Servo-type vibration detector and method for evaluating servo-type vibration detector

Abstract

To solve the issue that improvement of performance of an existing servo-type acceleration sensor is limited since the sensor had to be formed in a small range that satisfies the three contradicting objectives of (1) reduction of a resonance peak, (2) increase of responsivity, and (3) increase of the sensor sensitivity.SOLUTION: A mechanical attenuation action due to a dynamic fluid pressure in a gap part between electrodes is reduced by forming a circulation hole or a circulation groove in a relative moving surface of electrodes. The attenuation action is replaced by equivalent attenuation means in an electric circuit in a servo amplifier. In that way, the gap between electrodes, the sensor sensitivity determined by the external diameter, and the sensor dynamic characteristics, which has been in a trade-off relation, can be independently set.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a vibration sensor or a vibration control device that detects signals of acceleration, absolute velocity, or absolute displacement of a controlled object supported on a base and vibrating under disturbance in a wide frequency band.

Background Art

[0002] 1. Trend in the World In various fields such as semiconductor manufacturing processes, liquid crystal manufacturing processes, and precision machining, the use of vibration control for blocking and suppressing minute vibrations is expanding. Microfabrication and inspection devices such as scanning electron microscopes and semiconductor exposure apparatuses (steppers) used in these processes are required to have strict vibration tolerance conditions to ensure the performance of the devices. In the future, with the further high integration and miniaturization of products, the processing speed of processing processes and the enlargement of devices are progressing, and the vibration tolerance conditions tend to become increasingly strict.

[0003] 2. Disturbances to be Removed by a Vibration Isolation Device In recent years, an active vibration control technology that generates a control signal based on displacement, velocity, and acceleration information from vibration sensors arranged at multiple locations of a structure to be vibration-controlled (for example, a precision vibration isolation table) and controls a control device has become widespread.

[0004] Figure 32 shows a model diagram of a conventional active vibration isolation table. This active vibration isolation table is a well-known device as described in Patent Document 1 and Patent Document 2. A plurality of sets of pneumatic actuators (502a, 502b) for supporting the surface plate 501 are arranged on the floor surface 500. A precision device (not shown) is mounted on this surface plate 501. 503 is an acceleration sensor for detecting the acceleration of the surface plate 501 in the vertical and horizontal directions, and 504 is an acceleration sensor for detecting the acceleration of the floor surface 500 (the vibration state of the foundation). 505a and 505b are displacement sensors for detecting the vertical and horizontal relative displacements of the surface plate 501 with respect to the floor surface 500, respectively. Output signals from these respective sensors are input into the controller 506. A servo valve 508 controlled by the controller 506 is connected to the pneumatic actuator 502a via a pipe 507. By adjusting the flow rate of the compressed air supplied to and exhausted from the pneumatic actuator 502a by this servo valve 508, the internal pressure of the actuator 502a is controlled to drive the pneumatic actuator.

[0005] The disturbances to be removed in the vibration isolation device are roughly classified into ground motion disturbances caused by the vibration of the installation floor and linear motion disturbances input from the vibration isolation table.

[0006] As the sources of the vibrations that become ground motion disturbances, those caused by the movement of people called walking vibrations are 1 to 3 Hz, those caused by motors such as air conditioners are 6 to 35 Hz, and the resonance points of the floor and walls are about 10 to 100 Hz. Super high-rise and vibration-isolated buildings have natural frequencies in the vicinity of 0.2 to 0.3 Hz. In addition, due to wind sway, the building generates micro-vibrations of 0.1 to 1.0 Hz. Therefore, the vibration isolation table is required not only to suppress high-frequency vibrations but also to remove low-frequency vibrations.

[0007] As a source of high-frequency vibration caused by linear motion disturbance, when, for example, a positioning stage 509 is mounted on a vibration isolation table, the structure including the vibration isolation table is struck by the acceleration and deceleration operation of the stage and swings due to the driving reaction force. If the vibration caused by this impact and the sway caused by the driving reaction force are not suppressed, the performance of the stage cannot be maintained. In summary, in addition to "vibration isolation" caused by ground motion disturbance, the vibration isolation device is required to have a function that combines both "vibration control" caused by linear motion disturbance.

[0008] 3. Role of vibration sensors in active vibration isolation devices In active vibration control, a control method based on state feedback is adopted. This is a method of controlling a control device based on acceleration, velocity, and displacement information from vibration sensors arranged at multiple locations on the structure to be vibration-controlled. In order to obtain vibration isolation performance in a wide frequency range, for example, the acceleration signal is mainly used to control the state quantity above 10 Hz, the velocity signal is used to control the state quantity between 1 and 10 Hz, and the displacement signal is used to control the state quantity below 1 Hz. For example, a. If acceleration feedback is applied using the signal from the acceleration sensor arranged on the surface plate 501, effects such as being equivalent to an increase in mass M, decreasing the natural frequency, and reducing the resonance peak can be obtained. b. If the signal from the above acceleration sensor is converted into an absolute velocity or absolute displacement signal and feedback or feedforward is applied, significant improvement in vibration isolation performance can be achieved in a wide frequency range. c. If the signal from the acceleration sensor arranged directly below the surface plate 501 is used, the signal is converted into an absolute velocity or absolute displacement signal, and feedforward is similarly applied, the vibration isolation performance can be improved in a wide frequency range.

[0009] In order to perform the above controls b and c, velocity and position information with respect to the inertial space is required. Since the acceleration sensor can measure the acceleration with respect to the inertial space, by attaching the acceleration sensor to the object to be controlled, the acceleration applied to the object to be controlled can be detected. Therefore, in conventional active vibration isolation devices, a method is adopted in which the velocity signal is obtained by integrating the output of the acceleration sensor once, and the displacement signal is obtained by further integrating it twice.

[0010] 4. Basic Structure and Detection Principle of Acceleration Sensor Fig. 33 is a model diagram showing the basic structure and detection principle of a capacitance-type acceleration sensor. 301 is a main body that houses each component of the sensor, 302 is a mass body, 303 is a spring that mechanically supports the mass body 302 with respect to the vibration measurement surface A, and 304 is a damper. The mass body 302 also serves as the movable-side electrode of the capacitance-type sensor. 305 is a fixed-side electrode arranged on the opposite surface side of the movable-side electrode (mass body 302), and 306 is a gap between the two electrodes.

[0011] 307 is an electromagnetic actuator that drives the mass body 302 in the vertical direction with respect to the vibration measurement surface A. Since the capacitance C is determined by the size of the gap in the gap portion 306, by measuring this capacitance C, the relative displacement U - X, which is the difference between the ground motion absolute displacement U and the absolute displacement X of the mass body, can be detected. The servo circuit 310 (shown by a two-dot chain line) is composed of a proportional amplifier 312 that amplifies with a gain K P through a displacement detector 311 that detects the relative displacement signal U - X.

[0012] Hereinafter, the detection principle of the acceleration sensor will be described using mathematical formulas. Let the mass of the mass body 302 be m, the spring constant of the mechanical spring 303 that supports the mass body be k, the damping coefficient of the damper 304 be c, and the driving force of the actuator 307 be F = A f i0, then the following equation of motion holds.

[0013]

Equation

[0014] The current i0 of the actuator is controlled by an amplifier with a proportional gain constant K P so that the relative displacement u - x becomes zero.

[0015]

Equation

[0016] [Number]

[0017] Proportional gain constant K P is sufficiently large, and if the first and second terms on the right side of Equation (3) can be ignored compared to the third term, then

[0018] [Number]

[0019] If the current i0 flowing through the actuator is detected from Equation (2) and Equation (4), the acceleration of the mass 302 can be approximately obtained.

[0020] 5. Specific Structure of Conventional Servo-Type Accelerometers FIG. 34 is a front cross-sectional view showing a specific structural example of a conventional capacitance-type accelerometer [Patent Document (3)], and is configured according to the basic configuration and detection principle shown in FIG. 33. 11 is a permanent magnet, 12 is a pole piece portion, 13 is a pole piece convex portion, 14 is a permanent magnet side yoke material, 15 is a coil side yoke material, 16a is a force coil, 16b is a calibration coil, 17 is a coil bobbin, 18 and 19 are coil bobbin support members made of a non-magnetic and non-conductive material, 20 is a front side disk-shaped spring, 21 is a rear side disk-shaped spring, 22 is a front side connecting member between the front side disk-shaped spring 20 and the coil side yoke material 15, and 23 is a rear side connecting member between the rear side disk-shaped spring 21 and the coil side yoke material 15.

[0021] 24 is a movable side electrode, 25 is a fixed side electrode, 26 is a front side panel, 27 is a central plate, and 28 is a fastening member between the fixed side electrode 25 and the front side panel 26.

[0022] A magnetic air gap 29 in the radial direction is formed between the outer peripheral portion of the pole piece portion 12 and the inner peripheral portion of the coil side yoke member 15. 29a is the permanent magnet side air gap, and 29b is the yoke member side air gap. A closed-loop magnetic circuit is formed by "permanent magnet 11 → pole piece portion 12 → magnetic air gap 29 → coil side yoke member 15 → permanent magnet side yoke member 14". When a current flows through the force coil 16a disposed in the space of the magnetic air gap 29, a Lorentz force is generated to move the movable side electrode 24 in the axial direction. 30 is the air gap formed by the movable side electrode 24 and the fixed side electrode 25. Since the capacitance C is determined by the size of the gap of the air gap 30, the relative displacement U-X, which is the difference between the ground motion absolute displacement U and the absolute displacement X of the mass body, can be detected by measuring the capacitance C. The servo circuit is composed of a displacement detector 31, an amplifier 32, and a driver 33. The amplifier 32 and the driver 33 are displacement amplifiers that amplify the relative displacement signal U-X with a gain K P It is a displacement amplifier that amplifies with a proportional gain constant K so that the relative displacement u-x becomes zero P The current i0 of the actuator is controlled by an amplifier. If the current i0 flowing through the force coil 16a is detected, the acceleration acting on the movable part can be obtained as described above

Prior Art Documents

Patent Documents

[0023]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Non-Patent Documents

[0024]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0025] When applying a servo - type acceleration sensor to an active vibration isolation table, it is necessary to simultaneously satisfy the following three conditions. (1) The resonance frequency is sufficiently high and the resonance peak value is small (2) Improvement in responsiveness (reduction of phase delay in the frequency range below the resonance point) (3) Obtaining high sensor sensitivity

[0026] The resonance frequency of passive - type element devices such as pneumatic actuators that are the control targets of an active vibration isolation table is usually f0 = 4 - 6 Hz. To control this passive - type element, a control range of 40 - 50 Hz is required for the active vibration isolation table. To obtain sufficient control performance within this control range, high responsiveness is required for control elements other than the pneumatic actuator that is the control target (acceleration sensor, servo valve). As an evaluation index based on many empirical values, the phase delay (responsiveness) of the acceleration sensor at f = 100 Hz was ideally Δθ≦10 deg.

[0027] In the above (1), the reason why a high resonance frequency f0 is required for the acceleration sensor is that the higher the resonance frequency f0, the smaller the phase delay in the frequency range below the resonance point, which is the evaluation index in the above (2). However, the resonance frequency of a servo - type acceleration sensor is determined by the following formula based on the mass m of the movable part and the servo stiffness K that gives the restoring force to the actuator.

[0028]

Equation

[0029] The servo stiffness K is proportional to the control gain. The magnitude of this control gain is restricted by delay elements such as the inductance of the coil in addition to the second-order lag element due to the mass m and the stiffness K. Therefore, the resonance frequency of the servo type acceleration sensor has usually been limited to f0 = 350 to 500 Hz.

[0030] In the above (1), if evaluated on the premise that the range of the resonance frequency f0 of the acceleration sensor (f0 = 350 to 500 Hz) cannot be greatly changed, it is desirable to make the resonance peak value at the resonance point as small as possible. The reason is that the smaller the resonance peak value, the larger the gain margin that determines the stability margin of the control system, and the larger the control gain that determines the servo stiffness can be.

[0031] The larger the attenuation, the smaller this resonance peak value can be made. However, when the attenuation increases, the responsiveness decreases and the phase delay at f = 100 Hz increases. Therefore, the above (1) and (2) are in a contradictory relationship (a trade-off relationship). In the case of a capacitive servo type acceleration sensor, the magnitude of this attenuation is determined by the electrode gap and the electrode area (outer diameter of the electrode) for detecting the capacitance.

[0032]

Equation

[0033] In the above (3), the most effective measure to obtain high sensor sensitivity is to set the capacitance of the displacement detector large. As is well known, when the area of two parallel conductor plates is A, the gap is d, the permittivity is ε0, and the relative permittivity of air is ε r the capacitance C is

[0034]

Equation

[0035] That is, the larger the electrode area A and the narrower the electrode gap d, the larger the capacitance C and the higher the sensor sensitivity. However, from Equation (6), this approach increases attenuation and reduces responsiveness. Therefore, (1) to (3) are in a conflicting relationship (a trade-off relationship).

[0036] Fig. 35 shows the gain-phase characteristics of a conventional acceleration sensor with respect to frequency for cases where the electrode gap is h0 = 20, 30, and 40 μm. The specifications of the sensor are a resonance frequency f0 = 350 Hz, an electrode outer diameter of Φ13 mm, and a conventional sensor (excluding the attenuation coefficient) in Table 1 described later. (1) When the electrode gap h0 = 40 μm, the resonance peak value ΔP = 7 dB is the highest, and the phase delay Δθ at f = 100 Hz is the smallest at 10 deg or less. (2) When the electrode gap h0 = 30 μm, the resonance peak value ΔP ≒ 0 dB, and the phase delay Δθ at f = 100 Hz is 14.2 deg. Although the ideal target specification Δθ ≤ 10 deg is not satisfied, it has the most balanced characteristics among the three cases of the electrode gap. (3) When the electrode gap h0 = 20 μm, the resonance peak value ΔP = -10 dB is the lowest, and the phase delay Δθ at f = 100 Hz is the largest at 40 deg. It clearly has characteristics of excessive attenuation and is not applicable to an active vibration isolation table.

[0037] Therefore, from the results of (1) to (3), in the above specifications of the acceleration sensor, the electrode gap must be selected at one point as h0 = 30 μm. As a result, the sensor sensitivity determined by the size of the gap is restricted.

Means for Solving the Problem

[0038] As described above, in the case of a conventional servo-type acceleration sensor, the sensor had to be configured within a narrow range that simultaneously satisfied the three conflicting requirements for the sensor, namely, (1) reduction of resonance peaks, (2) improvement of responsiveness (reduction of phase lag), and (3) improvement of sensor sensitivity. Therefore, there was a limit to performance improvement. The origin of the idea of the present invention was to find a way to resolve these three trade-off relationships and independently solve each problem.

[0039] In the case of a capacitive servo-type acceleration sensor, attention was paid to the fact that, due to its structural characteristics, the damping effect that governs the above three characteristics is the dynamic fluid pressure (squeeze pressure) generated in the gap between the electrodes.

[0040] Thus, the servo-type vibration detector of the first invention of the present application includes a housing as a fixed member, a movable member provided so as to be movable in a predetermined direction with respect to the housing, an elastic member that supports the movable member so as to be disposed with a gap therebetween with respect to the housing, a displacement detection unit that detects displacement of the movable member in a predetermined direction, drive means that is driven by a servo amplifier so as to generate a driving force for returning the movable member to its origin position when a relative displacement from the origin position of the movable member is detected by the displacement detection unit, a movable-side electrode provided on the movable member, and a fixed-side electrode provided on the housing side facing the movable-side electrode. The displacement detection unit is configured to detect the capacitance formed in the gap between the movable-side electrode and the fixed-side electrode, and grooves or holes communicating with the atmosphere are formed on the relative movement surfaces of the movable-side electrode and the fixed-side electrode so as to reduce the damping effect caused by the dynamic fluid pressure generated in the gap between the movable-side electrode and the fixed-side electrode. The servo amplifier is provided with damping means by an electric circuit so as to compensate for the reduction of the damping effect.

[0041] That is, in the present invention, the mechanical damping effect due to the squeeze pressure in the inter-electrode gap is reduced by dividing the electrodes into a plurality of electrodes and setting the boundary portions of the respective electrodes to atmospheric pressure, or by forming a plurality of through holes communicating with the atmospheric pressure on the electrode surface. At the same time, this damping effect is replaced with an equivalent damping means by an electric circuit in a servo amplifier. As a result, appropriate sensor dynamic characteristics can be obtained without depending on the gap and outer diameter of the electrodes. That is, the sensor sensitivity determined by the gap and outer diameter of the electrodes, which was a trade-off relationship in the past, and the sensor dynamic characteristics can be set independently.

[0042] The servo type vibration detector of the second invention of the present application is characterized in that the plurality of groove portions are flow grooves having a substantially concentric shape with respect to the axis of the fixed-side electrode or the movable-side electrode, and / or a substantially radial shape in the radial direction.

[0043] That is, in the present invention, a flow groove communicating with the atmosphere is formed concentrically or radially with respect to the axis of the fixed-side electrode or the movable-side electrode. The groove width is narrow, and the groove depth is formed sufficiently deep compared to the inter-electrode gap, so that each groove portion can maintain atmospheric pressure regardless of the size of the inter-electrode gap. Since the ratio of the area of the groove portion to the total area of the electrodes can be made sufficiently small, the decrease in capacitance due to groove formation can be minimized.

[0044] The servo type vibration detector of the third invention of the present application is characterized in that the damping means by the electric circuit is composed of a differential circuit that differentiates the signal of the relative displacement.

[0045] That is, in the present invention, a differential circuit that differentiates the signal of the relative displacement is provided in an electric circuit that drives the driving means by a servo amplifier.

[0046] The servo-type vibration detector according to the fourth invention of the present application further includes a proportional amplification circuit that proportionally amplifies the signal of the relative displacement output from the displacement detection unit, and the driving means is configured to be driven by the sum signal of the proportional amplification circuit and the differential circuit, and the sum signal is used as a sensor output signal indicating the detected vibration.

[0047] That is, in the present invention, by using the sum signal of the proportional amplification circuit and the differential circuit as the sensor output signal, it has been found that ideal sensor dynamic characteristics (gain and phase characteristics) deviating from conventional common sense can be obtained. As a result of theoretical analysis, this unexpected effect lies in the fact that a phase advance element is introduced into the transfer characteristic of the sensor output with respect to the input acceleration. Due to the action brought about by this phase advance element, the resonance peak value at the resonance point can be reduced, and the phase lag in a wide frequency range can be significantly decreased. According to the present invention, a sensor that can independently set the following three conditions, which were conventionally in a contradictory relationship (a trade-off relationship), (1) improvement of sensor sensitivity, (2) improvement of responsiveness (reduction of phase lag), and (3) reduction of the resonance peak value, can be realized.

[0048] The servo-type vibration detector according to the fifth invention of the present application is characterized in that two sets of the displacement detection units are provided, each of which includes the individual movable-side electrode and the fixed-side electrode, and the gaps of the gaps between the movable-side electrodes and the fixed-side electrodes of the two sets of displacement detection units are configured to change in opposite phases.

[0049] That is, the present invention is configured as a differential sensor that includes two sets of capacitance-type displacement detection units, and the gaps of the displacement detection units change in opposite phases. Microgrooves and through holes are formed on the relative movement surfaces of the two sets of electrodes to reduce the damping effect caused by the dynamic fluid pressure generated in the gap portion, and the servo amplifier that processes the differential signal is provided with damping means by an electric circuit to compensate for the reduction of the damping effect. The mechanical damping of the two sets of electrode gaps is both C MSince it can be made approximately equal to 0, there is no need to sacrifice sensor sensitivity. For example, measures such as increasing the electrode gap to reduce mechanical attenuation are unnecessary.

[0050] Also, in the case of a differential type, the sum of the two electrode gaps, δ L +δ R = has a certain constraint condition, but in the present invention, the electrode gaps δ L , δ R The size and accuracy of do not affect the dynamic characteristics of the sensor. The necessary damping effect can be replaced by electrical damping C E Therefore, the above constraint condition of the differential type in which two sets of capacitance sensors are arranged is eliminated.

[0051] The servo type vibration detector of the sixth invention of the present application is characterized in that the movable side electrodes are provided at both axial ends of the movable member, the fixed side electrodes are provided on the housing side facing these movable side electrodes, and a differential type sensor is configured by detecting two capacitance differences formed between the movable side electrodes and the fixed side electrodes.

[0052] That is, the present invention is configured by an actuator structure in which movable side electrodes can be mounted at both axial ends of the movable member. Therefore, if the axis of the movable member is set as the Z axis and the Y axis orthogonal to the Z axis is set at the center of the movable member, the two sets of movable side electrodes and fixed side electrodes are arranged "mirror symmetrically" with respect to this Y axis. Further, with respect to the axis Z axis of the movable member, two sets of electrodes, permanent magnets, coils, etc. can be arranged "axis symmetrically". The aim of the differential type is to obtain a differential output in which noise and drift are canceled by adding noise and drift commonly to the outputs of the two sets of electrodes. In the present invention arranged "mirror symmetrically" and "axis symmetrically" structurally, the external disturbance noise and drift input to the outputs of the two sets of electrodes are exactly symmetric, so the differential type can be utilized more effectively.

[0053] The servo type vibration detector of the seventh invention of the present application includes a coil fixed to the fixed member side, a permanent magnet arranged so that magnetic flux flows through a gap between the housing and the movable member, and the movable member is composed of the permanent magnet and a movable side yoke material connecting a magnetic path therewith, or is composed of only the movable side yoke material. The movable member, the gap between the housing and the movable member, the fixed member, and the permanent magnet form a closed-loop magnetic circuit, thereby constituting the driving means by an electromagnetic force for moving the movable member in the axial direction.

[0054] That is, in the present invention, paying attention to the fact that the driving means by electromagnetic force is composed of elements such as a movable member, a magnetic air gap portion, a fixed member which is a yoke material, and a permanent magnet, an actuator is configured by a closed-loop magnetic circuit in which a coil is fixed. Therefore, the processes required during mass production of a conventional MC type servo sensor, (i) cutting and dividing of a disc-shaped spring, (ii) insulation of signal wires, (iii) a complicated production method involving soldering of ultra-fine wires, etc. are not required, and productivity can be significantly improved. Also, due to the structural feature that the coil does not move (1) Since both ends of the shaft can be an open structure, the support span L of the movable part can be set large. (2) Since division and cutting of the disc-shaped spring for supporting the movable part are not required, the torsional rigidity of the movable part can be set large. Due to the above (1) and (2), the dynamic stability of the sensor movable part can be achieved. For example, it becomes easy to improve the sensor sensitivity by increasing the diameter of the movable side electrode and increasing the capacitance.

[0055] The servo type vibration detector of the eighth invention of the present application is configured such that one end of the movable member is fixed to the housing and is supported by the elastic member so as to swing. A plurality of divided electrode surfaces or a plurality of through holes communicating with the atmospheric pressure are formed on the front and back surfaces of the movable member, the fixed side facing surfaces facing the front and back surfaces, and the relative moving surfaces of the front and back surfaces and the fixed side facing surfaces.

[0056] That is, the present invention is applicable to a swing motion type sensor in which a pendulum with one end fixed swings within a narrow gap. On the opposing surfaces of the electrodes formed on the front and back of the pendulum, for example, a semi-circular arc-shaped ring is provided, and micro grooves communicating with the atmosphere are formed in this semi-circular arc-shaped ring.

[0057] Even in the case of a swing motion type, the same effects as those of the linear motion type described above can be obtained. That is, (1) the effect of improving dynamic characteristics by the combination of micro grooves and electrical attenuation, (2) the noise and drift canceler effect by the differential type, (3) the improvement of productivity by relaxing the gap accuracy between both electrodes, etc. Further, in the case of the swing motion type, it has been pointed out that due to the asymmetry of the sensor structure, it is susceptible to thermal expansion of the members. However, by the micro grooves formed on the relative movement surface between the electrodes, the mechanical attenuation C M ≒0 can be achieved, and the influence of minute changes due to thermal expansion of the electrode gap on the sensor characteristics can be avoided.

[0058] In the servo type vibration detector of the ninth invention of the present application, when the inertial mass of the movable member is m (Kg), the sum of the electrical gain of the servo amplifier and the mechanical spring constant of the elastic member is the proportional gain K P (N / m), the mechanical damping coefficient C M (Ns / m), the electrical damping coefficient C of the servo amplifier E is defined as the mechanical damping ratio ζ M , and the electrical damping ratio ζ E when defined

[0059]

Equation

[0060] That is, the present invention obtains the range of ζ E , ζ M that satisfies the following (1) and (2) in the gain and phase characteristics of the servo type acceleration sensor. (1) The peak value at the resonance point is 15 dB or less (2) The phase delay Δθ at ω / ω n = 0.2 is 10 deg or less The mechanical damping ratio ζ that satisfies the above (1) M , the electrical damping ratio ζ E The conditions for are ζ M ≧0.2, ζ E ≧0.2. That is, regardless of electrical or mechanical factors, the peak value at the resonance point is determined only by the magnitude of the attenuation. The ζ that satisfies the above (2) M , ζ E The conditions for are ζ M ≦0.6. Even when the electrical damping ratio ζ E is sufficiently large, ζ E does not affect the conditions of the above (2), and the phase delay is determined only by the mechanical damping ζ M .

[0061] The tenth invention of the present application is a method for evaluating a servo type vibration detector, which obtains a damping coefficient from the measured value of the peak value of the gain at the resonance point, and evaluates the effect of the damping means provided in the electric circuit of the servo amplifier from the characteristics of the damping coefficient with respect to the size of the electrode gap.

Effect of the Invention

[0062] Thus, in the servo type vibration detector according to the present invention, the present invention reduces the mechanical damping action due to the squeeze pressure in the electrode gap portion by dividing the electrode into a plurality of electrodes and making the boundary portions of the respective electrodes at atmospheric pressure, or by forming a plurality of through holes communicating with the atmospheric pressure on the electrode surface, and can replace this damping action with an equivalent damping means by an electric circuit in the servo amplifier. As a result, appropriate sensor dynamic characteristics can be obtained without depending on the gap and outer diameter of the electrode. That is, the sensor sensitivity determined by the gap and outer diameter of the electrode, which was a conventional trade-off relationship, and the sensor dynamic characteristics can be set independently.

Brief Description of the Drawings

[0063]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15a

Figure 15b

Figure 15c

Figure 15d

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

Figure 35

[0064] [First Embodiment] [1] Model diagram of the present embodiment FIG. 1 is a model diagram showing the basic configuration and detection principle of the capacitance type acceleration sensor according to Embodiment 1 of the present invention. AA is the sensor main body, and the dashed line BB part is the displacement detection part for detecting capacitance, which also serves as an attenuator. Further, the dashed line CC part is a control circuit part in which a servo amplifier for driving the actuator is built in. Hereinafter, it will be described in comparison with FIG. 32 which is a conventional model diagram.

[0065] In the sensor main body AA, 11 is the main body for housing each member of the sensor, 12 is the mass body, 13 is a spring for mechanically supporting the mass body 12 with respect to the vibration measurement surface A, and 14 is a damper (mechanical damping coefficient C M ). The mass body 12 also serves as the movable side electrode of the capacitance type sensor. 15 is a fixed side electrode disposed on the opposing surface side of the movable side electrode (mass body 12), and 16 is the gap between the two electrodes. 17 is an electromagnetic actuator for driving the mass body 12 in the vertical direction with respect to the vibration measurement surface A. Since the capacitance C is determined by the size of the gap in the gap portion 16, by measuring this capacitance C, the relative displacement U-X, which is the difference between the ground motion absolute displacement U and the absolute displacement X of the mass body, can be detected, which is the same as the conventional type (FIG. 33).

[0066] In the displacement detection part BB, when the gap h0 of the gap part 16 between the two electrodes changes, a dynamic pressure (squeeze pressure) due to the viscosity of air is generated. The damping action by this squeeze fluid pressure becomes the damper 14 in the sensor main body AA. In order to reduce the damping action, a circumferential groove 18 (flow groove, or called a micro groove) is formed on the electrode surface of the present embodiment. A through hole 19 connecting to the outside (atmospheric pressure) is formed in this circumferential groove. Since the groove depth of the circumferential groove 18 is sufficiently deeper than the gap h0 of the gap part 16, the inside of the groove of the circumferential groove (micro groove) 18 becomes atmospheric pressure on the entire circumference.

[0067] In the control circuit part CC, a proportional amplifier 21 that amplifies through a displacement detector 20 for detecting the relative displacement signal U-X with a proportional gain K P and an electrical damping circuit 22 are provided in parallel. The electrical damping circuit 22 has a differential gain 23 (electrical damping coefficient C E) and a pseudo-differential circuit 24. The electrical damping circuit is provided to compensate for the reduction in the mechanical damping effect due to the circumferential groove 18 and the through hole 19. If the electromagnetic actuator 17 is driven by the parallel sum signal of the proportional amplifier 21 and the electrical damping circuit 22 to detect the current flowing through the actuator, the acceleration of the mass body 12 can be approximately obtained.

[0068] As will be clear from the theoretical analysis described later, the sensor dynamic characteristics (gain - phase characteristics) of this embodiment can obtain ideal characteristics. That is, a sensor can be realized in which three conditions that were conventionally in a contradictory relationship (a trade - off relationship), (1) improvement of sensor sensitivity, (2) improvement of responsiveness (reduction of phase delay), and (3) reduction of resonance peak value, can be set independently.

[0069] [2] Specific structure of this embodiment (hereinafter, MC type) FIG. 2 shows an example of a servo - type acceleration sensor according to Embodiment 1 of the present invention. FIG. 2(a) is a view taken along the DD arrow of FIG. 2(b), and FIG. 2(b) is a front cross - sectional view of the sensor body. The dashed - line AB part in FIG. 2(b) is a moving - coil type (MC type) actuator part that drives the movable part in the axial direction. The dashed - line BB part indicates a displacement detection part that detects capacitance. The dashed - line CC part is an outline of a control circuit part in which a servo amplifier for driving the actuator is built in. As described above, the MC type actuator part has a known structure. Hereinafter, the specific structure of this embodiment will be described by dividing it into an actuator part, a displacement detection part, and a control circuit part, similar to the model diagram of FIG. 1.

[0070] [2 - 1] Actuator part In the actuator section (two-dot chain line AB section) in Fig. 2(b), 201 is a permanent magnet, 202 is a pole piece section, 203 is a pole piece convex portion, 204 is a permanent magnet side yoke material, 205 is a coil side yoke material, 206a is a force coil, 206b is a calibration coil, 207 is a coil bobbin, 208 and 209 are coil bobbin support members made of a non-magnetic and non-conductive material, 210 is a front side disk-shaped spring, 211 is a rear side disk-shaped spring, 212 is a front side connecting member between the front side disk-shaped spring and the coil side yoke material, 213 is a rear side connecting member between the rear side disk-shaped spring and the coil side yoke material, 214 is a movable side electrode, and 215 is a joint portion between the movable side electrode and the front side disk-shaped spring 210.

[0071] A magnetic air gap portion 216 in the radial direction is formed between the outer peripheral portion of the pole piece section 202 and the inner peripheral portion of the coil side yoke material 205. 216a is a permanent magnet side air gap portion, and 216b is a yoke material side air gap portion. A closed-loop magnetic circuit B M is formed by "permanent magnet 201 → pole piece section 202 → magnetic air gap portion 216 → coil side yoke material 205 → permanent magnet side yoke material 204". When an electric current flows through the force coil 206a disposed in the space of the magnetic air gap portion 216, a Lorentz force is generated to move the movable side electrode 214 in the axial direction.

[0072] [2-2] Displacement detection section In the displacement detection section (two-dot chain line BB section) in Fig. 2(b), 217 is a fixed side electrode, 218 is an insulating ring, and 219 is a fixed ring. The fixed side electrode 217 is held against the fixed ring with the insulating ring interposed therebetween. 218 is an adhesive application portion for fixing the fixed ring 219 to the coil side yoke material 205. 220 is a gap portion between the fixed side electrode and the movable side electrode. At the assembly adjustment stage, the position of the fixed ring 219 with respect to the coil side yoke material 205 is adjusted so that the gap d of the gap portion 220 becomes the target value.

[0073] On the opposing surface of the fixed-side electrode 217 with respect to the movable-side electrode 214, a ring-shaped groove and a through-hole are formed. 221 is the central through-hole, 223 is the first ring-shaped groove, and 224 is the second ring-shaped groove. The former first ring-shaped groove and the second ring-shaped groove are formed to reduce the dynamic pressure (squeeze pressure) of the air viscous fluid generated between the electrodes.

[0074] Hereinafter, the grooves (the first and second ring-shaped grooves) formed on the electrode surface to reduce the squeeze pressure will be collectively referred to as micro-grooves. 223a, 223b, 223c, and 223d are through-holes formed inside the first ring-shaped groove 223. 224a, 224b, 224c, and 224d are through-holes formed inside the second ring-shaped groove 224. 225 is a recess open to the atmosphere.

[0075] With the first and second ring-shaped grooves and the through-holes communicating with the atmosphere formed in these grooves, the electrode surface having a large area is divided into a plurality of independent electrode surfaces. The electrode surface between the central through-hole 221 and the first ring-shaped groove 223 will be referred to as the first electrode 217a, the electrode surface between the first ring-shaped groove 223 and the second ring-shaped groove 224 will be referred to as the second electrode 217b, and the electrode surface between the second ring-shaped groove 224 and the outer peripheral portion of the fixed-side electrode 217 will be referred to as the third electrode 217c. The above three independent electrode surfaces 217a, 217b, and 217c may be configured, for example, by a method such as mounting three ring-shaped members on a flat plate. However, in the sensor structure of this embodiment, the method of forming by turning, etching, etc. was easy. For the formation of a plurality of electrodes, the best manufacturing method may be selected according to the form of the sensor.

[0076] In this embodiment, the groove widths of the first ring-shaped groove and the second ring-shaped groove are h G = 0.1 mm, and as described above, four through-holes communicating with the atmosphere are formed in each groove. Since the groove depth is formed sufficiently deep compared to the assumed electrode gap h0 (10 to 50 μm), each groove portion can maintain atmospheric pressure regardless of the size of the electrode gap h0.

[0077] In this embodiment, the central through-hole has a diameter of ΦD1 = 0.8 mm, and the outer diameter of the electrode is ΦD2 = 13 mm. The inner circumferential side of the first ring-shaped groove 223 is formed at a position with a radius r1 = 2.1 mm, and the inner circumferential side of the second ring-shaped groove 224 is formed at a position with a radius r2 = 4.5 mm. The total groove area of the first and second ring-shaped grooves 223 and 224 is S m = 6.15 mm 2 , and the electrode area when the above two grooves are not present is S T = 132 mm 2 . Therefore, the ratio of the area of the ring-shaped grooves 223 and 224 to the total electrode area is 4.66%. In summary, the reduction in capacitance due to the ring-shaped grooves is about 5%. Therefore, if the outer diameter of the electrode is slightly increased, in the case of this embodiment, from ΦD2 = 13 mm to 13.2 mm, the reduction in capacitance can be compensated for.

[0078] [2-3] Control Circuit Section The two-dot chain line CC section in Fig. 2(b) shows an overview of the control circuit section. 226 is a displacement detector, 227 is an electrical damping circuit, and 228 is a proportional amplification circuit. Since the capacitance C is determined by the gap of the gap portion 220 formed by the movable-side electrode 214 and the fixed-side electrode 217, by measuring the capacitance C, the relative displacement, which is the difference between the ground motion absolute displacement U and the absolute displacement X of the mass body, can be detected. The current i0 of the actuator is controlled through the electrical damping circuit 227 and the proportional amplification circuit 228 so that the relative displacement U-X becomes zero. By detecting the current i0 flowing through the force coil 206a, the acceleration acting on the movable part can be obtained.

[0079] [3] Theoretical Analysis This embodiment has found the effects by combining the following (1) and (2). (1) Micro-grooves (flow channels) connecting the gap portion to the outside are formed on the relative movement surfaces of the two electrodes so as to reduce the damping action due to the squeeze fluid pressure generated in the gap portion of the displacement detection section. (2) An attenuation means by an electric circuit is provided in the servo amplifier so as to compensate for the reduction in the damping action due to the micro-grooves. In order to specifically and quantitatively grasp the operations and effects of (1) and (2) above, theoretical analysis will be performed below.

[0080] [3-1] Viscous Fluid Analysis of Displacement Detection Unit When a viscous fluid (air) is interposed in a narrow gap (electrode gap) between opposing planes and the gap between the gaps changes steeply with time, a dynamic fluid pressure (squeeze pressure) due to the viscosity of the air is generated. The influence of this squeeze pressure on the function as an electrostatic acceleration sensor will be elucidated by solving the following Reynolds equation.

[0081] [Number]

[0082] (1) Assumption of Sine Wave Input Waveform Figure 3 shows the input waveform applied to the electrode gap. Assume that the electrode gap vibrates in a sine wave. Here, let the vibration center value be h0 = 0.030 mm, the vibration amplitude Δh = 0.001 mm, and the frequency f = 200 Hz.

[0083] [Number]

[0084] (2) Radial Distribution of Squeeze Pressure Figure 4 shows the squeeze pressure at the time of maximum load generation with respect to the radial position of the electrode. A comparison is made between a conventional type without forming a microgroove and the present invention with a microgroove formed. The boundary conditions for numerical analysis are at the positions of the outer periphery of the central through hole r = 0.4 mm and the outer periphery of the electrode r = 6.5 mm, and the pressure (gauge pressure) is atmospheric pressure P = 0. In the case of the conventional type, at the position of r = 2.5 mm, the squeeze pressure shows a maximum value and P max = 47 Pa.

[0085] In the case of the present invention in which microgrooves are formed, the pressure becomes atmospheric pressure P = 0 at the location where the first ring-shaped groove 223 is formed (radius r1 = 2.1 mm) and at the location where the second ring-shaped groove 224 is formed (radius r2 = 4.5 mm). As is clear from comparing the two, the generated pressure is significantly reduced by the formation of the microgrooves.

[0086] (3) Sine wave response of the generated load Fig. 5 shows the sine wave response of the generated load applied to the electrode. A comparison is made between the conventional type without microgrooves and the present invention with microgrooves formed. By forming the microgrooves, the maximum value of the generated load is reduced to approximately 1 / 10.

[0087] (4) Relationship between the attenuation coefficient and the electrode gap Fig. 6 shows the relationship between the attenuation coefficient and the electrode gap. In the same graph, the conventional type without microgrooves (dashed-dotted line) and the present invention with microgrooves formed (dashed line) are shown in contrast. Here, the mechanical attenuation of the electrode with microgrooves is C M , and the electrical attenuation provided in the electric circuit of the servo amplifier is C E . Let them be so. The mechanical attenuation C M and the electrical attenuation C E are added to obtain the hybrid attenuation C T = C M + C E , which is shown by the solid line.

[0088] The value of the electrical attenuation C E is determined by the electric circuit (described later) regardless of the electrode gap h0. The hybrid attenuation C T is set such that at the vibration center value (electrode gap) h0 = 30 μm, the mechanical attenuation C M = 0.393 Ns / m, the electrical attenuation C E = 3.46 Ns / m, and C T = C M + C E = 3.85 Ns / m. The above C TThe value is consistent with that of the conventional formula (dashed line). While the attenuation coefficient of the conventional formula depends greatly on the gap h0, the hybrid attenuation C T maintains an almost constant value when h0 > 20 μm.

[0089] [3-2] Theoretical Analysis of Control Circuit Using the results obtained from the viscous fluid analysis of the displacement detection unit, an analysis is performed to obtain the dynamic characteristics of the servo-type acceleration sensor. FIG. 7 is a control block diagram including the displacement detection unit and the control circuit unit of the present embodiment.

[0090] (1) Equation of Motion of Movable Body When the absolute displacement of ground motion is U(s), the displacement of the movable body (movable-side electrode of the sensor) is X(s), and the relative displacement is U(s)-X(s), and the driving force of the actuator is F(s), the equation of motion is as follows.

[0091] [Equation]

[0092] In the above equation, the Laplace operator s = d / dt, s 2 = d 2 / dt 2 is. The driving force F(s) of the actuator is

[0093] [Equation]

[0094] In Equations (10) and (11), k is the support spring stiffness, C M is the mechanical attenuation coefficient determined by the groove shape formed on the electrode, K P is the proportional gain, and C E is the electrical attenuation coefficient.

[0095] (2) Transfer Function of Movable Body Displacement with Respect to Relative Displacement The transfer function of the movable body displacement X(s) with respect to the relative displacement E(s) [= U(s)-X(s)] is

[0096] [Number]

[0097] [Number]

[0098] [Number]

[0099] If the input acceleration to be detected is Λ, then U = Λ / s 2 as

[0100] [Number]

[0101] (3) Proposal of Sensor Output Derivation Method Here, a method is proposed to extract the acceleration sensor output signal from the sum of the signal outputs of two circuits: a proportional amplification circuit and an electrical attenuation circuit. That is, if the sensor output Z = "proportional amplification circuit output + electrical attenuation circuit output", then point A in the control block diagram of Figure 7 becomes the extraction point of the sensor output (1). In this case, it has been found that a dramatic improvement effect on the dynamic characteristics of the sensor can be obtained. The theoretical basis is shown below.

[0102] [Number]

[0103] Substituting U and X in Equations (14) and (15), the mechanical damping C M and the electrical damping C E are in the denominator, and only the electrical damping C E is in the numerator, and the following sensor signal output Z is obtained.

[0104] [Number]

[0105] As described above, the characteristics of the sensor of the present invention obtained by formula (17) are as follows: (1) Microgrooves are formed on the electrode surface. (2) "Attenuation means by an electric circuit" is provided to compensate for the reduction in the attenuation effect by the above (1). The combination of the above (1) and (2) makes it achievable. The denominator of the transfer characteristic contains a differential (C M +C E )s that delays the phase, and the numerator contains a differential C E s that advances the phase.

[0106] (4) Transfer functions of the conventional sensor and the sensor of the present invention (ideal) (i) Sensor output of the conventional sensor

[0107]

Equation

[0108] E The above formula represents the transfer characteristics of a well-known second-order lag element, which is the characteristic of a conventional sensor in which microgrooves are not formed on the electrodes and the electrical attenuation C

[0109] (ii) Sensor output of the sensor of the present invention (ideal)

[0110]

Equation

[0111] The above formula is the ideal sensor output obtained by forming microgrooves on the electrodes, setting C M = 0 in formula (17), and introducing electrical attenuation C E into the servo amplifier. However, since the mechanical attenuation does not completely become C M = 0, the sensor output is obtained by giving mechanical attenuation C M (≈0) and electrical attenuation C E in formula (17).

[0112] [4] Analysis Results of Sensor Dynamic Characteristics The following is about the case where the sensor of the present invention is configured with the specifications in Table 1. Based on the comparison with the conventional sensor, the gain-phase characteristics are obtained. As described above, the items for performance evaluation as a servo-type acceleration sensor are: (1) the peak value at the resonance point, (2) responsiveness: the phase delay at f = 100 Hz, and (3) sensor sensitivity. These are the above three points.

[0113]

Table 1

[0114] (1) Comparison between the Sensor of the Present Invention and the Conventional Sensor The graph in Fig. 8 compares the sensor of the present invention configured in Case 2 of Table 1 with the conventional sensor when the electrode gap h0 = 30 μm. In the specifications of Table 1, the mechanical damping coefficient C of the conventional sensor M = 3.85 Ns / m, while for the sensor of the present invention, since the microgrooves are formed on the electrodes, C is reduced to 1 / 10 times, M = 0.393 Ns / m.

[0115] However, the sensor of the present invention sets the electrical damping coefficient C E = 3.461×2 Ns / m to compensate for the reduction in the damping effect by the microgrooves.

[0116] From the graph in Fig. 8, for both the sensor of the present invention and the conventional sensor, the peak value at the resonance point ≒ 0. However, the phase delay at f = 100 Hz is Δθ = -2.25 deg for the sensor of the present invention, compared with Δθ = -14.3 deg for the conventional sensor. Therefore, it can be seen that the sensor of the present invention satisfies the ideal conditions in both a. the peak value at the resonance point and b. responsiveness: the phase delay at f = 100 Hz.

[0117] Here, assume a case where the sensor output (2) is taken out from point B in the control block diagram of FIG. 7. That is, the sensor output Z = the output of the proportional amplification circuit. In this case, the effect of improving the dynamic characteristics as described above cannot be obtained, and the electrical attenuation C provided in the control circuit E has characteristics that only replace the mechanical attenuation C M . That is, it has the same characteristics as the conventional sensor in the graph of FIG. 8.

[0118] (2) Comparison of the sensor of the present invention: the electrode gap h0 is 20 to 40 μm The graph of FIG. 9 compares the gain-phase characteristics of the sensor of the present invention configured in Case 2 of Table 1 when the electrode gap h0 is 20 to 40 μm. The value of the mechanical attenuation coefficient C M at each gap h0 is shown in the graph of FIG. 9. The electrical attenuation coefficient is C E = 3.46 × 2. Here, the characteristics of the sensor of the present invention (FIG. 9) and the conventional sensor (FIG. 34) are compared from three viewpoints: (1) the peak value at the resonance point, (2) the responsiveness: the phase delay at f = 100 Hz, and (3) the sensor sensitivity. When the gap h0 is changed in the range of 20 to 40 μm, in the conventional sensor (FIG. 34), as described above, the above (1) to (3) are in a trade-off relationship. However, in the case of the sensor of the present invention (FIG. 9), even when the gap h0 is changed in the range of 20 to 40 μm, the above (1) and (2) hardly change. The sensor sensitivity (capacitance) of (3) is inversely proportional to the gap h0 (= d) as shown in Equation (7). Therefore, in the sensor of the present invention, the trade-off relationship of the above (1) to (3) is resolved, and each specification can be set independently. In summary, the present invention can simultaneously realize ideal sensor dynamic characteristics and improved sensor sensitivity.

[0119] Furthermore, in the mass production process of conventional acceleration sensors, "how to adjust the electrode gap h0 with high precision" has been an important issue. The reason is that, as shown in the graph of FIG. 35, a slight variation in the electrode gap h0 during assembly has a great influence on the sensor dynamic characteristics. With the sensor of the present invention, the assembly accuracy during mass production can be significantly relaxed.

[0120] (3) The sensor of the present invention: Electrical attenuation coefficient C E Comparison when changing The graph in Figure 10 compares the gain and phase characteristics of the sensors of the present invention configured in Cases 1 to 3 of Table 1. The mechanical damping in Case 1 is C M =0.393Ns / m, electrical damping is C E = 3.46 Ns / m, and hybrid damping C T =C M +C E = 3.85 Ns / m. Therefore, the mechanical damping C M (=3.85 Ns / m) and the above C T In the above Case 1, the resonance peak value near the resonance point f=400 Hz is 2.5 dB.

[0121] Mechanical damping above C M and the electrical decay C E Increase C E = 3.46 × 2Ns / m (Case 2 in Table 1), and C E = 3.46 × 3 Ns / m (Case 3 in Table 1), the resonance peak value near the resonance point f = 400 Hz is reduced to 0. The following points can be seen from the above results. (i) Mechanical damping C without microgrooves M0 Let's say. (ii) Mechanical damping C when microgrooves are formed MG Let's say. (iii)C E0 +C MG =C M0 So that the electrical damping C E0 Decide. (iv) Electrical decay C E and C in (iii) above. E0 If you set it to be more than twice the The peak value can be reduced. Therefore, the electrical decay C EBy obtaining this, more ideal gain-phase characteristics of the sensor can be obtained.

[0122] (4) Regarding other damping effects In the acceleration sensor, there are damping elements in addition to the viscous fluid of air intervening between the electrode gaps. For example, in Fig. 2(b), when a conductor (aluminum) is used for the coil bobbin 207, damping C ME is generated by eddy currents. This damping C ME does not depend on the electrode gap.

[0123] Damping C ME is sufficiently smaller than the damping by the viscous fluid in the electrode gap and is usually a negligible value. When it is difficult to theoretically obtain damping C ME , for example, the actuator can be driven with the electrodes removed, and C ME can be measured from its dynamic characteristics.

[0124] However, damping C ME does not provide the effect of improving the dynamic characteristics as in the case of extracting the sensor output (1) at point A in the control block diagram of Fig. 7. Therefore, the dynamic characteristics of the sensor of the present invention are equivalent to the case where the measured value C ME is added with the air viscous damping C M0 to obtain the mechanical damping C M (=C M0 +C ME ).

[0125] [5] General evaluation of the present invention and the conventional formula Hereinafter, in order to comprehensively evaluate the characteristics and effects of the present invention on the basis of comparison with the conventional sensor, the axis of the frequency f is made dimensionless, and the damping coefficient C, the inertial mass m, and the proportional gain K are replaced with the mechanical damping ratio ζ M and the electrical damping ratio ζ E which should also be called representative physical quantities and sorted out. The transfer function G (=Z / Λ) in Equation (17) is

[0126]

Equation

[0127] Here, let the electrical gain be K P , and the mechanical spring stiffness be k. Usually, since K P ≫k, the resonance frequency ω n can be approximated by the following equation.

[0128]

Equation

[0129] The mechanical damping ratio ζ M and the electrical damping ratio ζ E are given by the following equation.

[0130]

Equation

[0131]

Equation

[0132] In the above equation, when the value of the mechanical spring stiffness k cannot be ignored, let K P + k → K P , and the proportional gain K P can be defined as follows. Next, for the case where the frequency axis is made dimensionless, (i) When no microgrooves are formed and the electrical damping ratio ζ E = 0 (ii) When ζ M = 0 due to the formation of microgrooves and the electrical damping ratio ζ E is set Obtain the gain-phase characteristics of the above (i) and (ii). Fig. 11 shows the characteristics of the above (i), and in Equation (15), it is the case where ζ E = 0. Fig. 12 shows the characteristics of the above (ii), and in Equation (15), it is the case where ζ M = 0. Fig. 13 shows, in Fig. 11, 0.1 < ω / ω nIt shows phase characteristics limited to the range of <0.4. FIG. 14 shows, in FIG. 12, 0.1 <ω / ω n It shows phase characteristics limited to the range of <0.4.

[0133] From the gain-phase characteristics of (i) and (ii) above, the following conditions required for the servo type acceleration sensor are evaluated. That is, (1) The peak value at the resonance point is 10 dB or less (2) The phase delay Δθ at ω / ω n = 0.2 is 10 deg or less The ζ E , ζ M range that satisfies the above conditions (1) and (2) is obtained. The evaluation index in the above (2) was set under the following assumption. When the upper limit value of the resonance frequency of the servo type acceleration sensor is f n = 500 Hz, f = 100 Hz, that is, the phase delay at ω / ω n = 0.2 is used as the evaluation index. Since the achievable resonance frequency of an actual servo type acceleration sensor is f n = 500 Hz or less, the above (2) is a sufficient condition for satisfying the ideal sensor dynamic characteristics.

[0134] The ζ M , ζ E conditions for satisfying the above (1) are, from the graphs of FIGS. 11 and 12, ζ M ≧ 0.2, ζ E ≧ 0.2. That is, regardless of electrical and mechanical damping, the peak value at the resonance point is determined only by the magnitude of the damping. The reason is that the peak value at the resonance point is the magnitude of the damping that causes phase delay, that is, the damping term in the denominator of the transfer function in Equation (17) (ζ M + ζ E )ω n . Therefore, the condition for satisfying the above (1) is as follows.

[0135]

Equation

[0136] ζ that satisfies the phase delay condition of the above (2) M , ζ E The condition of is as follows. From the graph of Fig. 14, the electrical damping ratio ζ E has little influence on the phase delay. The mechanical damping ratio ζ E From the graph of Fig. 13, the condition that the phase delay at ω / ω n = 0.2 is 15 deg or less is

[0137]

Equation

[0138] The reason why the electrical damping ratio ζ E does not affect the phase delay condition is that the damping ratio ζ E has the effect of advancing the phase while delaying the phase.

[0139] In Eqs. (24) and (25), if m = K P / ω 2 n Then, ζ M = C M ω n / 2K P , ζ E = C E ω n / 2K P It may be evaluated as. When the mass m cannot be easily obtained as in the rocking motion type sensor described later, the above equation may be used.

[0140] [Second Embodiment] (MM-Type Accelerometer) FIG. 15 shows an example of the MM type servo acceleration sensor according to Embodiment 2 of the present invention. By doubling the electrode area compared to the first embodiment, the sensor sensitivity is improved. As described above, by applying the present invention, the sensor sensitivity determined by the electrode area and the electrode gap, which was a conventional trade-off relationship, and the sensor dynamic characteristics depending on the attenuation characteristics can be set independently. However, when the conventional MC type sensor (FIG. 33) is applied to the present invention, the following points have become problems in practical use. That is, when the diameter of the movable side electrode is increased in order to increase the capacitance, that is, when a movable electrode with a large mass is arranged at the shaft end, the dynamic stability of the sensor body. Therefore, attention was paid to the MM type (Moving Magnet type) as a sensor structure that can achieve the following. (1) Both ends of the shaft have an open structure, and the support span L of the movable part can be set large. (2) It is not necessary to divide and cut the disk-shaped spring that supports the movable part. Therefore, the torsional rigidity can be set large. The torsional rigidity is the rigidity in the X direction when the moving direction of the movable part is the X axis. θ It is the rigidity in the direction. According to the above (1) and (2), the MM type can have a movable part support structure that is dynamically more stable than the MC type.

[0141] [1] About the MM type servo acceleration sensor The above-described servo acceleration sensor (FIG. 33) had problems in the production technology aspect related to the yield and reliability during mass production. That problem is due to the basic operating principle of the moving coil type (hereinafter referred to as the MC type) in which the coils (16a, 16b), which are movable parts, move, and thus a sensor signal must be transmitted and received between the coils and the fixed side. The conduction paths through which a plurality of signals connecting the movable part and the fixed side flow must be formed using elastic members (disk-shaped springs 20, 21) that connect the two. As a result, (1) division by cutting the disk-shaped spring for forming a plurality of conduction paths, (2) insulation of the signal lines, (3) a complicated production method involving a soldering process for extremely thin wires, etc. are required, which has been a major factor in reducing the yield and reliability during mass production.

[0142] Now, the present inventors have proposed a moving magnet type (MM type) servo acceleration sensor that does not require fine wire processing by fixing the coil and moving the permanent magnet, and it is currently under application. However, the MM type servo acceleration sensor has no precedent in the past. It is considered that there was an implicit premise (blind spot), which could be called a fixed idea, that "since the inertial mass of the moving part increases in the MM type, the transmission characteristics and responsiveness in the high frequency range are disadvantageous." The previously proposed invention has overcome this "blind spot" by the following devices. That is, (i) Magnetic circuit configuration for reducing the weight of the moving part (ii) Magnetic pole shape for reducing the influence of leakage magnetic flux (iii) By devising the electromagnetic structure that can increase the coil volume, a coil specification (number of turns, wire diameter) that can achieve both increased generating force and heat generation suppression is found. By such means, the weaknesses of the MM type were eliminated, and by taking advantage of the characteristics of the MM type, sensor performance far superior to the MC type could be realized. The weakness of the MM type, compared with the MC type, that the inertial mass of the moving part is large is also an advantage that the moving part is easily movable even with a minute acceleration. As a result of eliminating this weakness, it has become an advantage that the sensor sensitivity, especially in the low frequency range, is improved. Also, as described above, since the MM type can have a more dynamically stable moving part support structure compared with the MC type, the outer diameter of the electrode can be increased and the sensor output can be increased. Hereinafter, the specific structure and the structural features of the present embodiment will be described in [2] below.

[0143] [2] Description of the Second Embodiment Fig. 15a is a view taken along the DD arrow of Fig. 15b, and Fig. 15b is a front sectional view of the sensor body. Fig. 15c is an external view of the front side disk, and Fig. 15d is an external view of the rear side permanent magnet.

[0144] The dashed line AA in Fig. 15b is a moving magnet type (MM type) actuator section that drives the movable section in the axial direction. The dashed line BB section indicates a displacement detection section that detects capacitance. The dashed line CC section outlines a control circuit section incorporating a servo amplifier that drives the actuator. As described above, the MM type servo acceleration sensor is currently under application by the present inventors, and its actuator section has been further improved. Hereinafter, the specific structure of this embodiment will be described separately for the actuator section, displacement detection section, and control circuit section.

[0145] [2-1] Actuator section In the actuator section (the section indicated by the dashed line AA in Fig. 15b), 101 is the front permanent magnet, 102 is the rear permanent magnet, and 103 is the pole piece section (movable side yoke material). As shown in Fig. 15d, both the front permanent magnet and the rear permanent magnet are composed of a plurality of segment type permanent magnets magnetized in the radial direction, and are mounted on the pole piece section 103. The radial magnetization directions of the front permanent magnet 101 and the rear permanent magnet 102 are opposite. 104 and 105 are cylindrical voids formed on the left and right of the pole piece section to reduce the weight of the pole piece section 103. 106 is the front spiral disk spring (hereinafter referred to as the front disk), and 107 is the rear spiral disk spring (hereinafter referred to as the rear disk). The distance between the front disk and the rear disk is the support span L. The magnitude of this support span L is sufficiently large compared to the conventional MC type sensor (Fig. 33). As shown in the external view of Fig. 15c, the front disk is formed by a peak portion 106a and a groove portion 106b, and the same applies to the rear disk. As will be described later, the front disk 106 also serves as a signal transmission path for propagating the electrical signal of the fixed side electrode to a control circuit installed outside. In this embodiment, as well as in the embodiments described later, the shape of the spring is not limited to this spiral curve. From the characteristics required for the acceleration sensor, a spring structure and specifications that can obtain low rigidity and low resonance frequency may be selected. For example, well-known arc springs and the like can also be applied. 108 is the movable side electrode supported by the front disk, 109 is the coil side yoke material, 110 is the coil bobbin, 111 is the front force coil, 112 is the rear force coil, and 113 is the coil bobbin fastening bolt (indicated by the imaginary line). The winding directions of the front force coil 111 and the rear force coil 112 are opposite. 114 is the magnetic void formed between the inner peripheral surface of the coil bobbin 110 and the two permanent magnets, 114a is the front magnetic void, and 114b is the rear magnetic void.A closed-loop magnetic circuit B is formed by "permanent magnet 102 → rear magnetic air gap 114b → coil-side yoke material 109 → front magnetic air gap 114a → permanent magnet 101 → pole piece part 103 → permanent magnet 102", as indicated by the dashed arrow. M The front disk 106 also serves as a support for the movable part and a conduction path for detecting capacitance. In order to detect a minute capacitance signal between the movable-side electrode 108 and a fixed-side electrode (described later), complete electrical insulation is achieved for the conduction path connecting the movable-side electrode 108 to the outside. That is, the front disk 106 is fastened to the coil-side yoke material 109 by bolts 116 with an outer peripheral ring 115 interposed therebetween. The outer peripheral ring 115 is made of a non-conductive material, and the front disk 106 and the outer peripheral ring 115 are pre-fixed by an adhesive. Also, by making the hole diameter for the bolts formed in the front disk 106 larger than the bolt diameter, the bolts 116 and the front disk 106 are electrically insulated. The movable-side electrode 108 is adhesively fixed to the front disk 106 at the joint 117. Also, the central part of the movable-side electrode 108 is adhesively fixed to the end face of the pole piece part via a non-conductive material 118. Eddy currents are generated in the pole piece part and the coil-side yoke material, but due to this electrical insulation measure (non-conductive materials 115, 118), the capacitance signal between the two electrodes (movable side and fixed side) can avoid the influence of these eddy currents. As the non-conductive material, mica (muscovite), porcelain (ceramics), glass, polyimide (super engineering plastic), etc., which are inorganic solid insulation materials, can also be applied. In order to control the current flowing through the two force coils 111, 112, the lead wires of these coils penetrate the coil-side yoke material 109 and are connected to a control circuit installed outside (not shown).

[0146] The movable part of this embodiment is composed of two permanent magnets 101 and 102, a pole piece part 103, and a movable-side electrode 108. The fixed part is composed of a coil bobbin 110 in which each coil is housed and a coil-side yoke material 109. In the acceleration sensor of this embodiment, the driving means of the movable part uses the proposed moving magnet type. As is well known, when an electric current flows through a conductor placed in a magnetic field, a Lorentz force, which is an electromagnetic force, is generated. Regardless of the type of driving principle, the force relationship between the fixed side and the moving side of any actuator is relative. That is, if either the fixed side or the moving side is fixed, the other side moves. In this embodiment, when an electric current flows through the force coils 111 and 112 housed in the coil bobbin 110, a reaction force of the Lorentz force that moves the movable part in the axial direction is generated.

[0147] [2-2] Displacement detection unit

[0148] In the displacement detection unit (dotted line BB part) in Fig. 15b, 119 is a fixed-side electrode, 120 is an insulating ring, and 121 is a fixed ring. The fixed-side electrode 119 is held against the fixed ring with the insulating ring interposed therebetween. 122 is an adhesive application part for fixing the fixed ring 121 to the coil-side yoke material 109. 123 is a gap between the fixed-side electrode 119 and the movable-side electrode 108. At the assembly adjustment stage, the position of the fixed ring 121 with respect to the coil-side yoke material 109 is adjusted so that the gap h0 of the gap part 123 becomes the target value.

[0149] On the fixed-side electrode 119 which is the opposing surface of the movable-side electrode 108, two rows of ring-shaped grooves and a plurality of through holes are formed in the same manner as in the first embodiment. 124 is a central through hole, 125 is the first ring-shaped groove, and 126a, 126b, 126c, 126d are through holes formed inside the first ring-shaped groove. 127 is the second ring-shaped groove, and 128a, 128b, 128c, 128d are through holes formed inside the second ring-shaped groove.

[0150] Furthermore, on the fixed-side electrode surface, in addition to the above two rows of circumferential grooves (ring-shaped grooves), cross-shaped radial grooves are formed centering around the central through-hole 124. In the electrode surface of the present embodiment, microgrooves are formed by both the circumferential grooves and the radial grooves in order to reduce the increase in the attenuation effect due to the squeeze pressure associated with the increase in the outer diameter of the electrode UP. 129a is the first radial groove, 129b is the second radial groove, 129c is the third radial groove, and 129d is the fourth radial groove.

[0151] [Supplementary Note: Capacitance reduction due to microgrooves] The groove widths of the first ring-shaped groove, the second ring-shaped groove, and the radial grooves are h G = 0.2 mm. The central through-hole ΦD1 = 1.0 mm, the outer diameter of the electrode is ΦD2 = 18.4 mm, the inner peripheral side of the first ring-shaped groove 125 is formed at a position with a radius r1 = 3.0 mm, and the inner peripheral side of the second ring-shaped groove 127 is formed at a position with a radius r2 = 6.0 mm. The total groove area of the first and second ring-shaped grooves 125 and 127 is S m1 = 11.3 mm 2 The total groove area of the four radial grooves 129a, 129b, 129c, and 129d is S m2 = 7.36 mm 2 The electrode area when there are no such grooves is S T = 266 mm 2 Therefore, the ratio of the area of the grooves to the total electrode area is 7.01%. In summary, the capacitance reduction due to the ring-shaped grooves and the radial grooves is about 7%.

[0152] [2-3] Control circuit section The dashed line CC part in FIG. 15b shows the outline of the control circuit section. 130 is a displacement detector, 131 is an electrical attenuation circuit, and 132 is a proportional amplification circuit. Since the capacitance C is determined by the gap h0 of the gap portion 123 formed by the movable-side electrode 108 and the fixed-side electrode 119, by measuring the capacitance C, it is possible to detect the relative displacement, which is the difference between the ground motion absolute displacement U and the absolute displacement X of the mass body, in the same manner as in the first embodiment. The current i0 of the actuator is controlled through the electrical attenuation circuit 131 and the proportional amplification circuit 132 so that the relative displacement U−X becomes zero. By detecting the current i0 flowing through the force coils 111 and 112, the acceleration acting on the movable part can be obtained.

[0153] [3] Characteristic analysis results Regarding the case where the sensor of the present invention according to the MM formula (with grooves on the electrodes) is configured with the specifications in Tables 2 and 3, the gain-phase characteristics are obtained on the basis of comparison with the conventional electrode structure (without grooves on the electrodes). Here, the conventional sensor compared with the present invention in the first embodiment is defined as the conventional reference sensor. FIG. 16 shows the conventional reference sensor, and FIG. 17 shows the MM formula sensor to which the present invention is applied, and only the electrode parts are shown in both figures. The outer diameter of the electrode of the conventional reference sensor in FIG. 16 is d1, and the gap between the electrodes is h 01 is. The outer diameter of the electrode of the sensor of the present invention in FIG. 17 is d2, and the gap between the electrodes is h 02 is. In order to improve the sensor sensitivity, (1) The electrode area is twice that of the reference sensor + the gap between the electrodes is the same (2) The electrode area is twice that of the reference sensor + the gap between the electrodes is 1 / 2 For each of the above cases (1) and (2), the influence on the sensor dynamic characteristics is evaluated. As described above, the evaluation items of the dynamic characteristics are (1) the peak value at the resonance point, (2) the responsiveness: the phase delay at f = 100 Hz, and (3) the sensor sensitivity, which are the above three points.

[0154] [3-1] Gap h between electrodes 02 = 30 μm (sensor sensitivity doubled) Figure 18 shows the gain-phase characteristics of the MM type sensor configured under the conditions of Table 2, comparing the case without a flow channel (dashed line) and the case with a flow channel (solid line: the present invention). In both cases, the electrode area is twice that of the conventional reference sensor. The characteristics of the conventional reference sensor (dotted line) are shown in the same graph.

[0155] From the graph of Figure 18, for both the MM type including the present invention and the conventional reference sensor, the peak value at the resonance point is approximately 0. However, the phase delay at f = 100 Hz is Δθ = -14.3 deg for the MM type without a flow channel and Δθ = -14.2 deg for the conventional reference sensor. In contrast, for the sensor of the present invention, Δθ = -2.00 deg. Therefore, even though the electrode area of the sensor of the present invention is increased by a factor of 2, a. the peak value at the resonance point, b. the responsiveness: the phase delay at f = 100 Hz, and (3) the sensor sensitivity all satisfy the ideal conditions.

[0156]

Table 2

[0157] [3-2] Electrode gap h 02 = 15 μm case (sensor sensitivity 4 times) Figure 19 shows the gain-phase characteristics of the MM type sensor configured under the conditions of Table 3, comparing the case with a flow channel (solid line: the present invention) and the case without a flow channel (dashed line). For both MM types, the electrode area is twice that of the conventional reference sensor and the electrode gap is halved. Therefore, from Equation (7), the sensor sensitivity is 4 times.

[0158] From the graph of Figure 19, compared with the sensor with a flow channel (the present invention), the gain at the resonance point of the sensor without a flow channel has decreased by about -20 dB. Also, the phase delay at f = 100 Hz is Δθ = -9.27 deg for the sensor with a flow channel (the present invention) and Δθ = -69.4 deg for the sensor without a flow channel. The sensor without a flow channel clearly has an overdamping characteristic.

[0159] Therefore, even though the sensor sensitivity of the present invention has been increased by a factor of four, none of a. the peak value at the resonance point, b. the responsiveness: the phase delay at f = 100 Hz, and (3) the sensor sensitivity satisfy the ideal conditions.

[0160]

Table 3

[0161] [3-3] Two-dimensional Viscous Fluid Analysis of Electrodes with "Circumferential Grooves + Radial Grooves" In this embodiment, the numerical analysis results of the two-dimensional viscous fluid in the electrode gap are shown below. FIG. 20 is a two-dimensional analysis model of the electrode surface formed with microgrooves by circumferential grooves and radial grooves. The pressure boundary conditions of the first ring-shaped groove, the second ring-shaped groove, the radial groove, the inside of these grooves, the through-hole 124, and the outer periphery of the electrode 119 are atmospheric pressure. FIG. 21 is the numerical analysis result of the squeeze pressure distribution.

[0162] FIGS. 22 to 24 show the transient response of the squeeze pressure distribution generated in the electrode gap when the electrode gap changes in a sine wave waveform. The vibration center value h0 = 0.015 mm, the displacement amplitude Δh = 0.001 mm, and the frequency f = 200 Hz. In this case, as shown in FIG. 22, the center value V0 of the velocity V is 0, and the velocity amplitude ΔV is 1.25 mm. FIG. 23 extracts the velocity waveform (FIG. 22) in the range of time 0.002 < t < 0.006 seconds.

[0163] FIGS. 24(a) to 24(d) show the squeeze pressure distribution that changes depending on the velocity. When the velocity V = 0, as shown in FIG. 24(a), the generated pressure becomes zero. When the velocity V = V max = 1.25 mm / s, as shown in FIG. 24(d), the generated pressure becomes maximum. The values of the mechanical damping coefficient C in Tables 2 and 3 M are obtained by the above analysis method.

[0164] [Third Embodiment] (Application to Differential Sensors) FIG. 25 shows an example of a differential servo type acceleration sensor according to Embodiment 3 of the present invention. FIG. 25(a) is a view taken in the direction of arrow DD of FIG. 25(b), and FIG. 25(b) is a front sectional view. That is, by paying attention to the structural feature of the linear motion type MM type in which both the left and right output shafts are open ends, and by providing electrodes for detecting capacitance at two locations on the left and right, a differential capacitance type sensor is configured. By making the acceleration sensor differential, as described in [Supplementary Note 2], a high-resolution sensor in which the sensor output is hardly affected by disturbance signals such as noise and drift can be realized.

[0165] In the case of the differential type, the origin position of the movable member is set so that the capacitances of the two sets of electrodes are equal. When the relative displacement of the movable member from the origin position is detected, it is driven by a servo amplifier so that a generating force for returning the movable member to the origin position is generated. Now, the present inventors are proposing a differential acceleration sensor that takes advantage of the characteristics of the linear motion type MM type in which the output shaft is an open end in a previously filed patent application. When the acceleration sensor is made differential, the following problems have been found.

[0166] (1) Problems in terms of performance In the case of a differential type in which two displacement sensors are arranged with the electrode gaps h0 on the left and right under the same conditions, the mechanical damping C of the electrode portion M is doubled. To maintain the sensor dynamic characteristics, that is, (i) reduction of the resonance peak and (ii) improvement of the responsiveness, for example, assume a case where the left and right electrode gaps are increased from h0 = 30 → 40 μm. In this case, from Equation (2), since the damping constant D increases inversely proportional to the cube of the gap h0, the mechanical damping C M can be reduced to about one. However, the sensor sensitivity has to be sacrificed.

[0167] (2) Problems in terms of production In the case of a normal sensor (FIG. 33) in which the electrodes are installed only on one side, the gap δ set through the electrode gap adjustment process X is usually the error Δδ = δ with respect to the target gap δ0 X- It has -δ0. This error is caused by factors such as component processing accuracy, assembly accuracy, and variations in the thickness of the adhesive in the adhesion between components during production. However, this error can be readjusted to the target gap δ0 by passing a bias current I0 through the coil. That is, it can be corrected with a bias current so that the error Δδ → 0.

[0168] Assume that the linear acceleration sensor is made differential, the adjustment of the electrode gap is completed, and the left and right gaps are δ L 、δ R are set. In this case, it is difficult to adjust both of the above gaps δ L 、δ R to the target value δ0 with a bias current I0. Because in the differential type, at the stage when the adjustment of the electrode gap is completed, it is restricted by "δ L +δ R = constant value". Since the electrode gap is on the order of several microns and has a great influence on the sensor dynamic characteristics, it is difficult to maintain the sensor dynamic characteristics of (i) and (ii) above.

[0169] In the case of the present invention where microgrooves are formed on the left and right electrodes, the mechanical damping C M ≈ 0 can be achieved. That is, the size and accuracy of the left and right electrode gaps δ L 、δ R do not affect the sensor dynamic characteristics. The necessary damping effect can be replaced by electrical damping C E , and the above problems (1) and (2) of the differential type with two arranged capacitance sensors are solved at once.

[0170] The applicable object for obtaining the above effects may be in any form as long as it is a differential sensor having two electrodes. In summary, it is sufficient to include two sets of displacement detection parts composed of the movable side electrode and the fixed side electrode, and the gaps of the void parts of the two sets of displacement detection parts are configured to change in opposite phases.

[0171] In this embodiment, a differential type is applied to a linear motion type MM type in which the movable part is an open end. As a result, the following effects are obtained simultaneously: (1) an effect of improving dynamic characteristics by combining a microgroove and electrical attenuation, (2) an improvement in sensitivity in the low frequency range due to an increase in the inertial mass of the MM type, (3) noise and a drift canceller effect due to the differential type, and (4) an improvement in productivity due to relaxation of the gap accuracy between both electrodes. In summary, this embodiment simultaneously has the above characteristics (1) to (4), and realizes a new sensor, which can be called a so-called "ultimate servo type" without precedent in the past.

[0172] (i) Description of the structure Unlike the second embodiment using a radially magnetized permanent magnet, the acceleration sensor of FIG. 25 forms a closed-loop magnetic circuit by arranging an axially magnetized permanent magnet at the center of the pole piece portion. Further, compared with the MM type of the second embodiment, the area of both electrodes is set to be twice as large. Considering a three-fold increase in the inertial mass of the movable part, the sensor sensitivity can be improved six-fold compared with the conventional MC type (FIG. 33).

[0173] 801 is an axially magnetized permanent magnet, 802a is a front-side pole piece portion, 802b is a rear-side pole piece portion, 803 is a coil-side yoke material, 804 is a coil bobbin, 805 is a fastening bolt for this coil bobbin and the coil-side yoke material, 806a is a front-side coil, and 806b is a rear-side coil. 807a and 807b are gap portions formed at the centers of the front-side and rear-side pole piece portions 802a and 802b. 808a is a front-side magnetic gap portion, and 808b is a rear-side magnetic gap portion, each indicating a radial gap between the two pole piece portions and the coil-side yoke material. A closed-loop magnetic circuit B is formed by "permanent magnet 801 → front-side pole piece portion 802a → front-side magnetic gap portion 808a → coil-side yoke material 803 → rear-side magnetic gap portion 808b → permanent magnet 801". Mis formed. 809a is the front-side inner peripheral support member of the pole piece portion, 809b is the rear-side inner peripheral support member, 810a is the front-side disk, and 810b is the rear-side disk. 811a is the front-side movable electrode, 811b is the rear-side movable electrode, 812a is the front-side fixed electrode, 812b is the rear-side fixed electrode, 813a is the front-side insulating ring, 813b is the rear-side insulating ring, 814a is the front-side fastening ring, 814b is the rear-side fastening ring, and 815a and 815b are the fastening bolts for the front-side and rear-side movable-side electrodes 811a, 811b and the inner peripheral support members 809a, 809b. 816a and 816b are the bolts for fastening the two disks 810a, 810b and the coil-side yoke member 803 on the outer peripheral side. The inner peripheral sides of the two disks 810a, 810b are adhesively fixed by the inner peripheral support members 809a, 809b and the support portions 817a, 817b.

[0174] On the front side, the fixed-side electrode 812a, which is the opposing surface of the movable-side electrode 811a, has two rows of ring-shaped grooves, a cross-shaped radial groove, and a plurality of through holes formed with the same specifications as in the second embodiment. The same applies to the fixed-side electrode 812b on the rear side. 818 is the central through hole, 819 is the first ring-shaped groove, and 820a, 820b, 820c, 820d are the through holes formed inside the first ring-shaped groove. 821 is the second ring-shaped groove, and 822a, 822b, 822c, 822d are the through holes formed inside the second ring-shaped groove.

[0175] FIG. 26 is a diagram showing that the MM type servo acceleration sensor applied to the present embodiment can be easily changed to a differential type by simply replacing or attaching the left and right fixed side electrode units with micro grooves. FIG. 26(a) is the front side fixed side electrode unit, FIG. 26(b) is a front cross-sectional view of the MM type sensor without micro grooves in a non-differential type, and FIG. 26(c) is the rear side fixed side electrode unit. On the left side of the sensor body in FIG. 26(b), the front side fixed side electrode unit is composed of a front side fixed electrode 812, a front side insulating ring 813, and a front side fastening ring 814. If this unit is replaced with the front side fixed electrode 812a, the front side insulating ring 813a, and the front side fastening ring 814a shown in FIG. 26(a), it will have micro grooves. Similarly, on the right side of the sensor body where the electrode is not attached, after fastening the rear side movable electrode 811b to the rear side inner peripheral support member 809b, a rear side fixed side electrode unit with micro grooves may be attached.

[0176] Therefore, in the present invention, the selection from "without micro grooves to with micro grooves" and the selection from "non-differential type to differential type" do not result in a large cost increase. Also, the adjustment of the electrode gap performed at the final mass production stage, that is, the adjustment of the gap between the rear side movable side electrode 811b and the rear side fixed side electrode 812b, can be performed independently of the front side. For example, after finishing the adjustment on the front side, the adjustment on the rear side may be performed.

[0177] The noise-drift canceller effect by the differential type sensor of the present embodiment and the effect when applied to an active vibration isolation table will be described in detail in [Supplementary Note 2].

[0178] [Fourth Embodiment] (Application to a Swing Type Sensor)

[0179] The specific structure of the servo accelerometer is roughly divided into two types: (1) a type in which the mass part moves linearly, and (2) a type in which the mass part moves swingingly. All the above-described embodiments of the present invention are cases where the present invention is applied to the linear motion type of (1). However, the same effects can be obtained even when the present invention is applied to the rocking motion type described in (2) above. The rocking motion type servo acceleration sensor is a known one as shown by way of example in Patent Document (4).

[0180] FIG. 27 is a front cross-sectional view of a rocking motion type servo acceleration sensor according to Embodiment 4 of the present invention, FIG. 28 is a view taken along arrow AA in FIG. 27, showing the specific shape of a micro groove attached to the vibrator, FIG. 29 is a view showing a semi-circular arc-shaped ring in which the micro groove is formed, FIG. 29(a) is a top view, and FIG. 29(b) is a cross-sectional view taken along line BB in FIG. 29(a).

[0181] (1) Structure of the rocking motion type sensor 550a is a vibrator and is located within the frame of a disc-shaped frame body 550. The vibrator 550a is formed in a tongue piece shape with a part of its circumference cut out, and is supported by the frame body 550 via a hinge 550b. These frame body 550, vibrator 550a, and hinge 550b are integrally formed of, for example, quartz glass. The hinge 550b is thin and elastically deformable, and the vibrator 550a can be displaced in the vertical direction in the figure by an input acceleration.

[0182] 551 and 552 are a pair of magnetic yokes, 553 is a pole piece bottom, 554 is a permanent magnet, and 555 is a pole piece stop. The permanent magnet 554 is magnetized in its plate thickness direction, and annular magnetic gaps 556 are respectively formed between the inner peripheral surfaces of the open ends of the magnetic yokes 551 and 552 and the outer peripheral surface of the pole piece stop 555. Coil bobbins 558 around which torque coils 557 are wound are respectively attached to both plate surfaces of the vibrator 550a so as to be located within these annular magnetic gaps 556. Capacitive electrodes 550c are respectively formed in an arc shape along the outer periphery of the tip side of the tongue piece shape on both plate surfaces of the vibrator 550a. 551e and 552e are electrode surfaces facing the capacitive electrodes 550c with a predetermined interval therebetween. The permanent magnet, the pole piece stop, the pole piece bottom, the torque coil, etc. are arranged symmetrically in the vertical direction as shown in the figure (detailed description is omitted).

[0183] In a servo-type accelerometer having such a configuration, the displacement of the vibrator 550a due to an acceleration input is detected as a change in the capacitance between the capacitance electrodes 550c and the electrode surfaces 551e, 552e. The electrode surfaces 551e, 552e on the fixed side are set to a common potential, and the detection signals of the capacitance electrodes 550c on both plate surfaces of the vibrator 550a are differentially amplified by a servo amplifier (not shown), and a torque current based on the capacitance difference is passed through a pair of torque coils 557. Due to the interaction between this torque current and the magnetic field by the permanent magnet 554, the displaced vibrator 550a returns to its original position and balances at the neutral point. Since the torque current at this time is proportional to the acceleration applied to the vibrator 550a, the input acceleration can be obtained from this current. The coil terminals of the torque coil 557 are adhered to the metal conductors on the vibrator 550a and are electrically joined.

[0184] The servo amplifier is provided with an electrical damping circuit and a proportional amplification circuit described in the first embodiment of the linear motion type acceleration sensor. This electrical damping circuit is provided to compensate for the reduction of the mechanical damping effect by the micro grooves described later (not shown).

[0185] 559A is a semi-circular ring A attached to the magnetic yoke 552. 559B is a semi-circular ring B attached to the magnetic yoke 551. The semi-circular ring A and the semi-circular ring B are formed with micro grooves for reducing the dynamic pressure (squeeze pressure) due to the viscosity of air. Each semi-circular ring also serves as a fixed-side electrode and is arranged opposite to the movable-side capacitance electrode 550c.

[0186] 560A and 560B are through holes formed in the semi-circular ring A and the semi-circular ring B, respectively. 561A and 561B are through passages formed in the magnetic yokes 551 and 552, respectively. One communicates with the through hole, and the other is open to the atmosphere.

[0187] (2) Structure of the vibrator and the semi-circular ring Figure 28 shows the conductor part formed on the pendulum by imaginary lines, and the shape of the semi-circular arc-shaped ring 559A with a microgroove formed by solid lines. The arc-shaped pendulum conductor A constitutes one input / output end of the torque current. The pendulum conductor B connects the two left and right torque coils 557 in series, and the arc-shaped pendulum conductor C constitutes the other input / output end of the torque current. The capacitance detection electrode D is formed in an arc shape along the outer edge of the pendulum 550a on one surface of the pendulum 550a. Also, the capacitance detection electrode E is formed on the other surface of the pendulum 550a in the same manner as the capacitance detection electrode D. Each end of the capacitance detection electrodes D and E is connected to a servo amplifier (not shown). Each of the above-described pendulum conductors is formed of a thin film in which gold (Au) is sputtered or vacuum-deposited on the surfaces of the frame 550, the pendulum 550a, and the hinge 550b made of quartz glass.

[0188] In FIG. 28, which is a view taken along the arrow AA in FIG. 27, the semi-circular arc-shaped ring 559A is disposed at a position surrounding the outer periphery of the capacitance detection electrode D. FIG. 29 shows a view of the semi-circular arc-shaped ring 559A alone. 562 is a microgroove formed on the surface of the semi-circular arc-shaped ring 559A, and 563 is a flow hole formed inside this microgroove. The flow hole 563 of the present embodiment is formed as a circle having the same diameter as the groove width of the microgroove 562, and the air flow path communicates with the through hole 560A. The flow holes 563 and the through holes 560A having the same shape are formed at three locations in the groove of the microgroove 562. That is, similar to the above-described embodiment, due to the air flow path of "the space inside the microgroove ⇔ the flow hole 563 ⇔ the through hole 560A ⇔ the through passage 561A ⇔ the atmospheric pressure", the pressure in the space inside the microgroove is not affected by the variation of the electrode gap and always maintains the atmospheric pressure. As a result, the mechanical attenuation C M is significantly reduced. In the present embodiment, the groove width of the microgroove 562 is formed to be 0.1 to 0.2 mm. The total area S m of this groove width is sufficiently smaller than the total area S T of the semi-circular arc-shaped ring 559A when the microgroove is not formed, so the influence on the capacitance is negligible.

[0189] In this embodiment, the case where one micro groove (arc-shaped flow channel) is formed in the semi-circular arc-shaped ring is shown. The measures for reducing the squeeze pressure are not limited to this structure. As shown in the example of the linear motion type / servo type acceleration sensor, the shape of the micro groove may be a radial groove. Alternatively, a large number of small-diameter holes may be formed in the semi-circular arc-shaped ring, a space communicating with the atmosphere may be provided on the bottom surface, and this space may be communicated with the small-diameter holes (not shown). When the shape of the micro groove is complicated, the pattern of the micro groove is formed on a thin plate by etching. This thin plate may be attached to the relative moving surface of the movable side electrode and the fixed side electrode (not shown). Any shape, structure, and construction method can be applied as long as the squeeze pressure generated in the gap between the electrodes can be reduced.

[0190] As a structure connecting the micro groove and the atmosphere, in this embodiment, a through passage is formed in the magnetic yoke. Instead of this through passage, a groove (hollow) may be formed along the outer peripheral side of the semi-circular arc-shaped ring. This hollow may be communicated with the atmosphere (not shown). The above-described measures can be applied not only to the linear motion type / oscillatory motion type sensors.

[0191] (3) Effects of this embodiment Even in the case of this embodiment applied to the oscillatory motion type, the same effects as those of the aforementioned linear motion type can be obtained. That is, (1) the effect of improving the dynamic characteristics by the combination of the micro groove and the electrical attenuation, (2) the improvement of productivity by relaxing the gap accuracy between the two electrodes, and the like.

[0192] Furthermore, in the case of the oscillatory motion type, when the oscillation direction of the movable part is set as the Z axis and the X axis orthogonal to this Z axis is set, each component constituting the sensor is not "mirror symmetric" with respect to the central part of the X axis. Also, the electrode shape is not "axisymmetric" with respect to the central axis of the electrode surface formed on the vibrator. Due to this asymmetry, it has been pointed out that the oscillatory motion type is susceptible to the thermal expansion of the members. However, by making the mechanical attenuation C M ≈0 by the micro groove formed on the relative moving surface between the electrodes, the influence of the minute change due to the thermal expansion of the electrode gap on the sensor characteristics can be reduced.

[0193] [Addendum] Brownian noise canceller effect The effect of the present invention described above is that when the present invention is applied to, for example, a servo-type acceleration sensor which is a control element of an active vibration isolation table, ideal sensor dynamic characteristics (gain-phase characteristics) can be obtained. The principle that brings about this effect could be explained within the scope of viscous fluid mechanics, control engineering, and mechanical mechanics. However, when turning to the microscopic world where an explanation based on molecular dynamics is required, it was found that the present invention has effects seen from different aspects. That is, the mechanical damping C M ≒0 can be achieved, yet the effect of the present invention that the sensor sensitivity (capacitance) can be maintained and further improved is expected to be extremely effective also in reducing mechanical noise due to the Brownian motion of gas molecules.

[0194] (1) Mechanical noise due to Brownian motion According to academic literature (1), it is said that the capacitance C that can be set for a capacitance-type sensor is restricted by mechanical noise (Brownian noise) due to the Brownian motion of gas molecules generated from a displacement detector. The magnitude of this noise is obtained by the following formula.

[0195]

Equation

[0196] The damping constant D in Equation (26) is

[0197]

Equation

[0198] Equation (26) is the same as the aforementioned Equation (6). When the electrode area is increased and the gap is narrowed to increase the capacitance, the damping constant D increases in proportion to the square of the electrode area A and inversely proportional to the cube of the gap d. The increase in the damping constant D increases the mechanical noise N in Equation (26). That is, the conventional method of improving the sensor sensitivity has limitations in terms of mechanical noise (Brownian noise) caused by Brownian motion. In terms of improving the sensor performance without being affected by the negative aspects, it is common with the effect (improvement of dynamic characteristics) of the present invention described above. That is, by applying the present invention, the sensor sensitivity can be improved without increasing the Brownian noise.

[0199] (2) Brownian noise reduction effect of the present invention Here, assuming that the damping constant of the reference acceleration sensor is D0 and the mass of the movable part of the sensor is M0, the reference value of the noise power spectrum in Equation (26) is defined as N0.

[0200]

Number

[0201] To compare with a sensor having a different damping constant and movable part mass, the relative ratio η of the noise power spectrum is defined.

[0202]

Number

[0203] Table 4 shows an example of organizing the relative ratio η of the noise power spectra of the conventional sensor and the sensor of the present invention, and the relative ratio of the sensor sensitivity. However, instead of the damping constant D, the attenuation coefficient C with different units is used.

[0204]

Table 4

[0205] [Supplement 1] Drift and noise reduction effect by differential method (1) In the case of a conventional acceleration sensor Hereinafter, the drift and noise reduction effect of the sensor of the present embodiment will be described based on a comparison with the conventional type. FIG. 30 shows the relationship between the electrode output, noise and drift, and the sensor acceleration output in the case of a conventional acceleration sensor (see FIG. 33). The electrode output is obtained by detecting the capacitance determined by the gap between the movable-side electrode 24 and the fixed electrode 25. The graph A of noise and drift is obtained by adding a minute positive bias value to a sine wave. The sensor acceleration output (graph C) is the result of applying noise and drift (graph A) to the electrode output (graph B).

[0206] Furthermore, in the active vibration isolation table, in order to obtain vibration isolation performance in the low frequency range, absolute velocity feedback and absolute displacement feedback are applied. In order to obtain an absolute velocity signal, it is necessary to integrate the acceleration signal once, and in order to obtain an absolute displacement signal, it is necessary to integrate the acceleration signal twice. The graph D in FIG. 30 shows the absolute velocity output obtained by integrating the sensor acceleration output (graph C) once by perfect integration. The velocity signal with noise superimposed diverges due to the influence of drift. In order to solve the above problems, in an actual active vibration isolation table, perfect integration 1 / s cannot be used, and the output of the acceleration sensor is integrated by imperfect integration 1 / (s + a) to obtain an approximate absolute velocity signal. Furthermore, a method of integrating this velocity signal by a similar integrator to obtain an approximate absolute displacement signal is adopted. However, since the phase characteristics in the low frequency region of the signal via imperfect integration are different from those in the case of the above perfect integration, an accurate negative feedback signal cannot be obtained. As a result, there are problems such as a phase delay in the low frequency region and an increase in gain, so that sufficient vibration isolation characteristics cannot be obtained.

[0207] (2) In the case of the acceleration sensor of the present invention FIG. 31 shows the relationship between the two electrode outputs, noise and drift, and the sensor acceleration output in the case of the differential acceleration sensor according to the third embodiment.

[0208] The electrode output B on the front side fis the capacitance determined by the gap between the movable electrode 811a and the fixed electrode 812a, and the electrode output B on the rear side r is obtained by detecting the capacitance determined by the gap between the movable electrode 811b and the fixed electrode 812b. Due to the common addition of noise and drift to these electrode outputs, the front-side electrode output B f is obtained, and the rear-side electrode output B r is obtained. As a result, the acceleration output C of the differential sensor s has a waveform in which noise and drift are canceled. Furthermore, the absolute velocity signal D obtained by completely integrating the acceleration output and the absolute displacement signal (not shown) do not diverge. Therefore, when the acceleration sensor of this embodiment is applied to an active vibration isolation table, in addition to the effect of improving the sensor sensitivity (for example, improving the positioning accuracy of the stage), a significant effect of improving the vibration isolation characteristics in the low-frequency region can be obtained.

[0209] Although the control circuit is not shown in the description of the third embodiment, as described in detail in the first embodiment, the effect of obtaining the conventional ideal sensor dynamic characteristics (gain and phase characteristics) can be obtained in the same manner by using the sum signal of the proportional amplification circuit and the differential circuit as the sensor output signal.

[0210] Furthermore, the MM type sensor applied to the third embodiment is a linear motion type, and its structure is a complete "axisymmetric" structure with respect to the axis of the movable part. Also, when the axis of the movable part is set as the Z axis and the X axis orthogonal to this Z axis is set, each component constituting the sensor has a complete "mirror symmetry" structure with respect to the center part of the X axis. Due to this "axisymmetry" and "mirror symmetry" structure, the influences received by the two electrodes on the front side and the rear side due to thermal expansion are exactly the same. Therefore, the differential signal of the two electrodes can avoid the influence of thermal expansion.

[0211] Background Art Background Art Background Art [Supplementary Note 2] Method for Confirming the Application Effect of the Present Invention (Part 1) 1. Regarding the configuration of the present invention As described above, the present invention is composed of the following combination of a and b. a. To reduce the dynamic pressure (squeeze pressure) of the air viscous fluid generated between the electrodes, solid A plurality of groove portions, hole portions, etc. are formed on the relative moving surfaces of the fixed-side and movable-side electrodes. b. A damping means by an electric circuit is provided in the servo amplifier so as to compensate for the reduction of the mechanical damping action. In the above a., by actually measuring the groove shape or the like formed on the electrode and performing numerical analysis of the viscous fluid, the reduction effect of damping can be theoretically predicted. However, the electrical damping effect of the above b. cannot be easily obtained. Therefore, without largely disassembling the mechanism part of the acceleration sensor, the following method for verifying that the present invention functions effectively by the combination of a. and b. is proposed.

[0212] 2. Method for obtaining the damping coefficient with respect to the electrode gap (1) The operating point of the movable electrode is changed so that it can be changed from the lower limit value to the upper limit value of the electrode gap h0. The operating point is set by the bias current flowing through the coil. Also, the electrode gap h0 can be estimated from the measured value of the capacitance [see Equation (6) of this specification]. (2) Under the conditions of the operating point set in the above (1), the servo-type vibration detector body is sweep-vibrated to obtain the sensor dynamic characteristics (gain - phase characteristics). (3) The damping coefficient C is obtained from the peak value of the gain at the resonance point. The damping coefficient C is obtained from the following equation by obtaining the damping ratio ζ from the graph of FIG. 11 or FIG. 12 of this specification.

[0213]

Equation

[0214] In Equation (30), m is the mass of the movable part of the sensor, and K P is the proportional gain including the stiffness of the support spring. K P is obtained from the relational expression between the resonance frequency f0 and the mass m [Equation (5) of this specification].

[0215] Therefore, through the steps (1) to (3) above, the damping coefficient C with respect to the electrode gap h0 can be obtained.

[0216] 3. Verification of the application effect of the present invention The graph of FIG. 6 described above shows the attenuation coefficient C with respect to the electrode gap h0. In this graph, a conventional formula (dashed-dotted line) without forming microgrooves (groove portions) and the present invention (dashed line) with formed microgrooves are shown in comparison. Here, the mechanical attenuation of the electrode with formed microgrooves is C M , and the electrical attenuation provided in the electric circuit of the servo amplifier is C E . Let the mechanical attenuation C M and the electrical attenuation C E . The hybrid attenuation C T = C M + C E is shown by a solid line.

[0217] The value of the electrical attenuation C E is determined by the electric circuit regardless of the electrode gap h0. Summarizing the results shown in the graph of FIG. 6, (1) In the case of a conventional sensor that does not form a plurality of groove portions (microgrooves), hole portions, etc. on the relative movement surface of the electrodes, when the electrode gap h0 decreases, the attenuation coefficient C M increases steeply in inverse proportion to the cube of the gap h0 [Equation (6) in this specification]. (2) A plurality of groove portions, hole portions, etc. are formed on the electrodes to sufficiently reduce the mechanical attenuation, and electrical attenuation is provided to compensate for the reduction in the mechanical attenuation effect. In this case, the attenuation coefficient C T maintains a substantially constant value. In this case, C T ≒ C E . Therefore, if the graph of the attenuation coefficient C with respect to the electrode gap h0 is between (1) and (2) above, it can be verified that the present invention is applied.

[0218] [Supplementary Note 3] Method for confirming the application effect of the present invention (Part 2) By the method shown below, it can be verified that the attenuation means by the electric circuit contributes to the sensor performance so as to compensate for the reduction in the attenuation effect due to the groove portions or hole portions formed between the electrodes. 1. Method for obtaining each attenuation coefficient (1) Estimate the electrode gap h0 from the measured capacitance value (see Equation (6) in this specification). (2) From the outer diameter of the electrode, the above electrode gap h0, the shape of the groove and / or hole formed on the electrode surface, obtain the mechanical damping coefficient C by the viscous fluid analysis shown in [3-1] of this specification M for the following two cases. (i) When a groove and / or hole is formed, the mechanical damping coefficient is C M = C MA . (ii) When no groove and / or hole is formed (conventional sensor), the mechanical damping coefficient is C M = C MB . In this case, C MB > C MA . (3) Obtain the sensor dynamic characteristics (gain-phase characteristics) obtained by sweep exciting the servo-type vibration detector body. From this dynamic characteristic, obtain the damping coefficient C T . The damping coefficient C T is obtained from the damping ratio ζ as shown in [Supplementary Note 2]. This damping coefficient C T includes all of the mechanical damping coefficient C M , the electrical damping C E incorporated in the servo amplifier, and the damping coefficient C ME caused by eddy currents generated in the coil bobbin, etc.

[0219] 2. Verification of the Application of the Present Invention Assume that the "resonance peak value" and "phase delay", which are evaluation indices of the sensor dynamic characteristics (gain-phase characteristics), satisfy the sensor target specifications. However, assume that the damping C ME ≈ 0. (1) When C T ≈ C MB (2) When C MA < C T < C MB In the case of (1) above, all of the electrical damping C E is replaced by the mechanical damping C M . In the case of (2) above, the electrical damping C M is added to the mechanical damping C Eis the hybrid attenuation with addition. All of them are derived from the graph in FIG. 6 of this specification (the relationship between the attenuation coefficient and the electrode gap).

Explanation of Signs

[0220] 101 Permanent magnet 102 Movable side member 105 Fixed side member 116, 104 Movable side yoke material 106 Coil 110 Movable part of displacement detector 117 Gap part

Claims

1. A housing which is a fixed member, A movable member provided so as to be movable in a predetermined direction with respect to the housing, An elastic member that supports the movable member with respect to the housing so that the movable member is disposed via a gap portion, A displacement detection unit that detects displacement of the movable member in a predetermined direction, Drive means that is driven by a servo amplifier so that a generating force that returns the movable member to the origin position is generated when a relative displacement from the origin position of the movable member is detected by the displacement detection unit, A movable-side electrode provided on the movable member, A fixed-side electrode provided on the housing side facing the movable-side electrode, and The displacement detection unit is configured to detect a capacitance formed in a gap portion between the movable-side electrode and the fixed-side electrode, A groove portion or a hole portion communicating with the atmosphere is formed on a relative movement surface of the movable-side electrode and the fixed-side electrode so as to reduce an attenuation effect due to a dynamic fluid pressure generated in a gap portion between the movable-side electrode and the fixed-side electrode, The servo amplifier is provided with attenuation means by an electric circuit so as to compensate for reduction of the attenuation effect, and a servo type vibration detector characterized by this.

2. The groove portion is a flow groove having a substantially concentric shape with respect to an axis of the fixed-side electrode or the movable-side electrode, and / or a substantially radial shape in a radial direction, and the servo type vibration detector according to Claim 1, characterized by this.

3. The attenuation means by the electric circuit is composed of a differential circuit that differentiates a signal of the relative displacement, and the servo type vibration detector according to Claim 1, characterized by this.

4. Further provided with a proportional amplification circuit that proportionally amplifies the signal of the relative displacement output from the displacement detection unit, The drive means is configured to be driven by a sum signal of the proportional amplification circuit and the differential circuit, and The servo type vibration detector according to Claim 3, characterized in that the sum signal is used as a sensor output signal indicating vibration detected.

5. Two sets of the displacement detection units are provided, each including an individual movable-side electrode and fixed-side electrode, and a gap between the movable-side electrode and the fixed-side electrode of each of the two sets of displacement detection units is configured to change in an inverse phase, and the servo type vibration detector according to Claim 1, characterized by this.

6. The movable-side electrodes are provided at both axial ends of the movable member, The fixed-side electrodes are respectively provided on the housing side facing these movable-side electrodes. The servo-type vibration detector according to claim 5, characterized in that a differential sensor is constituted by detecting two capacitance differences formed between the movable-side electrode and the fixed-side electrode.

7. A coil fixed to the fixed member side, A permanent magnet arranged so that magnetic flux flows through a gap between the housing and the movable member, The movable member is composed of the permanent magnet and a movable-side yoke material connecting a magnetic path therewith, or is composed only of the movable-side yoke material. The servo-type vibration detector according to claim 1, characterized in that a closed-loop magnetic circuit is formed by the movable member, a gap between the housing and the movable member, the fixed member, and the permanent magnet, thereby constituting the driving means by an electromagnetic force for moving the movable member in the axial direction.

8. One end of the movable member is fixed to the housing, and is configured to be supported by the elastic member and swing. The front and back surfaces of the movable member, and fixed-side opposing surfaces facing the front and back surfaces, The servo-type vibration detector according to claim 1, characterized in that electrode surfaces divided into a plurality of parts or a plurality of through holes communicating with the atmospheric pressure are formed on relative moving surfaces of the front and back surfaces and the fixed-side opposing surfaces.

9. The inertial mass of the movable member is m (Kg), The sum of the electrical gain of the servo amplifier and the mechanical spring constant of the elastic member is defined as the proportional gain K P (N / m), Mechanical damping coefficient C M (Ns / m), The electrical damping coefficient C of the servo amplifier E As the mechanical damping ratio ζ of the following formula M , and the electrical damping ratio ζ E When defined 【Number 31】 ζ M +ζ E ≧0.2, or ζ M The servo type vibration detector according to claim 1, characterized in that it is ≦0.

4.

10. An evaluation method for a servo-type vibration detector according to claim 1, An evaluation method for a servo-type vibration detector, which obtains a damping coefficient from a measured value of a peak value of a gain at a resonance point and evaluates the effect of the damping means provided in the electric circuit of the servo amplifier from the characteristics of the damping coefficient with respect to the size of the electrode gap.

Images