Method of manufacturing plurality of mechanical resonators in manufacturing wafer
Patent Information
- Application Number
- JP2022114391
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-07-20
- Filing Date
- 2022-07-15
- Publication Date
- 2025-07-24
AI Technical Summary
Existing methods struggle to compensate individual mechanical resonators in a wafer to achieve frequencies close to specification, leading to variability and accuracy issues.
A method involving fabricating resonators in a wafer, measuring their actual frequencies, grouping them, and applying tuning masses to compensate for frequency offsets, with optional pre-treatment steps like oxidation and silicon oxide deposition to achieve precise frequency tuning.
The method effectively brings resonators within close proximity to their reference specifications, reducing frequency variability and enhancing accuracy by adjusting their moment of inertia and orientation sensitivity.
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method for manufacturing a plurality of mechanical resonators in a manufacturing wafer, the resonators being intended to comprise regulating members of a timepiece, the method comprising the step of fabricating a plurality of resonators in at least one wafer according to reference specifications.
Background Art
[0002] WO 2021 / 053501 pamphlet teaches such a method, wherein, after the step of fabricating the resonator in the reference wafer according to reference specifications, which comprises: a) at least one lithography step for forming a pattern of a plurality of resonators on or above at least one reference wafer, and a step of machining the reference wafer through the pattern, the following steps are continued; (b) creating, for at least one reference plate, a map showing the variation of the rigidity of the resonator with respect to the average rigidity value; (c) dividing the mapping into regions and determining a correction to be applied to the dimensions of the resonator for at least one of the plurality of regions in order to reduce the variation; (d) modifying the reference specifications of the lithography step so as to apply the dimension correction to at least one of the regions in the lithography step; (e) fabricating a resonator in a manufacturing wafer using the modified specifications.
[0003] One problem associated with the method known from WO 2021 / 053501 pamphlet is that it is not possible to compensate each or substantially each individual resonator so that the resonator exhibits a frequency close to the specification.
Summary of the Invention
Problems to be Solved by the Invention
[0004] Therefore, the object of the present invention is to compensate for each resonator or at least most of the resonators, and to make the resonators as close to the specifications as possible. [Means for solving the problem]
[0005] According to the present invention, this objective is achieved by following the methods described in one or more of the appended claims.
[0006] The present invention is primarily embodied in a method for manufacturing a plurality of mechanical resonators in a wafer, wherein the resonators are intended to include adjustment members for a timepiece, and the method is (a) Fabricate multiple resonators on at least one wafer in accordance with the standard specifications; (b) Measuring the actual natural frequency of each of the multiple resonators; (c) Determining the actual frequency offset of the resonator to the resonator reference specification; (d) Compensating for the offset of a concerned resonator by applying at least two masses from a series of tuning masses to at least one of the multiple resonators so that the concerned resonator approaches the resonator reference specification. This process includes the following steps. In most cases, this is sufficient to bring the resonator within specifications.
[0007] Before applying step (d), the resonators are preferably removed from the wafer and sorted into groups (G1 to G4), where the resonators in a particular group have a first offset from the reference specification within a predetermined first range, and this first offset within the predetermined first range is different from a second offset within a predetermined second range for the resonators in another group, and preferably the same applies to the other groups. Thus, this provides a rough tuning step preceding the final tuning step. However, this is not mandatory, and the final tuning step may be the only tuning step.
[0008] It is preferable to use a silicon wafer or SOI wafer. This helps with antimagnetic behavior and provides high elasticity for continued adhesion to the tuning mass.
[0009] Prior to applying step (d), the resonator may be subjected to a step of oxidizing the resonator and / or depositing silicon oxide on the resonator, followed by controlled removal of the silicon oxide to provide the resonator, preferably all of the resonators, with a silicon oxide layer thickness that brings the resonators, preferably all of the resonators, closer to the standard specification.
[0010] Before applying step (d), it is preferable that each group be treated to a target frequency, preferably with a width of 1 Hz or less. In one embodiment, the group is treated by oxidizing the resonator and / or depositing silicon oxide on the resonator, followed by controlled removal of the silicon oxide to achieve the target frequency.
[0011] In this invention, for each resonator compensated for reasons of symmetry, preferably two or a multiple of two masses are applied.
[0012] In one embodiment, this involves applying at least two masses from a series of tuning masses, each of which has a center of mass outside the geometric center of the mass, and preferably the masses applied to the resonator are rotated in order to fine-tune the resonator to bring it closer to a reference specification.
[0013] The method of the present invention makes it possible for the series of tuning masses to cover a range of the resonator of 10-15% of the standard specifications of the resonator.
[0014] It is preferable that adjacent masses in the series of tuning masses have different weights so that the resonator can be tuned in a frequency step of 0.5% of the reference specification of the resonator.
[0015] In one embodiment, exactly two masses are applied to tune the resonator to its reference specifications. As already mentioned above, two masses are required for symmetry (each mass applied to each mass of the resonator). If only one tuning mass is used, it will create internal stress in the resonator due to the rising temperature, thus negatively affecting accuracy.
[0016] Preferably, the mass has a tolerance of 10% relative to the design weight. This allows for tolerances for manufacturing and assembly errors and ensures that orientation errors remain low.
[0017] It is even more preferable that the tuning range of each mass overlaps by 0-50% with the tuning range of adjacent masses in the series of tuning masses. In this way, manufacturing tolerances of the tuning masses can be easily addressed.
[0018] Adding the tuning mass to the resonator is more preferably done to increase the moment of inertia of the resonator by at least 1 to 100%, or more preferably 1 to 30%, compared to the same resonator without the tuning mass. This is particularly advantageous for lower-frequency resonators or when a smaller group of tuning masses is desired. This allows the frequency of the resonator to be reduced without using a narrower beam, which benefits the robustness of the resonator and reduces the possibility of manufacturing errors.
[0019] One embodiment of the method of the present invention is characterized by applying three masses to at least one of a plurality of resonators in order to tune the resonator frequency and to adjust the sensitivity of the resonator orientation.
[0020] Another embodiment of the method of the present invention is characterized by applying three masses to at least one of a plurality of resonators, wherein one of the three masses is applied to set the frequency of the resonator to a desired frequency, and two of the three masses, designated as tuning masses, are applied to fine-tune the frequency of the resonator.
[0021] International Publication No. 2017 / 068538 discloses an oscillator for regulating the movement of a mechanical timepiece, the oscillator comprising an escape wheel and a resonator forming the time base of the oscillator, the resonator including a mass element held in an oscillating state by at least two oscillating elements, the mass element including at least one anchor portion firmly coupled to the mass element and configured to directly engage with the escape wheel to maintain the oscillation of the resonator. The oscillator is made from a single substrate, preferably glass, ceramic, glass-ceramic, or silicon substrate, which are in the form of a wafer. The moment of inertia of the mass element can be altered by adding or removing weight from the mass element.
[0022] Swiss Patent No. 709291 relates to a rotary oscillator for a timepiece, which comprises a support element designed to enable the oscillator to be assembled in a timepiece, a balance, a plurality of flexible blades connecting the support element to the balance and capable of exerting a restoring torque on the balance, and a felloe fixedly attached to the balance.
[0023] European Patent No. 3182213 relates to a mechanism for adjusting the average speed within a timepiece movement, comprising a gang wheel and a mechanical oscillator. A plurality of blades that are elastically flexible in the oscillation plane support and return the balance so that the balance oscillates at an angle within the oscillation plane. The pallet fork has two hard pallets, which are firmly connected to the balance and are configured to cooperate alternately with the teeth of the gang wheel when the balance oscillates at an angle.
[0024] U.S. Patent Application Publication No. US2021 / 0026299 relates to a method including the steps of: a) providing a substrate including a first silicon layer, a second silicon layer, and an intermediate silicon oxide layer therebetween; b) etching the first silicon layer to form a timepiece component therein; c) releasing a wafer including the timepiece component formed by at least all or part of the etched first silicon layer from the substrate; d) thermally oxidizing and then deoxidizing the timepiece component; e) forming a silicon oxide layer on the timepiece component by thermal oxidation or deposition; and f) removing the timepiece component from the wafer.
[0025] The present invention will be further described hereinafter by referring to exemplary embodiments of the method according to the present invention, which are not restrictive with respect to the appended claims.
[0026] The following description is made with reference to the accompanying drawings.
Brief Description of the Drawings
[0027] [Figure 1A]Figure 1A shows a wafer with a resonator tuned using a tuning mass according to the present invention. [Figure 1B] Figure 1B shows a wafer with a resonator tuned using a tuning mass according to the present invention. [Figure 2] Figure 2 shows the grouped resonators along with the tuning masses. [Figure 3] Figure 3 shows a resonator equipped with two tuning masses. [Figure 4] Figure 4 shows the possible frequency range covered by the tuning mass. [Modes for carrying out the invention]
[0028] The resonators can be manufactured, for example, by lithography or any other suitable manufacturing method that provides the resonators to the wafer 1. This is shown in Figures 1A and 1B. The wafer 1 shown in Figures 1A and 1B is preferably made of silicon or is an SOI wafer. It is clearly shown that the resonators G1 to G4 are randomly dispersed while still in the wafer 1.
[0029] Two different manufacturing methods are possible; the first method is more precise than the second method and involves the following steps: a. A resonator is fabricated within wafer 1. The resulting resonator is dispersed over a certain frequency range (for example, with a width of approximately 6 Hz). See Figure 1B. b. Remove all resonators from wafer 1. c. Measure the natural frequency of each resonator. d. Resonators with similar frequencies are grouped together into coarse tuning groups, such as G1, G2, G3, G4, etc., with a frequency range of approximately 1 Hz (e.g., 50.0-51.0 Hz, 51.0-52.0 Hz, 53.0-54.0 Hz, and 54.0 Hz-55.0 Hz). This is shown on the right side of Figure 1B. e. The group is treated to a target frequency with a maximum bandwidth of approximately 1 Hz, for example, by oxidizing the resonator and / or depositing silicon oxide and removing the silicon oxide (resulting in, for example, a reduction in the total bandwidth from 6 Hz to 1 Hz). f. Select the precise mass for each resonator (for example, from a series of, e.g., 10, masses M1, M2, M3, etc., with progressively increasing weight; see Figure 2), and compensate for the offset from the target resonant frequency according to the specifications. For accuracy reasons, it is desirable that each tuning mass be able to tune a Δ of, for example, 0.1 Hz. Preferably, a series of 10 tuning masses are used, covering a range of 1 Hz before and after the target design frequency. This is shown in Figure 4.
[0030] In a second, faster manufacturing method according to the present invention, the following steps can be distinguished: a. A resonator is fabricated within wafer 1. The resulting resonator is distributed over a certain frequency range (for example, with a width of approximately 6 Hz). See Figure 1A. b. Measure all individual resonators while they are still part of wafer 1. c. All resonators are removed from wafer 1 and resonators with similar frequencies are grouped into, for example, tuning groups G1, G2, G3, G4, etc. d. Select the precise mass for each resonator (from a series of 60 masses) and compensate for the offset from the target design frequency according to the specifications. This is shown in Figure 1A, which relates to grouping multiple resonators G1-G4 with similar frequencies into fine tuning groups, for example, of about 0.1 Hz.
[0031] In summary, both Figure 1A and Figure 1B depict the results of manufacturing multiple mechanical resonators G1 to G4 on a wafer 1, where the multiple resonators G1 to G4 are manufactured according to standard specifications, and the following process is then applied. Measure the actual frequency of each of the multiple resonators G1 to G4; thereafter, Determine the actual frequency offset of the multiple resonators G1 to G4 relative to the standard specification; The offset of the target resonators G1 to G4 is compensated by applying at least two masses M1 from a series of tuning masses to at least one of the multiple resonators G1 to G4 (as shown in Figure 3) so that the multiple resonators G1 to G4 approach the reference specification.
[0032] Those skilled in the art will see from Figure 3 that each of the masses M1 has a center of mass outside the geometric center of the mass, and that the masses M1 applied to the multiple resonators G1 to G4 are rotated while being applied to the multiple resonators G1 to G4 in order to fine-tune the multiple resonators G1 to G4 to bring them closer to the standard specifications of the resonators.
[0033] Preferably, the series of tuning masses M1, M2, M3, etc., shown in Figure 2, cover a range of multiple resonators G1-G4, up to 15% of the reference specification of the resonators. Furthermore, it is desirable that adjacent masses M1, M2, or M2, M3, etc., in the series of tuning masses M1-M4 have different weights so as to enable tuning of each resonator G1-G4 at a frequency step of 0.5% of the reference specification of the resonators. For example, at a design frequency of 50 Hz, the tuning frequency range is preferably 2 Hz, and the frequency step in that case is 0.2 Hz. However, as shown in the exemplary embodiments in Figures 1A, 1B, and 4, different values are clearly feasible.
[0034] Referring to Figure 3, it is preferable that at least two masses M1 have a weight difference of less than 0.5% from each other. This figure also depicts an embodiment in which exactly two masses M1 are applied to tune the resonator G1 to its reference specifications. However, this can also be a multiple of two masses M1.
[0035] Ideally, each mass should have a tolerance of 10% relative to its respective design weight.
[0036] Figure 4 illustrates that the tuning range of each mass overlaps by 0-50% with the tuning range of adjacent masses in the series of tuning masses. This is illustrated by the center of mass frequencies of 43.0 Hz, 43.1 Hz, 43.2 Hz, etc. The center of mass frequency lies in the middle of the tuning range, and adjacent tuning ranges clearly overlap.
[0037] By adding tuning masses to the multiple resonators G1 to G4, it is preferable that the moment of inertia of the multiple resonators G1 to G4 increases by 1 to 30%, or preferably at least 1 to 100%, compared to the same resonators without tuning masses.
[0038] Although the present invention has been described above with reference to exemplary embodiments of the method of the present invention, the present invention is not limited to those specific embodiments, and can be modified in many ways without departing from the present invention. Therefore, the exemplary embodiments described herein should not be used to interpret the appended claims strictly in accordance with them. On the contrary, the embodiments are intended only to illustrate the language of the appended claims without any intention to limit the claims to those exemplary embodiments. Accordingly, the scope of protection of the present invention should be interpreted solely in accordance with the appended claims, and any possible ambiguity in the language of the claims should be resolved by using those exemplary embodiments.
[0039] For example, applying three masses to at least one of a plurality of resonators in order to adjust the resonator frequency and the sensitivity of the resonator orientation is within the scope of the present invention.
[0040] Furthermore, within the scope of the present invention, three masses are applied to at least one of the multiple resonators, where one of the three masses is applied to set the frequency of the resonator closer to its reference specification, and two of the three tuning masses are applied to fine-tune the frequency of the resonator closer to the reference specification of the resonator.
Claims
1. A method for manufacturing a plurality of mechanical resonators (G1 to G4) in a wafer (1) to be manufactured, wherein the plurality of resonators (G1 to G4) are intended to comprise regulating members of a timepiece, the method comprising: (a) manufacturing a plurality of resonators (G1 to G4) in at least one wafer (1) according to a reference specification; (b) measuring the actual frequency of each of the plurality of resonators (G1 to G4); (c) determining the offset of the actual frequency of the resonators (G1 to G4) with respect to the reference specification; (d) applying at least two masses from a series of tuning masses (M1 to M4) to at least one of the plurality of resonators (G1 to G4) to compensate for the offset of the target resonator (G1 to G4) so as to bring the target resonator (G1 to G4) closer to the reference specification. The method as described above, including the steps.
2. The method according to claim 1, characterized in that the resonators are sorted into groups (G1 to G4) before applying step (d), the resonators in a specific group having a first offset from the reference specification within a predetermined first range, and the first offset within the predetermined first range being different from a second offset within a predetermined second range of the resonators in a second group.
3. The method according to claim 1 or 2, characterized by providing a wafer (1) made of silicon or a SOI wafer.
4. The method according to claim 1 or 2, characterized in that, before applying step (d), the resonators are subjected to a process of controlled oxidation of the resonator and / or a process of depositing silicon oxide on the resonator, followed by a controlled removal of the silicon oxide to provide, for all the resonators, a thickness of a layer of silicon oxide that brings all the resonators closer to the reference specification.
5. The method according to claim 1 or 2, characterized in that, before applying step (d), the group is processed to a target frequency, preferably with a width of 1 Hz or less.
6. The method according to claim 5, characterized in that the group is processed by oxidizing the resonator and / or depositing silicon oxide on the resonator, followed by a controlled removal of the silicon oxide to reach the target frequency.
7. Applying at least two masses from a series of tuning masses (M1 to M4), where each of said masses (M1 to M4) has a center of mass outside the geometric center of the mass, and the masses M1 to M4 applied to the plurality of said resonators (G1 to G4) are rotated to fine-tune the plurality of said resonators (G1 to G4) closer to the reference specifications of the resonator, the method according to claim 1 or 2, characterized thereby.
8. The method according to claim 1 or 2, characterized in that the series of tuning masses (M1 to M4) covers a range of the plurality of said resonators (G1 to G4) up to a maximum of 15% of the reference specifications of the resonator.
9. Adjacent masses (M1, M2; M2, M3; M3, M4; etc.) in the series of tuning masses are provided with different weights so as to enable tuning the resonators in frequency steps of 0.5% of the reference specifications of the plurality of said resonators (G1 to G4), the method according to claim 1 or 2, characterized thereby.
10. The method according to claim 1 or 2, characterized in that exactly two masses or a multiple of two masses are applied to tune the resonators (G1 to G4) closer to the reference specifications of the resonator.
11. The method according to claim 1 or 2, characterized in that each mass (M1 to M4) has a tolerance of 10% with respect to its designed weight.
12. The method according to claim 1 or 2, characterized in that the tuning range of each mass (M1 to M4) overlaps with the tuning range of adjacent masses in the series of tuning masses (M1 to M4) by 0 to 50%.
13. The method according to claim 1 or 2, characterized in that by adding tuning masses (M1 to M4) to the plurality of said resonators (G1 to G4), the moment of inertia of the resonator (G1 to G4) increases by at least 1 to 100%, or preferably at least 1 to 30%, compared to the same resonator without tuning masses.
14. The method according to claim 1 or 2, characterized in that three masses are applied to at least one of the plurality of resonators (G1 to G4) in order to adjust the resonator frequency and to adjust the sensitivity of the orientation of the resonator.
15. Applying three masses to at least one of the plurality of resonators (G1 to G4), wherein one of the three masses is applied to set the frequency of the resonator (G1 to G4) close to its reference specification, and two of the three tuning masses are applied to fine-tune the frequency of the resonator (G1 to G4) closer to the reference specification of the resonator. The method according to claim 1 or 2, characterized by being applied.