JPH06288773A - Semiconductor yaw rate sensor - Google Patents

Abstract

PURPOSE:To provide a semiconductor yaw rate sensor to improve S/N. CONSTITUTION:A beam structure separated from a substrate is formed on one part of the silicon substrate. The AC power is applied to one surface of a weight formed on the tip of the beam and the wall surface of the substrate opposite to the weight surface, and the weight is excited by the staic electricity. Electrodes are arranged opposite to each other on one surface of the weight and the wall surface of the substrate opposite to the beam surface in the axial direction orthogonal to the exciting direction of the weight, the change in the capacity between the opposite electrodes is electrically detected, and the yaw rate to be exerted in the same direction is detected. The signal from the electrode for detecting the yaw rate is superposed on the carrier. In the band pass filter 70, the center frequency coincides with the frequency of the carrier, and the signal from a differential amplifier 69 passes through the band pass filter.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor yaw rate sensor.

[0002]

2. Description of the Related Art Conventionally, a type using a piezoelectric ceramic as a yaw rate sensor has been used for controlling the attitude of an automobile and preventing camera shake of a consumer video camera. By the way, there are other possible fields of application of this sensor,
The accuracy and cost limit the expansion to other areas. Therefore, the applicant of the present application has proposed a new semiconductor yaw rate sensor by Japanese Patent Application No. 4-223072.

[0003]

However, in the semiconductor yaw rate sensor disclosed in Japanese Patent Application No. 4-223072, noise (thermal noise, 1 / f noise) is also amplified when the signal of the sensing portion is amplified. S / N
The issue remains that it is difficult to improve

Therefore, an object of the present invention is to provide a semiconductor yaw rate sensor capable of improving S / N.

[0005]

According to the present invention, a beam structure is formed on a part of a semiconductor substrate so as to be separated from the substrate, and one surface of a weight formed at the tip of the beam is provided on a substrate wall surface facing the same weight surface. An AC power is applied to excite the weight by static electricity, and an electrode is placed to face the wall surface of the substrate that faces one surface of the weight and the beam surface in the axial direction orthogonal to the excitation direction of the weight. In a semiconductor yaw rate sensor configured to detect a yaw rate that works in the same direction by electrically detecting a change, an AM modulation circuit that superimposes a signal from a yaw rate detection electrode on a carrier wave, and a center frequency of the carrier wave matches the carrier wave, The gist is a semiconductor yaw rate sensor provided with a bandpass filter that passes a signal from the AM modulation circuit.

[0006]

[Function] AC power is applied to excite the weight by static electricity,
The change in the capacitance between the counter electrodes arranged to face each other in the axial direction orthogonal to the excitation direction is electrically detected. Then, this signal is AM-modulated by being superimposed on the carrier wave in the AM modulation circuit. Furthermore, the signal from the AM modulation circuit
It passes through a bandpass filter whose center frequency matches the frequency of the carrier.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 2 shows a plan view of the semiconductor yaw rate sensor, and FIG. 3 shows a sectional view taken along the line AA of FIG.
In the following description, when expressing the three-dimensional direction, FIG.
The left-right direction is the X-axis direction, the vertical direction is the Y-axis direction, and the direction orthogonal to the paper surface is the Z-axis direction.

The silicon substrate 51 is a flat plate, and a rectangular recess 52 (depth: T) is formed in the silicon substrate 51. Inside the recess 52, two rod-shaped beams 53 extend from the left side surface of FIG.
A weight 55 is formed at the tip of the. On the other hand, inside the recess 52, two rod-shaped beams 5 are seen from the right side surface of FIG.
4 is extended and a weight 56 is formed at the tip of the beam 54. The weights 55 and 56 are wider than the beams 53 and 54 and are formed in a rectangular shape. Beam 53,
54 and the weights 55 and 56 have the same thickness.

Further, one side surface of the weight 55 (upper surface in FIG. 2) and the inner surface of the recess 52 are slightly separated (distance a). Similarly, the other side surface (lower surface in FIG. 2) of the weight 55 and the inner surface of the recess 52 are slightly separated (distance a). Similarly, the bottom surface (lower surface in FIG. 3) of the weight 55 and the inner surface of the recess 52 are slightly separated (distance d1).

On the other hand, one side surface of the weight 56 (the upper surface in FIG. 2) is slightly separated from the inner surface of the recess 52 (distance a). Similarly, the other side surface (lower surface in FIG. 2) of the weight 56 and the inner surface of the recess 52 are slightly separated (distance a). Similarly, the weight 56
The bottom surface (the lower surface in FIG. 3) of the above is slightly separated from the inner surface of the concave portion 52 (distance d1).

As described above, this sensor has a cantilever structure. In this structure, the distance d1 is formed by sacrificial layer etching or the like using the surface micromachining technique. In FIG. 3, the weights 55, 5 on the bottom surface of the recess 52 are
Electrodes 57 and 58 are formed on the surface facing 6 and electrodes 59 and 60 are formed on the portions of the weights 55 and 56 facing the electrodes 57 and 58. Further, electrodes 59 and 60 are formed on the surface of the inner wall of the recess 52 facing the weights 55 and 56 (the upper surface of the recess 52 in FIG. 2), and the weights 55 and 60 facing the electrodes 59 and 60 are formed. Electrodes 61 and 62 are formed at the portion 56.

The electrode 6 is provided on the surface of the inner wall of the recess 52 facing the weights 55 and 56 (the surface below the recess 52 in FIG. 2).
3 and 64 are formed, and electrodes 65 and 66 are formed on portions of the weights 55 and 56 facing the electrodes 63 and 64.

In this structure, the electrodes 57, 58,
59, 60, 61, 62, 63, 64 are insulated
Has been. And a capacitor at electrodes 59 and 57
C_{s +}, Capacitor C at electrodes 60 and 58_s-But electrode 59
And 61 with capacitor C_{d +}, Electrodes 66 and 64
Sensor C _d-But capacitor C at electrodes 65 and 63_{e +},electrode
Capacitor C at 60 and 62_e-But each formed
There is.

Further, the beams 53 and 54 are respectively provided to the electrode 5
9 (61, 65) and 60 (62, 66) wiring regions are formed. Although 59, 61, and 65 are described as separate electrodes for description, they are the same electrode (same potential). 60,6
2 and 66 are the same electrode (same potential), although they are described as separate electrodes for the same description.

FIG. 1 shows an electric circuit diagram of the semiconductor yaw rate sensor. The processing circuit of the sensor includes an oscillator 67, a sensing unit 68, a differential amplifier 69, and a bandpass filter 7.
0, a sample hold circuit 71, and a post-stage amplifier 72.

The capacitor Cr in FIG. 1 is not shown in FIGS. 2 and 3, but it is connected in parallel with the resistor R,
It is a fixed capacitance designed as r = C _{s +} = C _s- . The capacitors C _{e +} and C _e− drive the weights 55 and 56 with an electrostatic force Fe. Further, the capacitors C _{s +} and C _s− are capacitances for detecting the amount by which the weights 55 and 56 receive the Coriolis force Fc and are displaced in the Z-axis direction.

The capacitors C _{d +} and C _d- shown in FIG. 2 are monitoring capacitors for detecting the amount of movement of the weights 55 and 56 by the driving capacitors C _{e +} and C _{e- in the} Y-axis direction. next,
A configuration other than the sensing unit 68 will be described in FIG.

The oscillator 67 has an oscillation frequency f of 10 KHz and a voltage (AC power) for driving the weights 55 and 56.
And a signal (carrier wave) to the capacitors C _{s +} and C _s− . The resistor R is a capacitor C _{s +} or C _s- and C
A bias voltage is applied to the connection part of r, and the value is set as R >> 1 / ωCr. By applying a bias, subsequent signal processing is enabled.

The differential amplifier circuit 69 amplifies the difference voltage between the inputs (capacitors C _{s +} and C _s− ). The center frequency of the bandpass filter 70 is 10 KHz, which matches the frequency of the carrier wave. The bandpass filter 70 also attenuates signals outside the predetermined frequency band (near the center frequency). In the present embodiment, the bandpass filter 7
0 is a switched capacitor filter (abbreviation: S.C.
F).

Sample hold circuit 71 (detection circuit)
Is for demodulating an AM-modulated signal described later. The operational amplifier 73 and the resistors 74 and 75 form a reference voltage in the processing circuit. The post-stage amplifier 72 amplifies the detected signal. The post-stage amplifier 72 may be omitted.

In this embodiment, the electrodes 57, 58, 59, 6
The yaw rate detection electrode is constituted by 0, and the AM modulation circuit is constituted by the oscillator 67 and the differential amplifier 69.
Next, the operation of the semiconductor yaw rate sensor thus configured will be described.

From the oscillator 67, the voltage V _IN (= V _CM · cos
When the omega _c t) is applied to the capacitor C _e- and C _{e +,}
An electrostatic force Fe shown by the equation (1) is generated. _{Fe = (ε 0 S / 2a} 2) · V IN 2 ··· (1) However, epsilon _0; dielectric constant a; capacitor C _e- and C _{e +} spacing S; capacitor C _e- and C _{e +} of opposed electrodes Area The weights 55 and 56 are displaced in the Y-axis direction by the electrostatic force Fe. If the displacement amount is Dy, then Dy = KFe (2) where K is a constant determined by the cantilever. However, at this time, the weights 55 and 56 move in different directions.

From the above equations (1) and (2), Y of the weights 55 and 56
The axial speed is V_y55,V _y56Then V_y55= -V_y56 = K ・ ((ε₀S / 4a²) ・ V_cm ²・ 2ω_c・ Sin2ω_ct ... (3)

At this time, when the X-axis is used as the rotation axis and is rotated at the angular velocity ω, the Coriolis force F _C55 = 2 mV _y55 ω, F is applied to the Z-axis.
_C56 = 2 mV _y56 ω is generated. By this, the weight 5
5, 56 are respectively displaced in the Z-axis direction. This is D
If _{Z55, D Z56, D Z55 =} L 55 · F C55 D Z56 = L 56 · F C56 ··· (4) However, L _55, L _56; represented by the constant determined by the cantilever.

If the weights 55 and 56 and the cantilever are designed to have the same size, L ₅₅ = L ₅₆ , and | D _{Z5 5} | = | D _Z56 | =
It can be expressed as Δd. That is, the capacities of C _{s +} and C _s− are C _{s +} = (ε ₀ · S) / (d + Δd), C _s− = (ε ₀ · S) / (d−Δd) (5) .

Therefore, the output V _pre of the differential amplifier 69 is V _pre = V _IN · {C _{s +} / (C _{s +} + Cr) −C _s− / (C _s− + Cr)} · AV 1 ≈V _IN · (−Δd / 2d) · AV1 (6) where AV1 is represented by the amplification factor of the differential amplifier 69.

[0027] In addition, before the formula (3), (4) by Δd is _{Δd = L 55 · 2m · K} (ε 0 · S / 4a 2) · V CM 2 · 2ω c · ω · sin2ω c t ··· It is represented by (7).

[0028] Further, (6), (7) a ^{_{V pre = AV1 · V CM 3}} · L 55 · 2m · K (ε 0 · S / 4a 2) · ω c · ω · (sinω c t + sin3ω c t ) (8)

Here, V _{CM on} the right side of the above equation (8)
Portion of the _{^{3 · L 55 · 2m · K}} (ε 0 · S / 4a 2) · ω c is a constant determined the structure of the cantilever, at the conditions of the input voltage. From this equation (8), it is shown that V _pre represents a voltage proportional to the angular velocity ω to be detected, and
This is the frequency of the input signal f _IN = ω _C / 2π and its 3
It will be represented as a voltage output that is AM-modulated to double the frequency.

Here, only the detection signal has been described so far, but there is also noise generated in its own circuit element when processing the signal by the differential amplifier 69 and noise injected from the outside into the power supply system. This is also amplified by the differential amplifier 69. Thus (8) is ^{_{V pre = AV1 · V CM 3}} · L 55 · 2m · K (ε 0 · S / 4a 2) · ω c · ω · (sinω c t + sin3ω c t) + AV1 · V N. (9) A noise item AV ₁ · V _N that reduces the S / N of the detected angular velocity ω is generated.

Therefore, the angular velocity to be detected is AM-modulated as shown in equation (9), and the center frequency f _c = ω _C / 2π of the bandpass filter 70 is passed. As a result, S / N
Improve.

[0032] Then, when the output of the band pass filter 70 of f _c = ω _C / 2π and _{V BPF, V BPF = AV1 ·} V CM 3 · L 55 · 2m · K (ε 0 · S / 4a 2) · _{ω c · ω · sinω c t} + AV1 · V N (f c) becomes (10).

V _BPF is expressed by the equation (10),
Term noise _{AV1 · V N (f c)} , that is, the frequency component is not left only the noise component of the same components as _{f c, AV1 · V N »AV1} · V N (f c) ··· (11) And an output proportional to the angular velocity ω of high S / N is obtained. This is processed by the sample hold circuit (detection circuit) 71 as necessary, so that the output V proportional to the angular velocity ω is obtained.
_{out V out ≒ AV1 · V CM} 3 · L 55 · 2m · K (ε 0 · S / 4a 2) · ω c · ω ··· (12) is obtained.

If necessary, it is amplified by the post-stage amplifier 72. As described above, in this embodiment, the oscillator 67 and the differential amplifier 69 (AM modulation circuit) are used for the electrodes 57, 57
Signals from 59, 58, and 60 (electrodes for yaw rate detection) are superposed, and the signal from the differential amplifier 69 is passed by the bandpass filter 70 whose center frequency coincides with that of the carrier. Therefore, when the differential amplifier 69 processes a signal, there is noise generated in its own circuit element or noise injected from the outside into the power supply system, but these noises are removed. That is, noise (thermal noise, 1 / f noise) is amplified, and S / N can be improved.

[0035]

As described above in detail, according to the present invention,
It has an excellent effect of improving S / N.

[Brief description of drawings]

FIG. 1 is an electric circuit diagram of a semiconductor yaw rate sensor.

FIG. 2 is a plan view of a semiconductor yaw rate sensor.

FIG. 3 is a cross-sectional view taken along the line AA of FIG.

[Explanation of symbols]

51 Silicon substrate as semiconductor substrate 55, 56 Weight 57, 58, 59, 60 Electrode as yaw rate detecting electrode 67 Oscillator as AM modulation circuit 69 Differential amplifier as AM modulation circuit 70 Bandpass filter

Claims (1)

[Claims] 1. A beam structure is formed on a part of a semiconductor substrate so as to be separated from the substrate, and one side of a weight formed at the tip of the beam and AC power are applied to a substrate wall surface facing the weight surface to statically weight the weight. And the electrodes are arranged to face the wall surface of the substrate that faces one surface of the weight and the same beam surface in the axial direction orthogonal to the excitation direction of the weight, and the change in capacitance between the counter electrodes is electrically detected. In a semiconductor yaw rate sensor that detects yaw rate acting in the same direction, a signal from the yaw rate detection electrode is superimposed on the carrier wave A
A semiconductor yaw rate sensor comprising: an M modulation circuit; and a bandpass filter having a center frequency that matches the carrier wave and passing a signal from the AM modulation circuit.