JP2015017819A - Acceleration sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an acceleration sensor capable of reducing output errors.SOLUTION: An acceleration sensor is a differential capacitance acceleration sensor that comprises: an X detection unit 10 detecting and outputting an X-direction acceleration using a first movable electrode 11; a Y detection unit 20 detecting and outputting a Y-direction acceleration using a second movable electrode 21; a Z detection unit 30 detecting and outputting a Z-direction acceleration using a third movable electrode 31; and an automatic correction circuit 220 subtracting a correction value represented by a polynominal expression (Aa^3+Ba^5+...) from an acceleration a, where a is the acceleration output from each of the X detection unit 10, the Y detection unit 20, and the Z detection unit 30 and A and B are correction coefficients.

Description

  The present invention relates to an acceleration sensor.

  Conventionally, a technique for correcting the output of a differential capacitance type acceleration sensor is known. For example, in paragraph 0040 of Patent Document 1, “a plurality of measurement results and calibration samples are arranged in a table of XY pairs forming a DAC correction table. The XY array uses cubic spline interpolation. Provides a precisely set calibration output voltage level. " Other prior art documents related to acceleration sensors and angular velocity sensors having a correction function include Patent Documents 2 to 4.

Special table 2011-520128 gazette JP 2008-170294 A JP 2005-308531 A Japanese Patent Laid-Open No. 06-331647

  According to an ideal acceleration sensor, the output is proportional to the applied acceleration, but the actual output currently includes an error. As the application field of acceleration sensors increases, it is desired to realize high accuracy and high reliability of acceleration sensors.

  Therefore, an object of the present invention is to obtain an acceleration sensor that can reduce an output error.

  The present invention is a differential capacitance type acceleration sensor, a movable electrode movable according to an acceleration given from the outside, a detection unit for detecting and outputting an acceleration in a predetermined direction using the movable electrode, If the acceleration output from the detection unit is a and the correction coefficients are A, B,..., The correction is expressed by a polynomial (Aa ^ 3 + Ba ^ 5 +. And an automatic correction circuit for subtracting the value from the acceleration a.

In the present invention, the automatic correction circuit detects the acceleration by changing the parallel plate type electrostatic gap d, the dielectric constant is ε, the opposing area of the electrode is S, and the initial electrostatic gap is d _0. When the gap change amount is Δd, the acceleration a may be Y shown below, and the correction coefficients may be A, B... N shown below.

Y = εS × {−2Δd / (d ₀ ² −Δd ² )}
A = −2 · Δd ³ εS / d ₀ ⁴
B = −2 · Δd ⁵ εS / d ₀ ⁶
:
N = −2 · Δd ^{2n + 1} εS / d ₀ ^{2n + 2}
In the present invention, the automatic correction circuit detects the acceleration by changing the torsion type electrostatic gap d, the dielectric constant is ε, the opposing area of the electrode is S, the initial electrostatic gap is d ₀ , When the gap change amount is Δd, the acceleration a may be Y shown below, and the correction coefficients may be A, B... N shown below.

Y = εS / Δd × ln {(d ₀ ² −Δd ² ) / d ₀ ² }
A = −Δd ³ εS / 2d ₀ ⁴
B = −Δd ⁵ εS / 3d ₀ ⁶
:
N = −Δd ^{2n + 1} εS / {(n × 1) × d ₀ ^{2n + 2} }

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the acceleration sensor which can reduce an output error.

FIG. 1 is a perspective view illustrating an internal configuration example of a package incorporating the acceleration sensor according to the embodiment. FIG. 2 is an exploded perspective view of the acceleration sensor according to the embodiment. 3A and 3B are cross-sectional views of the acceleration sensor according to the embodiment, in which FIG. 3A is a cross-sectional view of the X detection unit, and FIG. 3B is a cross-sectional view of the Z detection unit. FIG. 4 is a cross-sectional view of the X detection unit in the state where no acceleration in the X direction is applied in the acceleration sensor according to the embodiment. FIG. 5 is a diagram for explaining the principle of detecting the acceleration in the X direction in the state shown in FIG. FIG. 6 is a cross-sectional view of the X detection unit in a state where 1 G acceleration is applied in the X direction in the acceleration sensor according to the embodiment. FIG. 7 is a diagram for explaining the principle of detecting the acceleration in the X direction in the state shown in FIG. FIG. 8 is a cross-sectional view of the Z detection unit in a state where 1 G acceleration is applied in the Z direction in the acceleration sensor according to the embodiment. FIG. 9 is a diagram for explaining the principle of detecting the acceleration in the Z direction in the state shown in FIG. FIG. 10 is a graph showing output characteristics of an ideal acceleration sensor and an actual acceleration sensor (comparative example). FIG. 11 is a functional block diagram of a main part of the acceleration sensor according to the embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In addition, below, while attaching | subjecting a common code | symbol to the same component, the overlapping description is abbreviate | omitted.

[Internal structure of the package]
FIG. 1 is a perspective view illustrating an internal configuration example of a package 300 on which the acceleration sensor according to the embodiment is mounted. Here, a state in which the lid of the package 300 mounted on the substrate 500 is opened is shown. As shown in this figure, the package 300 is mounted with a sensor chip 100, an ASIC 200 that performs various calculations based on the output from the sensor chip 100, and the like. Terminals 400 are drawn from the package 300 and connected to the substrate 500. This acceleration sensor is a capacitance type acceleration sensor and is manufactured by MEMS technology. In order to detect the acceleration in the three axis directions of XYZ, a weight (movable electrode) for each axis is formed and arranged in the sensor chip 100.

[Configuration of acceleration sensor]
FIG. 2 is an exploded perspective view of the acceleration sensor (sensor chip 100) according to the embodiment. As shown in the figure, the upper and lower surfaces of the sensor unit 1 are sandwiched between an upper fixing plate 2a and a lower fixing plate 2b. The sensor unit 1 is formed of a silicon SOI substrate or the like, and the upper fixing plate 2a and the lower fixing plate 2b are formed of an insulator such as glass.

  Hereinafter, in the sensor unit 1, the part that detects the acceleration in the X direction is “X detection part 10”, the part that detects the acceleration in the Y direction is “Y detection part 20”, and the part that detects the acceleration in the Z direction is “ This will be referred to as “Z detection unit 30”. The X direction is one of the planar directions. The Y direction is one of the planar directions and is a direction orthogonal to the X direction. The Z direction is the vertical direction.

  The X detector 10 detects the acceleration in the X direction by swinging the first movable electrode 11 around the pair of beam portions 12a and 12b. That is, the first fixed electrodes 13a and 13b are arranged so as to face one side and the other side of the surface of the first movable electrode 11 with a straight line connecting the pair of beam portions 12a and 12b as a boundary line. Thereby, the acceleration in the X direction can be detected based on the change in capacitance between the first movable electrode 11 and the first fixed electrodes 13a and 13b.

  The Y detector 20 detects the acceleration in the Y direction by swinging the second movable electrode 21 around the pair of beam portions 22a and 22b. That is, the second fixed electrodes 23a and 23b are arranged to face one side and the other side of the surface of the second movable electrode 21 with a straight line connecting the pair of beam portions 22a and 22b as a boundary line. Thereby, the acceleration of a Y direction is detectable based on the change of the electrostatic capacitance between the 2nd movable electrode 21 and 2nd fixed electrode 23a, 23b.

  The Z detection unit 30 detects the acceleration in the Z direction by translating the third movable electrode 31 held by the two pairs of beam units 32a, 32b, 32c, and 32d in the vertical direction. That is, the third fixed electrodes 33a and 33b are arranged to face the front and back surfaces of the third movable electrode 31. Thereby, the acceleration of a Z direction is detectable based on the change of the electrostatic capacitance between the 3rd movable electrode 31 and the 3rd fixed electrodes 33a and 33b.

  The X detection unit 10 and the Y detection unit 20 have the same shape that is simply rotated by 90 °, and are arranged on one side of the Z detection unit 30 in a different shape. That is, as shown in FIG. 2, the frame portion 3 is formed with three rectangular frames 10a, 20a, 30a arranged in a straight line. The first movable electrode 11 is disposed in the rectangular frame 10a, the second movable electrode 21 is disposed in the rectangular frame 20a, and the third movable electrode 31 is disposed in the rectangular frame 30a. Each of the first to third movable electrodes 11, 21, 31 has a substantially rectangular shape. A gap of a predetermined size is left between the first to third movable electrodes 11, 21, 31 and the side walls of the rectangular frames 10a, 20a, 30a.

  3A and 3B are cross-sectional views of the acceleration sensor according to the embodiment, in which FIG. 3A shows a cross section of the X detector 10 and FIG. 3B shows a cross section of the Z detector 30. Since the cross section of the Y detection unit 20 is the same as that of the X detection unit 10, the illustration is omitted here.

  First, the cross section of the X detector 10 is as shown in FIG. That is, the first movable electrode 11 is connected to the frame portion by connecting the substantially central portion of the two opposing sides of the surface of the first movable electrode 11 and the side wall portion of the rectangular frame 10a with the pair of beam portions 12a and 12b. 3 is swingably supported. On the side of the upper fixed plate 2a facing the first movable electrode 11, first fixed electrodes 13a and 13b are provided with a straight line connecting the beam portion 12a and the beam portion 12b as a boundary line. The first fixed electrodes 13a and 13b are drawn out to the upper surface (one side) of the upper fixed plate 2a using the first through electrodes 14a and 14b. The material of the first through electrodes 14a and 14b is a conductor such as silicon, tungsten, or copper, and the surrounding material that holds the first through electrodes 14a and 14b is an insulator such as glass.

  The same applies to the Y detector 20. That is, the second movable electrode 21 is connected to the frame portion by connecting the substantially central portion of two opposing sides of the surface of the second movable electrode 21 and the side wall portion of the rectangular frame 20a with a pair of beam portions 22a and 22b. 3 is swingably supported. On the side of the upper fixed plate 2a facing the second movable electrode 21, second fixed electrodes 23a and 23b are provided with a straight line connecting the beam portion 22a and the beam portion 22b as a boundary line. The second fixed electrodes 23a and 23b are drawn to the upper surface of the upper fixed plate 2a using the second through electrodes 24a and 24b. The material of the second through electrodes 24a and 24b is a conductor such as silicon, tungsten, or copper, and the surrounding material that holds the second through electrodes 24a and 24b is an insulator such as glass.

  Furthermore, the cross section of the Z detection unit 30 is as shown in FIG. That is, by connecting the four corners of the third movable electrode 31 and the side wall portion of the rectangular frame 30a by two pairs of L-shaped beam portions 32a, 32b, 32c, and 32d, the third movable electrode 31 is moved vertically. It can be translated. The shapes of the beam portions 32a, 32b, 32c, and 32d are not particularly limited. However, if the beam portions 32a, 32b, 32c, and 32d are L-shaped, the beam portions 32a, 32b, 32c, and 32d can be lengthened. A third fixed electrode 33a is provided on the side of the upper fixed plate 2a facing the third movable electrode 31, and a third fixed electrode 33b is provided on the side of the lower fixed plate 2b facing the third movable electrode 31. Is provided. The third fixed electrode 33a is drawn to the upper surface of the upper fixed plate 2a using the third through electrode 34a. The third fixed electrode 33b includes a protruding area 33b2 protruding from the rectangular area 33b1 (see FIG. 2). The protruding region 33b2 is connected to a columnar fixed electrode 34c separated from the third movable electrode 31, and the columnar fixed electrode 34c is connected to a third through electrode 34b provided on the upper fixed plate 2a. It has a configuration. Thus, the third fixed electrode 33b can be drawn out to the upper surface of the upper fixed plate 2a using the columnar fixed electrode 34c and the third through electrode 34b. The material of the third through electrodes 34a and 34b is a conductor such as silicon, tungsten, or copper, and the surrounding material that holds the third through electrodes 34a and 34b is an insulator such as glass.

[X-direction acceleration detection principle]
Next, the principle of acceleration detection in the X direction will be described. First, when the dielectric constant is ε, the opposing area of the electrode is S, and the opposing gap of the electrode is d, the capacitance C can be calculated by C = εS / d. When the movable electrode rotates due to acceleration, the facing gap d changes, so that the capacitance C changes. Therefore, the differential capacity (C1-C2) is CV-converted by the ASIC 200. A method for reducing the output error will be described later.

  FIG. 4 shows a cross section of the X detection unit 10 in a state where no acceleration in the X direction is applied. In this case, as shown in FIG. 5, the capacitances C1 and C2 between the first movable electrode 11 and the first fixed electrodes 13a and 13b are equal. The ASIC 200 calculates a difference value (C1−C2 = 0) between the electrostatic capacitance C1 and the electrostatic capacitance C2 and outputs it as an X output.

  FIG. 6 shows a cross section of the X detection unit 10 in a state where 1 G acceleration is applied in the X direction. In this case, as shown in FIG. 7, the electrostatic capacitance C1 between the first movable electrode 11 and the first fixed electrode 13a becomes a parasitic capacitance + ΔC, and the first movable electrode 11 and the first fixed electrode 13b. The capacitance C2 between and is a parasitic capacitance −ΔC. The ASIC 200 calculates a difference value (C1−C2 = 2ΔC) between the electrostatic capacitance C1 and the electrostatic capacitance C2 and outputs it as an X output.

  As described above, the X detection unit 10 detects the acceleration in the X direction based on the change in capacitance. The principle by which the Y detection unit 20 detects acceleration in the Y direction is the same.

[Z-direction acceleration detection principle]
FIG. 8 shows a cross section of the Z detection unit 30 in a state where 1 G acceleration is applied in the Z direction. In this case, as shown in FIG. 9, the capacitance C5 between the third movable electrode 31 and the third fixed electrode 33a becomes a parasitic capacitance + ΔC, and the third movable electrode 31 and the third fixed electrode 33b. The capacitance C6 between and is a parasitic capacitance −ΔC. The ASIC 200 calculates a difference value (C5−C6 = 2ΔC) between the capacitance C5 and the capacitance C6, and outputs it as a Z output. Thus, the Z detection unit 30 detects the acceleration in the Z direction based on the change in capacitance.

[Output error]
FIG. 10 is a graph showing output characteristics of an ideal acceleration sensor and an actual acceleration sensor (comparative example). As shown in this figure, an ideal acceleration sensor output (ideal output) O1 is proportional to the applied acceleration. On the other hand, the actual acceleration sensor output O2 currently includes an error as indicated by an arrow in the figure. Hereinafter, a case where acceleration is detected by changing the parallel plate type electrostatic gap d (see FIG. 8) will be described as an example.

  First, the amount of change of the third movable electrode 31 when acceleration is applied can be expressed by the following equation (Hooke's law). a is an applied acceleration, m is the weight of the third movable electrode 31, k is a spring constant, and Δd is a change amount (gap change amount) of the third movable electrode 31.

Δd = ma / k
Further, the capacitances C5 and C6 at the time of acceleration application can be expressed by the following equations. ε is the dielectric constant, d ₀ is the initial electrostatic gap, and S is the opposing area of the electrodes.

C5 = ε × S / (d ₀ + Δd)
C6 = ε × S / (d ₀ −Δd)
Therefore, the differential capacitance is as follows:

C5-C6 = εS × {−2Δd / (d ₀ ² −Δd ² )}
Such a differential capacitor does not accurately output a straight line but includes an error. Therefore, the acceleration sensor according to the present embodiment employs the following configuration in order to reduce output errors.

[Function block diagram]
FIG. 11 is a functional block diagram of a main part of the acceleration sensor according to the embodiment. As shown in this figure, the ASIC 200 includes an automatic correction circuit 220 that corrects a deviation (error) from the ideal output O1 by subtracting a correction value from the input acceleration a. The correction value is calculated from a polynomial (Aa ^ 3 + Ba ^ 5 +...) With the input acceleration a as a variable.

  Specifically, the arithmetic circuit 221 subtracts the correction value (Aa ^ 3 + Ba ^ 5 +...) From the acceleration a output from the X detection unit 10 and outputs the result. The arithmetic circuit 222 subtracts the correction value (Aa ^ 3 + Ba ^ 5 +...) From the acceleration a output from the Y detector 20 and outputs the result. Further, the arithmetic circuit 223 subtracts the correction value (Aa ^ 3 + Ba ^ 5 +...) From the acceleration a output from the Z detector 30 and outputs the result. As described above, when a polynomial (odd function) composed of odd-order terms of the third order or higher is used as a correction value, linearity correction can be realized. That is, the output of the acceleration sensor can be made closer to the ideal output O1 as compared with the conventional case.

  Here, when the acceleration is detected by changing the parallel plate electrostatic gap d (see FIG. 8), such as when the Z detection unit 30 detects acceleration in the Z direction, the acceleration a is represented by the following Y Can be expressed as In this case, the correction coefficients are preferably set to A, B. Thereby, the output of the acceleration sensor can be brought closer to the ideal output O1.

Y = εS × {−2Δd / (d ₀ ² −Δd ² )}
A = −2 · Δd ³ εS / d ₀ ⁴
B = −2 · Δd ⁵ εS / d ₀ ⁶
:
N = −2 · Δd ^{2n + 1} εS / d ₀ ^{2n + 2}
Of course, the correction value may be a cubic equation (Aa ^ 3). In this case as well, the acceleration a can be expressed by Y, and the correction coefficient A is also set to A = −2 · Δd ³ εS / d ₀ ^{4 as} described above.

  On the other hand, when the X detection unit 10 detects acceleration in the X direction, or when the Y detection unit 20 detects acceleration in the Y direction, the acceleration is detected by changing the torsional electrostatic gap d (see FIG. 6), acceleration a can be represented by Y shown below. In this case, the correction coefficients are preferably set to A, B. Thereby, the output of the acceleration sensor can be brought closer to the ideal output O1.

Y = εS / Δd × ln {(d ₀ ² −Δd ² ) / d ₀ ² }
A = −Δd ³ εS / 2d ₀ ⁴
B = −Δd ⁵ εS / 3d ₀ ⁶
:
N = −Δd ^{2n + 1} εS / {(n × 1) × d ₀ ^{2n + 2} }
Of course, the correction value may be a cubic equation (Aa ^ 3). Also in this case, the acceleration a can be expressed by the above Y, and the correction coefficient A is also set to A = −Δd ³ εS / 2d ₀ ^{4 as} described above.

  As described above, the acceleration sensor according to the embodiment is a differential capacitance type acceleration sensor, and is a movable electrode (first movable electrode 11, second movable electrode 21, or third movable electrode 31). A detection unit (X detection unit 10, Y detection unit 20, or Z detection unit 30) and an automatic correction circuit 220. The movable electrode moves in accordance with the acceleration given from the outside. The detection unit detects and outputs an acceleration in a predetermined direction using the movable electrode. The automatic correction circuit 220 is a polynomial (Aa ^ 3 + Ba ^ 5 +...) Composed of odd-order terms of the third order or higher, where a is the acceleration output from the detection unit and A, B,. Is subtracted from the acceleration a. As a result, the output of the acceleration sensor can be made closer to the ideal output O1 as compared with the conventional case, so that the output error can be reduced as compared with the conventional case.

Further, when the automatic correction circuit 220 detects acceleration by changing the parallel plate type electrostatic gap d, the acceleration a is set to Y shown below, and the correction coefficients are set to A, B... N shown below. Is preferable. Here, the dielectric constant is ε, the opposing area of the electrodes is S, the initial electrostatic gap is d ₀ , and the gap change amount is Δd. As a result, the output of the acceleration sensor can be brought closer to the ideal output O1, and the output error can be further reduced.

Y = εS × {−2Δd / (d ₀ ² −Δd ² )}
A = −2 · Δd ³ εS / d ₀ ⁴
B = −2 · Δd ⁵ εS / d ₀ ⁶
:
N = −2 · Δd ^{2n + 1} εS / d ₀ ^{2n + 2}
On the other hand, when the automatic correction circuit 220 detects acceleration by changing the electrostatic gap d of the torsion method, the acceleration a is set to Y shown below, and the correction coefficients are set to A, B. It is preferable to do this. Here, the dielectric constant is ε, the opposing area of the electrodes is S, the initial electrostatic gap is d ₀ , and the gap change amount is Δd. As a result, the output of the acceleration sensor can be brought closer to the ideal output O1, and the output error can be further reduced.

Y = εS / Δd × ln {(d ₀ ² −Δd ² ) / d ₀ ² }
A = −Δd ³ εS / 2d ₀ ⁴
B = −Δd ⁵ εS / 3d ₀ ⁶
:
N = −Δd ^{2n + 1} εS / {(n × 1) × d ₀ ^{2n + 2} }
The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above embodiments, and various modifications can be made. For example, in the above embodiment, an acceleration sensor in which the X detection unit 10, the Y detection unit 20, and the Z detection unit 30 are arranged in one chip is illustrated, but the present invention is an X detection unit 10 / Y detection. Only the unit 20 and the Z detection unit 30 can be applied to an acceleration sensor arranged in one chip. Of course, the detailed specifications (shape, size, layout, etc.) of the sensor chip 100 can be changed as appropriate.

10 X detector (detector)
11 First movable electrode (movable electrode)
20 Y detector (detector)
21 Second movable electrode (movable electrode)
30 Z detector (detector)
31 Third movable electrode (movable electrode)
220 Automatic correction circuit ε Dielectric constant S Electrode facing area d ₀ Initial electrostatic gap Δd Gap variation a, Y Acceleration A, B... N Correction coefficient

Claims (3)

A differential capacitance type acceleration sensor,
A movable electrode that can move according to acceleration given from the outside;
A detection unit that detects and outputs acceleration in a predetermined direction using the movable electrode;
If the acceleration output from the detection unit is a and the correction coefficients are A, B,..., The correction is expressed by a polynomial (Aa ^ 3 + Ba ^ 5 +. An acceleration sensor comprising: an automatic correction circuit that subtracts a value from the acceleration a. The automatic correction circuit detects the acceleration by changing the parallel plate type electrostatic gap d, the dielectric constant is ε, the opposing area of the electrode is S, the initial electrostatic gap is d ₀ , and the gap change amount is Δd. Then, the acceleration a is set to Y shown below, and the correction coefficient is set to A, B... N shown below.
Y = εS × {−2Δd / (d ₀ ² −Δd ² )}
A = −2 · Δd ³ εS / d ₀ ⁴
B = −2 · Δd ⁵ εS / d ₀ ⁶
:
N = −2 · Δd ^{2n + 1} εS / d ₀ ^{2n + 2} When the acceleration is detected by changing the torsion type electrostatic gap d, the automatic correction circuit sets the dielectric constant to ε, the electrode facing area to S, the initial electrostatic gap to d ₀ , and the gap change amount to Δd. The acceleration sensor according to claim 1, wherein the acceleration a is Y shown below, and the correction coefficients are A, B... N shown below.
Y = εS / Δd × ln {(d ₀ ² −Δd ² ) / d ₀ ² }
A = −Δd ³ εS / 2d ₀ ⁴
B = −Δd ⁵ εS / 3d ₀ ⁶
:
N = −Δd ^{2n + 1} εS / {(n × 1) × d ₀ ^{2n + 2} }

Images