in Lagrangian mechanics, the theory is identical and the mechanisms will have the same isotropic harmonic properties as the translational XY mechanisms.
(160)
(161) TABLE-US-00003 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes Yes Yes No
(162)
(163)
(164)
(165) TABLE-US-00004 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
6.3 Simplified Escapements
(166) The advantage of using an escapement is that the oscillator will not be continuously in contact with the energy source (via the gear train) which can be a source of chronometric error. The escapements will therefore be free escapements in which the oscillator is left to vibrate without disturbance from the escapement for a significant portion of its oscillation.
(167) The escapements are simplified compared to balance wheel escapements since the oscillator is turning in a single direction. Since a balance wheel has a back and forth motion, watch escapements generally require a lever in order to impulse in one of the two directions.
(168) The first watch escapement which directly applies to our oscillator is the chronometer or detent escapement [6, 224-233]. This escapement can be applied in either spring detent or pivoted detent form without any modification other than eliminating passing spring whose function occurs during the opposite rotation of the ordinary watch balance wheel, see [6, FIG. 471c]. For example, in
(169) H. Bouasse describes a detent escapement for the conical pendulum [3, 247-248] with similarities to the one presented here. However, Bouasse considers that it is a mistake to apply intermittent impulse to the conical pendulum. This could be related to his assumption that the conical pendulum should always operate at constant speed, as explained above.
(170) 6.4 Improvement of the Detent Escapement for the Isotropic Harmonic Oscillator
(171) Embodiments of possible detent escapements for the isotropic harmonic oscillator are shown in
(172)
(173)
(174)
(175)
7 Difference with Previous Mechanisms
(176) 7.1 Difference with the Conical Pendulum
(177) The conical pendulum is a pendulum rotating around a vertical axis, that is, perpendicular to the force of gravity, see
(178) However, as with cycloidal cheeks for the ordinary pendulum, Huygens' modification is based on a flexible pendulum and in practice does not improve timekeeping. The conical pendulum has never been used as a timebase for a precision clock.
(179) Despite its potential for accurate timekeeping, the conical pendulum has been consistently described as a method for obtaining uniform motion in order to measure small time intervals accurately, for example, by Defossez in his description of the conical pendulum see reference [8, p. 534].
(180) Theoretical analysis of the conical pendulum has been given by Haag see reference [11] [12, p. 199-201] with the conclusion that its potential as a timebase is intrinsically worse than the circular pendulum due to its inherent lack of isochronism.
(181) The conical pendulum has been used in precision clocks, but never as a time base. In particular, in the 1860's, William Bond constructed a precision clock having a conical pendulum, but this was part of the escapement, the timebase being a circular pendulum see references [10] and [25, p. 139-143].
(182) Our invention is therefore a superior to the conical pendulum as choice of time base because our oscillator has inherent isochronism. Moreover, our invention can be used in a watch or other portable timekeeper, as it is based on a spring, whereas this is impossible for the conical pendulum which depends on the timekeeper having constant orientation with respect to gravity.
(183) 7.2 Difference with Governors
(184) Governors are mechanisms which maintain a constant speed, the simplest example being the Watt governor for the steam engine. In the 19th Century, these governors were used in applications where smooth operation, that is, without the stop and go intermittent motion of a clock mechanism based on an oscillator with escapement, was more important than high precision. In particular, such mechanisms were required for telescopes in order to follow the motion of the celestial sphere and track the motion of stars over relatively short intervals of time. High chronometric precision was not required in these cases due to the short time interval of use.
(185) An example of such a mechanism was built by Antoine Breguet, see reference [4], to regulate the Paris Observatory telescope and the theory was described by Yvon Villarceau, see reference [24], it is based on a Watt governor and is also intended to maintain a relatively constant speed, so despite being called a regulateur isochrone (isochronous governor), it cannot be a true isochronous oscillator as described above. According to Breguet, the precision was between 30 seconds/day and 60 seconds/day, see reference [4].
(186) Due to the intrinsic properties of harmonic oscillators following from the wave equation, see Section 8, constant speed mechanisms are not true oscillators and all such mechanisms have intrinsically limited chronometric precision.
(187) Governors have been used in precision clocks, but never as the time base. In particular, in 1869 William Thomson, Lord Kelvin, designed and built an astronomical clock whose escapement mechanism was based on a governor, though the time base was a pendulum, see references [23] [21, p. 133-136] [25, p. 144-149]. Indeed, the title of his communication regarding the clock states that it features uniform motion, see reference [23], so is clearly distinct in its purpose from the present invention.
(188) 7.3 Difference with Other Continuous Motion Timekeepers
(189) There have been at least two continuous motion wristwatches in which the mechanism does not have intermittent stop & go motion so does not suffer from needless repeated accelerations. The two examples are the so-called Salto watch by Asulab, see reference [2], and Spring Drive by Seiko, see reference [22]. While both these mechanism attain a high level of chronometric precision, they are completely different from the present invention as they do not use an isotropic oscillator as a time base and instead rely on the oscillations of a quartz tuning fork. Moreover, this tuning fork requires piezoelectricity to maintain and count oscillations and an integrated circuit to control maintenance and counting. The continuous motion of the movement is only possible due to electromagnetic braking which is once again controlled by the integrated circuit which also requires a buffer of up to 12 seconds in its memory in order to correct chronometric errors due to shock.
(190) Our invention uses a mechanical oscillator as time base and does not require electricity or electronics in order to operate correctly. The continuous motion of the movement is regulated by the isotropic oscillator itself and not by an integrated circuit.
8 Realization of an Isotropic Harmonic Oscillator
(191) In some embodiments some already discussed above and detailed hereunder, the present invention was conceived as a realization of the isotropic harmonic oscillator for use as a time base. Indeed, in order to realize the isotropic harmonic oscillator as a time base, there requires a physical construction of the central restoring force. One first notes that the theory of a mass moving with respect to a central restoring force is such that the resulting motion lies in a plane. It follows that for practical reasons, that the physical construction should realize planar isotropy. Therefore, the constructions described here will mostly be of planar isotropy, but not limited to this, and there will also be an example of 3-dimensional isotropy. Planar isotropy can be realized in two ways: isotropic springs and translational isotropic springs.
(192) Isotropic springs have one degree of freedom and rotate with the support holding both the spring and the mass. This architecture leads naturally to isotropy. While the mass follows the orbit, it rotates about itself at the same angular velocity as the support. This leads to a spurious moment of inertia so that the mass no longer acts as a point mass and the departure from the ideal model described in Section 1.1 and therefore to a theoretical isochronism defect.
(193) Translational isotropic springs have two translational degrees of freedom in which the mass does not rotate but translates along an elliptical orbit around the neutral point. This does away with spurious moment of inertia and removes the theoretical obstacle to isochronism.
9 Isotropic Spring Invention
(194) A.1. As already discussed above, a rotating turntable 1 on which is fixed a spring 2 of rigidity k with the spring's neutral point at the center of rotation of the turntable is illustrated in
(195) We now list which of the theoretical properties of Section 3 hold for these realizations. In particular, for the rotating cantilever spring.
(196) TABLE-US-00005 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes No Yes Yes One direction No No
(197) Note that gravity does not affect the spring when it is in the axial direction. However, these inventions have the disadvantage of having the spring and its support both rotating around their own axes, which introduces spurious moment of inertia terms which reduce the theoretical isochronism of the model. Indeed, considering the point mass of mass m and then including an isotropic support of moment of inertia I and constant total angular momentum L, then if friction is ignored, the equations of motion reduce to
(198)
(199) This equation can be solved explicitly in terms of Jacobi elliptic functions and the period expressed in terms of elliptic integrals of the first kind, see [17] for definitions and similar applications to mechanics. A numerical analysis of these solutions shows that the divergence from isochronism is significant unless the moment of inertia I is minimized.
10 Translational Isotropic Springs: Background
(200) In this section we will describe the background leading to our principal invention of isotropic springs. From now on and unless otherwise specified, isotropic spring will denote planar translational isotropic spring.
(201) 10.1 Isotropic Springs: Technological Background
(202) The invention is based on compliant XY-stages, see references [26, 27, 29, 30] and
(203) In the literature Simon Henein, see reference [14, p. 166, 168], has proposed two XY-stages which exhibit planar isotropy. The first one, illustrated in
(204) 10.2 Isotropic Springs: Simplest Invention and Description of Concept
(205) Isotropic springs are one object of the present invention and they appear most suitable to preserve the theoretical characteristics of the harmonic oscillator are the ones in which the central force is realized by an isotropic spring, where the term isotropic is again used to mean same in all directions.
(206) The basic concept used in all the embodiment of the invention is to combine two orthogonal springs in a plane which ideally should be independent of each other. This will produce a planar isotropic spring, as is shown in this section.
(207) As described above, the simplest version is given in
(208) Sy of rigidity k are placed that spring 12 S.sub.x acts in the horizontal x-axis and spring 11 S.sub.y acts in the vertical y-axis.
(209) There is a mass 10 attached to both these springs and having mass m. The geometry is chosen such that at the point (0, 0) both springs are in their neutral positions.
(210) One can now show that this mechanism exhibits isotropy to first order, see
F(dr)=(kdx,kdy)=kdr
and the central linear restoring force of Section 2 is verified. It follows that this mechanism is, up to first order, a realization of a central linear restoring force, as claimed.
(211) In these realizations, gravity affects the spring in all directions as it changes the effective spring constant. However, the spring does not rotate around its own axis, minimizing spurious moments of inertia, and the central force is directly realized by the spring itself. We now list which of the theoretical properties of Section 3 hold for these embodiments (up to first order).
(212) TABLE-US-00006 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes No No No
(213) Since a timekeeper needs to be very precise, at least 1/10000 for 10 second/day accuracy, an isotropic spring realization must itself be quite precise. This is the subject of embodiments of the present invention.
(214) Since the invention closely models an isotropic spring and minimizes the isotropy defect, the orbits of a mass supported by the invention will closely model isochronous elliptical orbits with neutral point as center of the ellipse.
(215) The principle exposed hereunder by reference to
(216) 10.3 in Plane Orthogonal Non-Compensated Parallel Spring Stages.
(217) The idea of combining two springs is refined by replacing linear springs with parallel springs 171, 172 as shown in
(218) We now list which of the theoretical properties of Section 3 hold for these embodiments.
(219) TABLE-US-00007 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock No Yes Yes No Yes One direction No No
(220) This model has two degrees of freedom as opposed to the model of Section 11.2 which has six degrees of freedom. Therefore, this model is truly planar, as is required for the theoretical model of Section 2. Finally, this model is insensitive to gravity when its plane is orthogonal to gravity.
(221) We have explicitly estimated the isotropy defect of this mechanism and we will use this estimate to compare with the compensated mechanism isotropy defect.
11 Embodiment Minimizing m but not k Isotropy Defect
(222) The presence of intermediate blocks leads to reduced masses which are different in different directions. The ideal mathematical model of Section 2 is therefore no longer valid and there is a theoretical isochronism defect. The invention of this section shown in
(223) In
(224) As a result of the construction, the reduced mass in the x and y directions are identical and therefore the same in every planar direction, thus in theory minimizing reduced mass isotropy defect.
(225) We now list which of the theoretical properties of Section 3 hold for these embodiments.
(226) TABLE-US-00008 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock No Yes Yes Yes Yes One direction No No
12 Embodiment Minimizing k but not m Isotropy Defect
(227) The goal of this mechanism is to provide an isotropic spring stiffness. Isotropy defect, that is, the variation from perfect spring stiffness isotropy, will be the factor minimized in our invention. Our inventions will be presented in order of increasing complexity corresponding to compensation of factors leading to isotropy defects. In plane orthogonal compensated parallel spring stages. Out of plane orthogonal compensated parallel spring stages.
12.1 in Plane Orthogonal Compensated Parallel Spring Stages Embodiment
(228) This embodiment is shown in
(229) In particular,
(230) We now list which of the theoretical properties of Section 3 hold for these embodiments.
(231) TABLE-US-00009 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes No Yes One direction No No
12.2 Alternative in Plane Orthogonal Compensated Parallel Spring Stages Embodiment
(232) An alternative embodiment to the in plane orthogonal compensated parallel spring stages is given in
(233) Instead of having the sequence of parallel leaf springs 192, 194, 196, 198 as in
(234) We now list which of the theoretical properties of Section 3 hold for these embodiments.
(235) TABLE-US-00010 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes No Yes One direction No No
12.3 Compensated Isotropic Planar Spring: Isotropy Defect Comparison
(236) In a specific example computed, the in-plane orthogonal non-compensated parallel spring stages mechanism has a worst case isotropy defect of 6.301%. On the other hand, for the compensated mechanism, worst case isotropy is 0.027%. The compensated mechanism therefore reduces the worst case isotropy stiffness defect by a factor of 200.
(237) A general estimate depends on the exact construction, but the above example estimate indicates that the improvement is of two orders of magnitude.
13 Embodiment Minimizing k and m Isotropy Defect
(238) The presence of intermediate blocks leads to reduced masses which are different for different angles. The ideal mathematical model of Section 2 is therefore no longer valid and there is a theoretical isochronism defect. The invention of this section shown in
(239) Accordingly,
(240) A first plate 201 is mounted on top of a second plate 202 and the numbering has the same significance as in
(241) As a result of this embodiment, the reduced mass in the x and y directions are identical and therefore identical in every direction, thus in theory minimizing reduced mass isotropy defect.
(242) We now list which of the theoretical properties of Section 3 hold for this embodiment.
(243) TABLE-US-00011 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes One direction No No
13.1 Out of Plane Orthogonal Compensated Isotropic Spring Embodiment
(244) Another out of plane orthogonal compensated isotropic spring embodiment is illustrated in
(245) A fixed base 301 holds first pair of parallel leaf springs 302 connected to intermediate block 303. Second pair of leaf springs 304 (parallel to 302) connect to second intermediate block 305. Intermediate block 305 holds third pair of parallel leaf springs 306 (orthogonal to springs 302 and 304) connected to third intermediate block 307. Intermediate block 307 holds parallel leaf springs 308 (parallel to 306) which are connected to orbiting mass 309 (or alternatively frame holding the orbiting mass 309).
(246) We now list which of the theoretical properties of Section 3 hold for this embodiment.
(247) TABLE-US-00012 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes One direction No No
13.2 Reduced Isotropy Defect by Copying and Stacking in Parallel or in Series
(248) We can reduce the isotropy defect by making a copy of the isotropic spring and stacking the copy on top of the original, with a precise angle offset.
(249)
(250) TABLE-US-00013 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
(251)
(252) TABLE-US-00014 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
(253) Typically, the embodiments illustrated in
(254)
(255) TABLE-US-00015 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
(256)
(257) The stiffness isotropy defect of the complete assembly is significantly smaller (typically a factor 100 to 500) than that of a single XY parallel spring stage. The stiffness isotropy can be further improved by stacking more than two stages rotated by angles smaller than 45 degrees. Its properties are
(258) TABLE-US-00016 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes no No No
14 Gravity and Shock Compensation
(259) In order to place the new oscillator in a portable timekeeper, it is necessary to address forces that could influence the correct functioning of the oscillator. These include gravity and shocks.
(260) 14.1 Compensation for Gravity
(261) The first method to address the force of gravity is to make a planar isotropic spring which when in horizontal position with respect to gravity does not feel its effect as described above.
(262) However, this is adequate only for a stationary clock. For a portable timekeeper, compensation is required. This can be achieved by making a copy of the oscillator and connecting both copies through a ball or universal joint as described above in reference to
(263) We now list which of the theoretical properties of Section 3 hold for this embodiment
(264) TABLE-US-00017 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes Yes Yes No
14.2 Dynamical Balancing for Linear Acceleration
(265) Linear shocks are a form of linear acceleration, so include gravity as a special case. Thus, the mechanism of
(266) 14.3 Dynamical Balancing for Angular Acceleration
(267) Effects due to angular accelerations can be minimized by reducing the distance between the centers of gravity of the two masses as shown in
(268) TABLE-US-00018 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes Partially Partially No
(269) Another embodiment is given in
(270) TABLE-US-00019 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes Yes Yes Yes
(271) Another embodiment is given in
(272) We now list which of the theoretical properties of Section 3 hold for this embodiment
(273) TABLE-US-00020 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes Yes Yes No
16 Three Dimensional Translational Isotropic Spring Invention
(274) The three dimensional translational isotropic spring invention is illustrated in
(275) TABLE-US-00021 Isotropic k Radial k Zero J Isotropic m Radial m Gravity Linear shock Angular shock Yes Yes Yes Yes Yes No No No
17 Application to Accelerometers, Chronographs and Governors
(276) By adding a radial display to isotropic spring embodiments described herein, the invention can constitute an entirely mechanical two degree-of-freedom accelerometer, for example, suitable for measuring lateral g forces in a passenger automobile.
(277) In an another application, the oscillators and systems described in the present application may be used as a time base for a chronograph measuring fractions of seconds requiring only an extended speed multiplicative gear train, for example to obtain 100 Hz frequency so as to measure 1/100.sup.th of a second. Of course, other time interval measurement is possible and the gear train final ratio may be adapted in consequence.
(278) In a further application, the oscillator described herein may be used as a speed governor where only constant average speed over small intervals is required, for example, to regulate striking or musical clocks and watches, as well as music boxes. The use of a harmonic oscillator, as opposed to a frictional governor, means that friction is minimized and quality factor optimized thus minimizing unwanted noise, decreasing energy consumption and therefore energy storage, and in a striking or musical watch application, thereby improving musical or striking rhythm stability.
(279) The embodiments given herein are for illustrative purposes and should not be construed in a limiting manner. Many variants are possible within the scope of the present invention, for example by using equivalent means. Also, different embodiments described herein may be combined as desired, according to circumstances.
(280) Further, other applications for the oscillator may be envisaged within the scope and spirit of the present invention and it is not limited to the several ones described herein.
Main Features and Advantages of Some Embodiments of the Present Invention
(281) A.1. A mechanical realization of the isotropic harmonic oscillator. A.2. Utilization of isotropic springs which are the physical realization of a planar central linear restoring force (Hooke's Law). A.3. A precise timekeeper due to a harmonic oscillator as timebase. A.4. A timekeeper without escapement with resulting higher efficiency reduced mechanical complexity. A.5. A continuous motion mechanical timekeeper with resulting efficiency gain due to elimination of intermittent stop & go motion of the running train and associated wasteful shocks and damping effects as well as repeated accelerations of the running train and escapement mechanisms. A.6. Compensation for gravity. A.7. Dynamic balancing of linear shocks. A.8. Dynamic balancing of angular shocks. A.9. Improving chronometric precision by using a free escapement, that is, which liberates the oscillator from all mechanical disturbance for a portion of its oscillation. A.10. A new family of escapements which are simplified compared to balance wheel escapements since oscillator rotation does not change direction. A.11. Improvement on the classical detent escapement for isotropic oscillator.
Innovation of Some Embodiments
(282) B.1. The first application of the isotropic harmonic oscillator as timebase in a timekeeper. B.2. Elimination of the escapement from a timekeeper with harmonic oscillator timebase. B.3. New mechanism compensating for gravity. B.4. New mechanisms for dynamic balancing for linear and angular shocks. B.5. New simplified escapements.
Summary, Isotropic Harmonic Oscillators According to the Present Invention (Isotropic Spring)
Exemplary Features 1. Isotropic harmonic oscillator minimizing spring stiffness isotropy defect. 2. Isotropic harmonic oscillator minimizing reduced mass isotropy defect. 3. Isotropic harmonic oscillator minimizing spring stiffness and reduced mass isotropy defect. 4. Isotropic oscillator minimizing spring stiffness, reduced mass isotropy defect and insensitive to linear acceleration in all directions, in particular, insensitive to the force of gravity for all orientations of the mechanism. 5. Isotropic harmonic oscillator insensitive to angular accelerations. 6. Isotropic harmonic oscillator combining all the above properties: Minimizes spring stiffness and reduced mass isotropy and insensitive to linear and angular accelerations.
Applications of Invention A.1. The invention is the physical realization of a central linear restoring force (Hooke's Law). A.2. Invention provides a physical realization of the isotropic harmonic oscillator as a timebase for a timekeeper. A.3. Invention minimizes deviation from planar isotropy. A.4. Invention free oscillations are a close approximation to closed elliptical orbits with spring's neutral point as center of ellipse.
(283) A.5. Invention free oscillations have a high degree of isochronism: period of oscillation is highly independent of total energy (amplitude). A.5. Invention is easily mated to a mechanism transmitting external energy used to maintain oscillation total energy relatively constant over long periods of time. A.6. Mechanism can be modified to provide 3-dimensional isotropy.
Features N.1. Isotropic harmonic oscillator with high degree of spring stiffness and reduced mass isotropy and insensitive to linear and angular accelerations. N.2. Deviation from perfect isotropy is at least one order of magnitude smaller, and usually two degrees of magnitude smaller, than previous mechanisms. N.3. Deviation from perfect isotropy is for the first time sufficiently small that the invention can be used as part of a timebase for an accurate timekeeper. N.4. Invention is the first realization of a harmonic oscillator not requiring an escapement with intermittent motion for supplying energy to maintain oscillations at same energy level.
REFERENCES (ALL INCORPORATED BY REFERENCE IN THE PRESENT APPLICATION)
(284) [1] Joseph Bertrand, Theoreme relatif au mouvement d'un point attire vers un centre fixe, C. R. Acad. Sci. 77 (1873), 849-853. [2] Jean-Jacques Born, Rudolf Dinger, Pierre-Andr Farine, SaltoUn mouvement mcanique a remontage automatique ayant la prcision d'un mouvement a quartz, Societe Suisse de Chronomtrie, Actes de la Journe d'Etude 1997. [3] H. Bouasse, Pendule Spiral Diapason II, Librairie Delagrave, Paris 1920. [4] Antoine Breguet, Rgulateur isochrone de M. Yvon Villarceau, La Nature 1876 (premier semestre), 187-190. [5] Louis-Clment Breguet, Brevet d'Invention 73414, 8 juin 1867, Ministre de l'agriculture, du Commerce et des Travaux publics (France). [6] George Daniels, Watchmaking, Updated 2011 Edition, Philip Wilson, London 2011. [7] Leopold Defossez, Les savants du XVIleme siecle et la mesure du temps, Edition du Journal Suisse d'Horlogerie, Lausanne 1946. [8] Leopold Defossez, Theorie Generale de l'Horlogerie, Tome Premier, La Chambre suisse d'horlogerie, La Chaux-de-Fonds 1950. [9] Rupert T. Gould, The Marine Chronometer, Second Edition, The Antique Collector's Club, Woodbrige, England, 2013. [10] R. J. Griffiths, William Bond astronomical regulator No. 395, Antiquarian Horology 17 (1987), 137-144. [11] Jules Haag, Sur le pendule conique, Comptes Rendus de l'Acadmie des Sciences, 1947, 1234-1236. [12] Jules Haag, Les mouvements vibratoires, Tome second, Presses Universitaires de France, 1955. [13] K. Josic and R. W. Hall, Planetary Motion and the Duality of Force Laws, SIAM Review 42 (2000), 114-125. [14] Simon Henein, Conception des guidages flexibles, Presses Polytechniques et Universitaires Romandes, Lausanne 2004. [16] Christiaan Huygens, Horologium Oscillatorium, Latin with English translation by Ian Bruce, www.17centurymaths.com/contents/huygenscontents.html [17] Derek F. Lawden, Elliptic Functions and Applications, Springer-Verlag, New York 2010. [18] J. C. Maxwell, On Governors, Bulletin of the Royal Society 100 (1868), 270-83. en.wikipedia.org/wiki/File:On_Governors.pdf [19] Isaac Newton, The Mathematical Principles of Natural Philosophy, Volume 1, Translated by Andrew Motte 1729, Google eBook, retrieved Jan. 10, 2014. [20] Niaudet-Breguet, Application du diapason a l'horlogerie. Sance de lundi 10 dcembre 1866. Comptes Rendus de l'Acadmie des Sciences 63, 991-992. [21] Derek Roberts, Precision Pendulum Clocks, Schiffer Publishing Ltd., Atglen, Pa., 2003. [22] Seiko Spring Drive official website, www.seikospringdrive.com, retrieved Jan. 10, 2014. [23] William Thomson, On a new astronomical clock, and a pendulum governor for uniform motion, Proceedings of the Royal Society 17 (1869), 468-470. [24] Yvon Villarceau, Sur les rgulateurs isochrones, drivs du systme de Watt, Comptes Rendus de l'Acadmie des Sciences, 1872, 1437-1445. [25] Philip Woodward, My Own Right Time, Oxford University Press 1995. [26] Awtar, S., Synthesis and analysis of parallel kinematic XY flexure mechanisms. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, 2006. [27] M. Dinesh, G. K. Ananthasuresh, Micro-mechanical stages with enhanced range, International Journal of Advances in Engineering Sciences and Applied Mathematics, 2010. [28] L. L. Howell, Compliant Mechanisms, Wiley, 2001. [29] Yangmin Li, and Qingsong Xu, Design of a New Decoupled XY Flexure Parallel Kinematic Manipulator with Actuator Isolation, IEEE 2008 [30] Yangmin Li, Jiming Huang, and Hui Tang, A Compliant Parallel XY Micromotion Stage With Complete Kinematic Decoupling, IEEE, 2012