CN206772316U - Mems gyroscope and its electronic system - Google Patents

Abstract

Disclosure MEMS gyroscope and its electronic system.Specially a kind of MEMS gyroscope (60,100), wherein, suspended mass (111 114) is removable relative to supporting construction (125,127).The removable mass is influenceed by quadrature error caused by orthogonal torque；Driving structure (77) is coupled to the suspended mass to control the removable movement of mass in the driven direction with driving frequency.The movement of the removable mass on sensing direction is detected coupled to the motion sensing electrode (130) that may move mass, and quadrature compensation electrode (121 124) may move mass to generate the compensating torque opposite with the orthogonal torque coupled to this.The gyroscope is configured for being biased these quadrature compensation electrodes using offset voltage, so that the resonant frequency of the removable mass and the difference of the driving frequency have predeterminated frequency mismatch value.

Description

MEMS gyroscope and electronic system thereof

Technical Field

The utility model relates to a MEMS gyroscope and electronic system thereof.

Background

As is known, Microelectromechanical systems (MEMS) are increasingly used for a variety of applications by virtue of their small size, cost compatible with consumer applications, and increased reliability. Specifically, inertial sensors such as a micro-integrated gyroscope and an electromechanical oscillator using such a technique are formed.

MEMS devices of this type generally comprise a support body and at least one movable mass suspended above and coupled to the support body by springs or "flexures". The spring is configured for enabling the movable mass to oscillate relative to the support according to one or more degrees of freedom. The movable mass is capacitively coupled to a plurality of fixed electrodes on the support body, forming a capacitor with variable capacitance. When the MEMS device is operated as a sensor, the movement of the movable mass with respect to the fixed electrode on the support modifies the capacitance of the capacitor due to the action of the forces acting thereon. This variation allows to detect the displacement of the movable mass with respect to the support and, according to the latter, to detect the external forces that have caused the displacement. Conversely, when the MEMS device is operated as an actuator, an appropriate bias voltage is applied to the movable mass, for example by a separate set of actuation or drive electrodes, so that the movable mass is subjected to an electrostatic force that causes the desired movement.

Among MEMS sensors, in particular gyroscopes have a complex electromechanical structure, which generally comprises at least two masses movable with respect to a support, coupled to each other so as to have a plurality of degrees of freedom (depending on the architecture of the system). In most cases, each movable mass has one or at most two degrees of freedom. The movable mass is capacitively coupled to the support body by the fixed and movable sensing electrodes and by the actuation or drive electrodes.

In an embodiment with two movable masses, the first movable mass is dedicated to driving and keeps oscillating at a resonant frequency with a controlled oscillation amplitude. The second movable mass is driven by the first movable mass with an oscillating motion (translational or rotational) and, when the microstructure rotates around the axis of the gyroscope with an angular velocity, it is subjected to a Coriolis Force (Coriolis Force) proportional to the angular velocity itself and perpendicular to the driving direction. In practice, the second (driven) movable mass acts as an accelerometer, which enables the detection of coriolis forces as well as the detection of angular velocities.

In another embodiment, a single suspension mass is coupled to the support body so as to be movable according to two independent degrees of freedom (i.e. a driving degree of freedom and a sensing degree of freedom) with respect to the support body. The sensing degree of freedom may enable movement along the plane of the movable mass (in-plane movement) or perpendicular to the plane (out-of-plane movement). The actuating or driving device keeps the suspended mass oscillating in a controlled manner according to one of these two degrees of freedom. The suspended mass moves based on another degree of freedom in response to rotation of the support (due to coriolis forces).

However, MEMS gyroscopes have a complex structure and frequently non-ideal electromechanical interactions between the suspended mass and the support body occur, for example due to production defects and process expansions. Thus, the useful signal component is mixed with parasitic components, which do not contribute to the measurement of angular velocity and are potential noise sources, the influence of which is unpredictable.

For example, the drawbacks of the elastic connection between the suspended mass and the supporting body may cause oscillations of the suspended mass in directions that do not coincide exactly with the degrees of freedom desired in the design phase. Such defects may also cause the onset of forces having components oriented along the detection degrees of freedom of angular velocity. This force in turn generates a signal component of unknown amplitude, of the same frequency as the carrier wave, and with a 90 ° phase shift that causes an error known as "quadrature error".

This effect can be understood from a comparison of fig. 1A and 1B representing the movement of an ideal gyroscope 1 (fig. 1A) and a non-ideal gyroscope 1 '(fig. 1B) subject to quadrature errors, respectively, wherein the gyroscopes 1, 1' are represented schematically only with respect to the parts discussed hereinafter.

The gyroscopes 1 and 1' have a sensing mass 5 driven in a first direction (driving direction a, here parallel to the axis X of the cartesian reference system) by a driving unit 6 represented by fixed electrodes 7 (rigidly connected to the substrate, not shown) and movable electrodes 8 (rigidly connected to the sensing mass 5). The sensing unit 10, represented by the fixed electrodes 11 (rigidly connected to the substrate, not shown) and by the movable electrodes 12 rigidly connected to the sensing mass 5, detects the movement of the coriolis force induced in the second direction (sensing direction B, here parallel to the axis Z of the cartesian reference system).

In an ideal gyroscope 1, the sense mass 5 is properly driven in the drive direction a. In contrast, in the non-ideal gyroscope 1', the sense mass 5 is driven in a lateral direction W having a drive component in the sense direction B.

Parasitic movements in the sensing direction B cause detection of movements of the sense mass 5 that will be affected by quadrature errors.

In order to compensate for quadrature errors, it is possible in known gyroscopes to act in various points of the sensing chain.

In particular, a solution enabling gyroscopes to have both high temperature stability and high stability over time is the so-called electrostatic elimination method, which consists in providing electrodes under each suspended mass.

For example, fig. 2 shows in a simplified manner a gyroscope 10 with a suspension structure, here forming four sense masses 11, 12, 13 and 14. Here, the sensing masses 11 to 14 have a generally quadrangular shape (e.g., rectangular shape) and are arranged symmetrically in pairs with respect to the center C of the gyroscope 10 and parallel to the drawing plane (plane XZ) in the rest state. In particular, each having a mass m₁And m₂Are driven along a first drive axis D1 (here parallel to axis X) and are arranged symmetrically to each other with respect to a second drive axis D2 (the drive axes of the third and fourth sense masses 13, 14) perpendicular to the first drive axis D1 and parallel to axis Y. As mentioned, the third and fourth sense masses 13, 14 are arranged symmetrically to each other with respect to the first drive shaft D1 and driven along the second drive shaft D2. In the following, the present description will refer only to the first pair of sensing masses 11, 12, but the following applies also to the second pair of masses 13, 14, obviously taking into account the corresponding drive and sensing axes.

The sense masses 11, 12 are anchored to a substrate (not shown) by a plurality of elastic springs or springs, wherein the figure only shows the springs 16 arranged between the sense masses 11 to 14 and a central mass 15 hinged to the substrate at a center C so as to be rotatable about an axis, not shown, parallel to the axis X, Y and extending through the center C. The spring 16 provides the sense masses 11, 12 with two degrees of freedom and more specifically enables a translational movement along the first drive axis D1 and a sensing movement having a component in a perpendicular direction D3 parallel to the axis Z.

Each sensing mass 11, 12 has an opening 17, 18, respectively, near the center of mass. Two pairs of compensation electrodes 20, 21 and 22, 23, shown in the schematic side view of fig. 3, are arranged below each opening 17, 18, respectively. Fig. 3 also shows a substrate 25, which extends below the plane of the suspension structure of fig. 2, here schematically represented by its mechanical equivalent. In particular, here, the springs 16 for connection to the central mass 15 are schematically represented as hinges, like further springs 26 arranged at opposite ends of the sensing masses 11, 12 in the representation of fig. 3 and connecting the movable masses 11, 12 to a fixed structure 27, which is generally rigidly connected to the substrate 25. The pairs of compensation electrodes 20, 21 and 22, 23 are arranged in the vicinity of the corresponding openings 17, 18, respectively, indicated with dashed lines, such that each compensation electrode 20 to 23 extends half under the corresponding sensing mass 11, 12 and half under the corresponding opening 17, 18.

In the presence of quadrature errors, a non-ideal direction M outside the plane of the gyroscope 10 (plane XZ of fig. 2)_D1、M_D2(fig. 3) upper drive sense masses 11, 12. In particular, the non-ideal direction M_D1、M_D2Possibly having a component along a vertical sensing axis (denoted herein by D3) parallel to axis Z. Subsequently, the gyroscope 10 of fig. 2, 3 may be subjected to an orthogonal force F_Q(these orthogonal forces are assumed to be the same as for the two sense masses 11, 12 and are here shown as being applied at the ends of sense masses 11, 12 coupled to central mass 15), causing sense masses 11, 12 to rotate in a sense direction D3 (not shown, but see fig. 6A).

The electrostatic quadrature error elimination method includes applying the corresponding dc-type compensation voltages V1, V2, V3, V4 to the compensation electrodes 20 to 23. In particular, the applied compensation voltages V1-V4 typically have the following values:

V1＝V3＝VCM-ΔV

V2＝V4＝VCM+ΔV

where VCM and Δ V are determined during the calibration phase for each gyroscope 10. By rotor voltage V_RFurther biasing the sensing masses 11, 12.

In practice, the compensation electrodes 20 to 24 generate a compensation force F_C(here shown as applied at an intermediate position relative to electrodes 20-23) is intended to mechanically balance out normal forces F_QThe movement in the sensing direction D3 due to the quadrature error is eliminated.

However, this does not completely solve the problem. In fact, the compensation electrodes 20 to 24 increase the range of motion due to the asymmetry and unbalance of the gyroscope structure. Further, the application of the additional voltage increases the electrostatic softening effect, i.e. the variation of the resonant frequency of these devices due to the variation of the elastic constant of the gyroscope caused by the potential difference existing between the movable part and the fixed part. For example, in the case of the electrostatic compensation discussed above, the resonant frequency f may be required for the electrostatic softening effect₀Sensible change (from 22kHz to 18 kHz). This variation is commonly referred to as "frequency mismatch".

On the other hand, the presence of a high frequency mismatch determines a significant change in the sensitivity of the gyroscope and a degradation of its performance.

To overcome the problem of frequency mismatch due to electrostatic softening, it is possible to introduce further frequency tuning electrodes, one for each sensing axis of the gyroscope.

In this way, the gyroscope has fifteen electrodes (four electrodes for quadrature error compensation plus three electrodes for frequency mismatch compensation for each sense axis). Subsequently, as shown in fig. 4, at least nine drivers 30, 31 and at least six DACs 32, 33 would be required.

In fact, in this manner, with six driver circuits 30 and three DACs 32, equation [1 ] provided below is used]The quadrature error cancellation electrodes designated by Q1X, Q2X, Q1Y, Q2Y, Q1Z, Q2Z are driven in a differential manner (it should be noted that fig. 4 only shows one of the two electrode sets at the same voltage). For a constant common-mode voltage VCM as shown in fig. 5A, the differential voltages 2 Δ Vx, 2 Δ Vy, 2 Δ Vz applied to each pair of electrodes are optimized to generate a cancellation quadrature error Q_γCompensated quadrature Q of_el. Specifically, fig. 5A shows: with this solution, the applied compensation voltages V1, V2 depend linearly on the compensated quadrature Q to be generated_el。

In contrast, fig. 5B shows the resonant frequency f resulting from the electrostatic softening according to the applied tuning voltage VQ discussed above₀A change in (c). As shown in fig. 4, three single-ended DACs 33 may be used to apply the tuning voltage VQ.

This approach will cause the device size due to the presence of the further electrodes Q3x, Q3y, Q3z to increase and the corresponding drive circuits (DAC 33 and buffer 31) to increase in size.

SUMMERY OF THE UTILITY MODEL

An object of the utility model is to provide a can overcome the solution of prior art's shortcoming.

According to the present invention, as defined in the appended claims, a MEMS gyroscope, a method for controlling a MEMS gyroscope, and a method for setting compensation parameters of a MEMS gyroscope are provided.

Specifically, a MEMS gyroscope (60, 100) is provided, comprising: a support structure (125, 127); a mass (111) movable relative to the support structure in a drive direction (D1) and a sense direction (D3) perpendicular to each other, the movable mass being affected by quadrature errors caused by quadrature moments; a drive structure (77) coupled to the movable mass for controlling movement of the movable mass in the drive direction at a drive frequency; motion sensing electrodes (130) coupled to the movable mass for detecting movement of the movable mass in the sensing direction; and quadrature compensation electrodes (121-124) coupled to the movable mass for generating a compensation moment opposite to the quadrature moment; the movable mass has a variable resonant frequency, the difference between the resonant frequency and the drive frequency being a frequency mismatch; the gyroscope is configured to bias the quadrature compensation electrodes with a compensation voltage to drive the movable mass with a preset frequency mismatch.

According to one embodiment, the compensation voltage may vary quadratically with the quadrature error.

According to one embodiment, the quadrature compensation electrode (121-124) comprises a first quadrature compensation electrode and a second quadrature compensation electrode configured for being respectively at a first compensation voltage V₁And a second compensation voltage V₂Is biased, wherein V₁And V₂Is selected to satisfy the equation:

wherein: Δ f₀Is the preset frequency mismatch; v_RIs a bias voltage of the movable mass (111) and (114); omega_S0Is the resonant frequency of the movable mass; ks/J is a parameter related to a mechanical constant of the movable mass; and f_dIs the drive frequency.

According to one embodiment, the first compensation voltage V₁And a second compensation voltage V₂Satisfies the equation:

Q_el+Q_γ＝0

wherein: q_elIs the compensation quadrature given by the following equation:

Q_el＝k_Q[(V_R-V₁)²-(V_R-V₂)²]，

Q_γis the quadrature error, k_qIs to make the compensation quadrature Q_elWith said compensation moment M_elThe associated proportionality constant.

According to one embodiment, a memory element (91) is included, the memory element being configured to store a value of the compensation voltage.

An electronic system is also provided, comprising a control unit (410) and the above-mentioned MEMS gyroscope (60), which is coupled to the control unit (410).

In practice, the present gyroscope is configured such that the same quadrature error compensation electrodes function to also control frequency mismatch. This is achieved by biasing the quadrature error compensation electrode with a voltage having a parabolic relationship with the frequency mismatch, as discussed in detail below.

Drawings

For a better understanding of the present invention, a preferred embodiment thereof will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

fig. 1A and 1B are schematic representations of the movement of an ideal gyroscope and a non-ideal gyroscope subject to quadrature errors, respectively;

figure 2 is a simplified top plan view of a known gyroscope with quadrature error compensation;

fig. 3 is a schematic cross section of the gyroscope with quadrature error compensation of fig. 2;

fig. 4 shows a simplified circuit diagram of the driving circuit of the gyroscope of fig. 2 and 3;

figures 5A and 5B show the compensation quantities that can be used with the gyroscopes of figures 2 and 3;

fig. 6A is a schematic cross section of a gyroscope with quadrature error;

FIG. 6B is a schematic cross section of an embodiment of the present gyroscope;

FIG. 6C is a detailed top plan view of the gyroscope of FIG. 6B;

fig. 7 shows a graph of the amount of electrical compensation that can be used for the gyroscope of fig. 6B for a desired frequency mismatch value as a function of the variation of the elastic characteristics of the gyroscope of fig. 6B;

FIG. 8 shows the compensated quadrature Q for a desired frequency mismatch value as a function of the variation of the elastic characteristic and the compensation quantity of the gyroscope of FIG. 6B_elThe curve of (d);

figure 9 shows a graph of the amount of electrical compensation for the gyroscope of figure 6B as a function of the compensation quadrature;

FIG. 10 shows a flow chart of a test method for setting the amount of electrical compensation for the gyroscope of FIG. 6B;

fig. 10A shows a table used in the test method of fig. 10;

fig. 11 shows a block diagram of a device comprising the gyroscope and the control section of fig. 6B;

fig. 12 shows a circuit diagram of the driving part of the gyroscope of fig. 6B; and is

Figure 13 shows a simplified block diagram of an electronic device incorporating the gyroscope of figure 11.

Detailed Description

As mentioned, the present gyroscope is designed such that the compensation electrodes intended to eliminate the quadrature error also adjust for the frequency mismatch so that this has an imposed preset value.

For this purpose, reference may be made to fig. 6A and 6B, which schematically show an uncompensated gyroscope 300 and a gyroscope 60 with quadrature error compensation and frequency adjustment according to an embodiment of the invention, respectively. The gyroscopes 300, 60 have the same basic structure as the gyroscope 10 of fig. 2 and 3. Subsequently, in fig. 6A to 6C, parts similar to those of fig. 3 are denoted by the same reference numerals increased from 100 in fig. 6B and 6C and by the same reference numerals increased from 300 in fig. 6A.

In detail (fig. 6A), the gyroscope 300 comprises a pair of movable masses 311, 312, which in the rest state extend parallel to the axes X and Y of cartesian space XYZ, are driven in a driving direction D1 parallel to axis X, and oscillate due to coriolis forces so as to have a movement component oriented in a sensing direction D3 parallel to axis Z. The movable masses 311, 312 are hinged to a central mass 315 via a first spring 316 and to a fixed structure 327 via a second spring 326.

Fig. 6A shows the effect of quadrature error on the system of quality blocks 311, 312. As shown, the quadrature error is caused by having a component in the direction Z, acting on the two masses 311, 312 and generating a quadrature moment M_γParasitic force D of_eAnd then produced.

As can be noted, the orthogonal moment M will be_γHalf value (M)_γ/2) is applied to each movable mass 311, 312 and the orthogonal moment has an arm b_γThe arm is equidistant from the center of mass of each movable mass 311, 312 to a vertical line extending through the second spring 326. From fig. 6A, it can be further noted that, as described in detail in us patent application 2015/0114112, this moment is due to the movable masses 311, 312 passing the first massThe interconnection of the springs 316 and the central mass 315 causes rotation of each movable mass in opposite directions.

Fig. 6B shows a gyroscope 60 having a basic structure similar to that of gyroscope 300, but having a compensation structure. In particular, the gyroscope 60 has compensation electrodes 121 to 124 which extend under the movable masses 111, 112 and more precisely under the openings 117, 118 of the movable masses 111, 112. The compensation electrodes 121 to 124 have a rectangular shape and size as shown in fig. 6C, and more precisely have a length L₀(in the direction X parallel to the drive direction D1) and a width p (in the direction Y). Further, the compensation electrodes 121 to 124 are arranged at a distance a (distance measured in direction Z) from the movable masses 111, 112.

Fig. 6B further illustrates fixed sensing electrodes 130, 131 arranged under the respective movable masses 111, 112 and capacitively coupled to the movable masses. In a known manner and not discussed herein, the fixed sensing electrodes 130, 131 supply output voltage signals s1, s2 related to the displacement of the movable mass in a sensing direction D3, here oriented parallel to the axis Z.

Fig. 6B also shows that the compensation electrodes 121 to 124 are generated on both movable masses 111, 112 and are intended to couple the orthogonal moments M_γCompensating moment M set to zero_el. In a manner not shown and described hereinafter, the compensation moment M_elFrom four components M_el1To M_el4Given that each component is generated by a corresponding compensation electrode 121 to 124. These components have corresponding arms b₁、b₂、b₃、b₄These arms are equidistant from each compensation electrode 121 to 124 from a vertical line extending through a corresponding second spring 126 for coupling to the fixed structure 127.

Based on these assumptions, the total compensation moment M_elEqual to:

M_el＝(M_el1+M_el2)-(M_el3+M_el4) [1]

wherein:

and is₀Is the vacuum dielectric constant; l is₀P is the size of the compensation electrodes 121 to 124 shown in fig. 6C; x is the number of_dIs the amplitude of the driving movement in direction D1;ais the distance between the compensation electrodes 121 to 124 and the plane of the movable masses 111, 112 in the rest state; v_RIs the voltage applied to the central mass 115; v₁、V₂、V₃、V₄Is a compensation voltage applied to the compensation electrodes 121 to 124; and b is₁、b₂、b₃、b₄Is the distance of the above-mentioned compensation electrodes 121 to 124.

By setting b₁＝b₄And b is₂＝b₃And applying a compensation voltage V₁＝V₃And V₂＝V₄We obtain:

consider thatAccording to the preceding equation, the compensation moment is given by the following equation:

compensating moment M_elCan be used to correct the quadrature error Q_γInduced orthogonal moment M_γSet to zero, generating and compensating a moment M_elProportional compensated quadrature Q_el. It can thus be written as:

Q_el＝k_Q[(V_R-V₁)²-(V_R-V₂)²][3]

wherein,is to compensate for the quadrature Q_elCompensating moment M orthogonal to the generated compensation_elThe associated proportionality constant.

As can be noted, the quadrature Q is compensated_elDependent on the compensation voltage V₁、V₂. Therefore, by applying a differential voltage Δ V between the compensation electrodes 121 and 122 equal to the differential voltage applied between the compensation electrodes 123 and 124, it is possible to generate a voltage causing a drive-dependent movement x_dAnd depends on the compensation voltage V₁、V₂Compensating moment M_elThe electrostatic force of (2). Such electrostatic forces can thus be used to cancel the quadrature moment according to the following equation:

M_el+M_γ＝0 [4]

this equation corresponds to such that:

Q_el+Q_γ＝0 [4’]

as shown below, by pairing V₁And V₂Appropriate selection of the relationship betweenVia the compensation electrodes 121 to 124 it is further possible to adjust the frequency mismatch.

In fact, the compensation voltage V is applied to the compensation electrodes 121 to 124₁To V₄The change in the overall spring constant of the system of masses 111, 112 and thus the resonance frequency f is determined_s. In particular, the resonance frequency f_sGiven by the following equation:

wherein J is the moment of inertia, K_mIs the mechanical elastic constant, K_{el_s}Is an electrostatic elastic constant due to a potential difference applied between the central mass 115 and the fixed sensing electrodes 130, 131 (FIG. 6B), and K_{el_q}Is the electrostatic spring constant due to the differential voltage applied between the central mass 115 and the compensation electrodes 121, 122, 123, 124.

Specifically, the electrostatic elastic constant is given by the following equation:

consider V₁＝V₃And V is₂＝V₄We get:

wherein

From equation [6]]It can be seen that the change in voltage across the electrodes 121 to 124 requires a resonant frequency f_sA change in (c). Thereby, frequencyMismatch Δ f₀Equal to:

point out

And substituting equation [6] into equation [7], we obtain

It should be noted that in equation [9 ]]Middle, omega_S0When is at V_RThe sensing resonance frequency when the compensation electrodes 121 to 124 are biased.

Equation [3 ]]It is shown that quadrature Q is compensated for because the two terms in square brackets are subtracted_elDepending on the differential voltage between the electrodes 121 to 124. In contrast, equation [9 ]]It is shown that the frequency mismatch Δ f is due to the addition of the same two terms in square brackets₀Depending on the common mode voltage of these electrodes. Then, for a given quadrature error Q_γIt is possible to find a frequency mismatch Δ f that compensates for the quadrature error and enables gyroscope 60 to be mismatched at the desired frequency₀Value-operated single pair compensation voltage V₁And V₂The value is obtained.

For a predetermined frequency mismatch value Δ f₀(here, 1kHz) and a different spring constant k that increases from-12% (for the bottom curve) to 12% (for the top curve) in 2% increments of the theoretically expected value_sValue, considering that the compensation voltage V applied to the compensation electrodes 121 to 124 in the gyroscope 60 of fig. 6B is shown₁、V₂This is particularly clear in fig. 7 and 8, which are a ratio of each other.

It should be noted that the curves and curves of fig. 7 and 8 are relative to each otherApplying a compensation voltage V₁、V₂Correspondingly, these compensation voltages generate a modulus equal to Q_γAnd of opposite sign_elAnd follows the law represented in figure 9. As can be noted, the curve of the compensation voltage applied according to the frequency mismatch is no longer linear as in fig. 5A, but quadratic, and the common mode voltage is no longer constant.

The test method for the present gyroscope will be described hereinafter. In fact, equation [9 ]]There are three unknowns: driving frequency f_dThe drive frequency may vary with respect to a design value; sensing resonant frequency omega_S0(ii) a And k_sand/J. Of these unknowns, f can be measured directly_dBut can be compensated by applying a suitable compensation voltage V₁、V₂And measuring the frequency mismatch Δ f₀To indirectly measure omega_S0And k_s/J。

Knowing these three variables f_d、ω_S0And k_sJ and seeks to provide a predetermined frequency mismatch Δ f_0d(here, 1kHz) of a compensation voltage V₁、V₂By value pair, it is possible to understand which of the curves of fig. 7 describes the behavior of a particular gyroscope 60 under test, i.e. which curve is the elastic constant k affecting this gyroscope_sA change in (c). Using the curves of FIG. 8 and the spring constant k determined therefrom_sIs changed in percentage, it is possible to determine the elimination of the quadrature error Q_γFirst compensation voltage V₁The value of (c). This can again be based on the previously identified curve of fig. 7 or on equation [9 ]]To obtain a second compensation voltage V₂The value of (c).

According to an aspect of the present description, during testing, each gyroscope 60 is tested to determine a compensation voltage V to be applied during operation₁、V₂The value of (c). In particular, the test procedure may comprise the following steps (see also the flow chart of fig. 10):

measuring the drive frequency f_dStep 200;

by applying an appropriate compensation voltage V as indicated above₁、V₂Value to indirectly measure the sensing resonance frequency omega_S0And a parameter k_s/J, step 202;

-obtaining the desired frequency mismatch Δ f_0dA value (which is inversely related to the system gain and is typically fixed during the design phase), step 204;

by a step Δ V (e.g. 1V, from 1V to V)_R(e.g., 10V)) applying a plurality of compensation voltages V₁、V₂Value and corresponding frequency mismatch deltaf is measured₀A value; the frequency mismatch Δ f thus obtained₀The values are saved, for example, in a table 220, shown illustratively in FIG. 10A, step 206;

search in table 220 by expected frequency mismatch Δ f_0dPoints characterized by values, step 208; in practice, the set of points represents the elastic constant k that may be imposed on the gyroscope under test_sA curve corresponding to the change in (f) (from among the curves shown in fig. 7);

identifying a generating-compensated quadrature Q among the set of points identified in step 208_elFirst compensation voltage V₁The compensation is orthogonal in equation [3 ]]And equation [4 ]]On the basis of compensating the quadrature error Q_γStep 210;

from the curve identified in step 208 or using equation [9 ]]To identify the second compensation voltage V₂The point value of (2), step 212; and

-compensating voltage V₁、V₂The value pairs are stored in a memory associated with the gyroscope under test, step 214, as described below with reference to fig. 11.

Fig. 11 shows a block diagram of an electronic device 100 using the frequency mismatch compensation principle described above.

In fig. 11, the electronic device 100 includes a gyroscope 60 (here, represented via its functional blocks) and a control element 65.

The gyroscope 60 is integrated in a schematically represented semiconductor chip 70 and has the same structure as the gyroscope 10 of fig. 2. Accordingly, elements described with reference to fig. 6B are denoted by the same reference numerals. In detail, the central mass 115 is anchored to a substrate (similar to substrate 125 of fig. 6B), not shown, which is rotatable about an axis perpendicular to gyroscope 60 (parallel to axis Z) and extends through center C. The central mass 115 is coupled to four movable masses 111 to 114, only one of which (movable mass 117) is shown in more detail in fig. 11, via a first spring 116. The movable masses 111 to 114 are arranged symmetrically in pairs with respect to the center C and extend in parallel to the drawing plane (plane XY) of the gyroscope of fig. 2 in the rest state. The movable masses 111 to 114 have two degrees of freedom and are subject to a driving motion along a corresponding drive axis and to a sensing motion having a component along a corresponding sensing axis perpendicular to the drive axis. In a manner known per se, the movable masses 111 to 114 thus have a similar basic structure but are configured for detecting movements about different axes. The present description thus refers, mutatis mutandis, only to the movable mass 111 shown in more detail, but may also be applied to the other movable masses 112 to 114.

In particular, the movable mass 111 is subjected to a driving motion along a driving axis D1 parallel to axis X and to a sensing motion having a component along a sensing axis D3 parallel to axis Z.

Here, the movable mass 111 is schematically represented by the sense mass 72 and the compensation mass 73 rigidly connected with respect to each other. The sense mass 72 may have a generally trapezoidal shape as shown in FIG. 2. The proof mass 73 has a generally hollow rectangular shape, covering the pair of compensation electrodes 121, 122. In particular, an opening 117 is formed in the proof mass 73 and has two sides 117A, 117B parallel to the axis Y. Each compensation electrode 121, 122 extends along a respective side 117A, 117B of opening 117, half under proof mass 73 and half under opening 117.

The movable mass 111 is connected via a second spring 76 enabling a sensing movement to an actuation module 77 driving the movable mass 111 in a driving direction D1, and by a third spring 78 to a sensor having a detected effective driving parameter (including a driving frequency f as discussed above) for sensing the driving movement 79_d) And (5) modules of tasks.

The actuation module 77 and the module for sensing the driving movement 79 are connected to a driving control module 85 formed in the control element 65. The control element 65 is typically integrated in a different chip 90 and comprises, for example, an ASIC (application specific integrated circuit).

Control element 65 further includes a sense module 88 connected to sense electrode 130, storing a compensation voltage V calculated during testing as discussed above₁、V₂A parameter memory 91 of values, a DAC 92 and a buffer 93. DAC 92 forms voltage sources, each configured to apply a compensation voltage V₁、V₂These compensation voltage values are specified by the contents of the memory 91.

For example, each DAC 92 and associated buffer 93 may be arranged as shown in fig. 12. In practice, in the illustrated embodiment, DAC 92 is coupled to a reference voltage V_{Reference to}And a resistive voltage divider 95 between the masses and comprising a plurality of resistors 96 which may be coupled to the buffer 93 by means of switches 97 driven on the basis of the contents of the memory 91.

The MEMS gyroscope 60 described herein thus enables the frequency mismatch to be electrostatically tuned using the same electrodes that tune the quadrature error and thus has a reduced size and low consumption level.

Fig. 13 illustrates a portion of an electronic system 400 that incorporates electronic device 100 and that may be used in an apparatus such as a palmtop computer (personal digital assistant, PDA), a laptop or portable computer that may have wireless capability, a cellular telephone, a messaging device, a digital music player, a digital camera, or other apparatus designed to process, store, transmit, or receive information. For example, the electronic device 100 may be used in a digital camera to detect movement and perform image stabilization. In a possible embodiment, the electronic device 100 is included in a motion-activated user interface for a computer or video game console. In a further embodiment, the electronic device 100 is incorporated in a satellite navigation device and used for temporary position tracking in case of loss of satellite positioning signals.

The electronic system 400 of fig. 13 includes a control unit 410, an input/output (I/O) unit 420 (e.g., a keyboard or a display), the electronic device 100, a wireless interface 440, and a volatile or non-volatile type memory 460, which are coupled to each other by a bus 150. Alternatively, the memory 460 may be internal to the control unit 410 or may replace the memory 91 internal to the control element 90 of fig. 11 and store parameters and quantities, such as the compensation voltage V, that may be used to operate the electronic device 100₁、V₂And the like. In one embodiment, a battery 480 may be used to power the system 400. However, the electronic system 400 may even include only some of the units shown in fig. 13.

The control unit 410 may comprise, for example, one or more microprocessors, microcontrollers, or the like. In different embodiments, it may integrate the functionality of the control element 90 of fig. 11, and the electronic device 100 of fig. 13 may be formed by a gyroscope 60.

I/O unit 420 may be used to generate messages. System 400 may use wireless interface 440 to transmit and receive messages to and from a wireless communication network (not shown) via Radio Frequency (RF) signals. Examples of a wireless interface may include an antenna, a wireless transceiver such as a dipole antenna, although the scope of the invention is not limited in this respect. Further, I/O cell 420 may provide a voltage representative of the stored content as a digital output (if digital information has been stored) or as an analog output (if analog information has been stored).

Finally, it is clear that modifications and variations can be made to the gyroscope, to the control method and to the regulation method described and illustrated herein, without thereby departing from the scope of the present invention as defined in the appended claims.

Claims (6)

1. A MEMS gyroscope (60, 100), comprising:

a support structure (125, 127);

a mass (111) movable relative to the support structure in a drive direction (D1) and a sense direction (D3) perpendicular to each other, the movable mass being affected by quadrature errors caused by quadrature moments;

a drive structure (77) coupled to the movable mass for controlling movement of the movable mass in the drive direction at a drive frequency;

motion sensing electrodes (130) coupled to the movable mass for detecting movement of the movable mass in the sensing direction; and

quadrature compensation electrodes (121) coupled to the movable mass to generate a compensation moment opposite to the quadrature moment;

the movable mass has a variable resonant frequency, the difference between the resonant frequency and the drive frequency being a frequency mismatch;

the gyroscope is configured to bias the quadrature compensation electrodes with a compensation voltage to drive the movable mass with a preset frequency mismatch.

2. The gyroscope of claim 1, wherein the compensation voltage varies quadratically with the quadrature error.

3. The gyroscope of claim 1, wherein the quadrature compensation electrode (121) comprises a first quadrature compensation electrode and a second quadrature compensation electrode, the first quadrature compensation electrode and the second quadrature compensation electrode being configured for being respectively at a first compensation voltage V₁And a second compensation voltage V₂Is biased, wherein V₁And V₂Is selected to satisfy the equation:

<mrow> <msub> <mi>&amp;Delta;f</mi> <mn>0</mn> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mn>2</mn> <mi>&amp;pi;</mi> </mrow> </mfrac> <msqrt> <mrow> <msubsup> <mi>&amp;omega;</mi> <mrow> <mi>s</mi> <mn>0</mn> </mrow> <mn>2</mn> </msubsup> <mo>-</mo> <mfrac> <msub> <mi>k</mi> <mi>S</mi> </msub> <mi>J</mi> </mfrac> <mo>&amp;lsqb;</mo> <msup> <mrow> <mo>(</mo> <msub> <mi>y</mi> <mi>R</mi> </msub> <mo>-</mo> <msub> <mi>V</mi> <mn>1</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>+</mo> <msup> <mrow> <mo>(</mo> <msub> <mi>V</mi> <mi>R</mi> </msub> <mo>-</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>&amp;rsqb;</mo> </mrow> </msqrt> <mo>-</mo> <msub> <mi>f</mi> <mi>d</mi> </msub> </mrow>

wherein:

Δf₀is the preset frequency mismatch;

V_Ris a bias voltage of the movable mass (111) and (114);

ω_S0is the resonant frequency of the movable mass;

ks/J is a parameter related to a mechanical constant of the movable mass; and is

f_dIs the drive frequency.

4. The gyroscope of claim 3, wherein the first compensation voltage V₁And said second compensation voltage V₂Satisfies the equation:

Q_el+Q_γ＝0

wherein:

Q_elis the compensation quadrature given by the following equation:

Q_el＝k_Q[(V_R-V₁)²-(V_R-V₂)²]，

Q_γis the error in the quadrature phase of the signal,

k_qis to make the compensation quadrature Q_elWith said compensation moment M_elThe associated proportionality constant.

5. The gyroscope according to any of claims 1 to 4, comprising a memory element (91) configured for storing a value of the compensation voltage.

6. An electronic system, characterized in that it comprises a control unit (410) and a MEMS gyroscope (60) according to any one of claims 1 to 5, the MEMS gyroscope being coupled to the control unit (410).