Numerous items such as smart phones, smart watches, tablets, automobiles, aerial drones, appliances, aircraft, exercise aids, and game controllers may utilize motion sensors during their operation. In many applications, various types of motion sensors such as accelerometers and gyroscopes may be analyzed independently or together in order to determine varied information for particular applications. For example, gyroscopes and accelerometers may be used in gaming applications (e.g., smart phones or game controllers) to capture complex movements by a user, drones and other aircraft may determine orientation based on gyroscope measurements (e.g., roll, pitch, and yaw), and vehicles may utilize measurements for determining direction (e.g., for dead reckoning) and safety (e.g., to recognizing skid or roll-over conditions).
Motion sensors such as accelerometers and gyroscopes may be manufactured as microelectromechanical (MEMS) sensors that are fabricated using semiconductor manufacturing techniques. A MEMS sensor may include movable proof masses that can respond to forces such as linear acceleration (e.g., for MEMS accelerometers) and angular velocity (e.g., for MEMS gyroscopes). The operation of these forces on the movable proof masses may be measured based on the movement of the proof masses in response to the forces. In some implementations, this movement is measured based on distance between parallel surfaces of the movable proof masses and sense electrodes, which form capacitors for sensing the movement.
The capacitance is based on distance and the voltages of the proof mass and sense electrode. However, other components of the system such as a shield layer on a substrate or cap may also have a voltage and may be located in positions (e.g., parallel) relative to the proof mass such that these other components also form a capacitor with the proof mass. Based on the relative voltage of the movable proof mass and fixed shield, this capacitor may result in a force on the proof mass and a resulting displacement of the proof mass. This displacement of the proof mass occurs is in the absence of any inertial and results in an error when attempting to measure a sensed motion of the proof mass (e.g., as a result of linear acceleration or angular velocity).
In an exemplary embodiment of the present disclosure, a microelectromechanical (MEMS) sensor may comprise a proof mass having a plurality of planar surfaces and having a proof mass voltage. The MEMS sensor may also comprise a sense electrode having one or more planar surfaces, wherein at least one of the one or more planar surfaces of the sense electrode is located in parallel to at least one of the planar surfaces of the proof mass, wherein the sense electrode has a sense electrode voltage, and wherein the sense electrode and the proof mass form a sense capacitor. The MEMS sensor may also comprise a shield located on one or more planar surfaces of the microelectromechanical sensor, wherein at least a portion of the shield is located in parallel to at least one of the planar surfaces of the proof mass, and wherein the shield has a modifiable shield voltage. The MEMS sensor may also comprise processing circuitry coupled to the proof mass, the sense electrode, and the shield, wherein the processing circuitry provides a plurality of test voltages for the modifiable shield voltage, measures an offset value at each of the test voltages, and sets the modifiable shield voltage to an operating shield voltage based on the plurality of measured offset values, and wherein the movement of the proof mass relative to the sense electrode is sensed based on changes in the value of the sense capacitor.
In an exemplary embodiment of the present disclosure, a method for operating a microelectromechanical (MEMS) sensor may comprise providing, to a shield of the MEMS sensor, a plurality of test voltages, determining, based on measured movement of a proof mass of the MEMS sensor, an offset value associated with each of the plurality of test voltages, identifying, based on the measured movements of the proof mass, a voltage-induced sensor offset associated with an operating voltage of the shield, and identifying, based on the measured movements of the proof mass, a mechanical sensor offset for the proof mass. The method may further comprise measuring, based on a capacitor formed by the proof mass and a sense electrode of the MEMS sensor, a measured value for the sensor, and correcting the measured value based on the voltage-induced sensor offset and the mechanical sensor offset.
In an exemplary embodiment of the present disclosure, a microelectromechanical (MEMS) sensor may comprise a proof mass having a plurality of planar surfaces and having a proof mass voltage. The MEMS sensor may further comprise a sense electrode having one or more planar surfaces, wherein at least one of the one or more planar surfaces of the sense electrode is located in parallel to at least one of the planar surfaces of the proof mass, wherein the sense electrode has a sense electrode voltage, and wherein the sense electrode and the proof mass form a sense capacitor. The MEMS sensor may further comprise a shield located on one or more planar surfaces of the microelectromechanical sensor, wherein at least a portion of the shield is located in parallel to at least one of the planar surfaces of the proof mass. The MEMS sensor may further comprise processing circuitry coupled to the proof mass, the sense electrode, and the shield, wherein the processing circuitry provides a plurality of test voltages to one or more of the proof mass or the shield, measures an offset value at each of the test voltages, identifies a minimum offset value based on the measured offset values, determines a voltage-induced sensor offset based on the minimum offset value and an operating voltage of the shield, determines a mechanical sensor offset based on the minimum offset value, and modifies a sensed signal from the sense capacitor based on the voltage-induced sensor offset and the mechanical sensor offset.
The above and other features of the present disclosure, its nature and various advantages will be more apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings in which:
A MEMS device is constructed of a number of layers such as a CMOS layer, a MEMS device layer, and a cap layer. The MEMS device layer includes a movable proof mass that is suspended from the MEMS device layer by components such as springs masses, and lever arms. At least one sense electrode is located parallel to a surface of the proof mass (e.g., on a substrate such as the top of the CMOS layer, or on a post that extends into the MEMS plane) for use in sensing a position or orientation of the proof mass. Each of the proof mass and sense electrode are conductive and have respective voltages. At least a portion of the proof mass includes a planar surface that is located opposite and parallel to the sense electrode to form a capacitor. The proof mass is suspended in a manner such that its primary movement relative the sense electrode is due to an inertial force that is desired to be measured, such as linear acceleration along the axis along which the proof mass is displaced, or a Coriolis force along the sense axis due to an angular velocity that is perpendicular to the sense axis. Processing circuitry measures the capacitance based on signals received from the sense electrode or proof mass, to determine a value indicative of the movement of the electrode. Based on a change in the capacitance and scaling factors, the processing circuitry determines a motion parameter indicative of motion (e.g., linear acceleration or angular velocity) of the MEMS device. As an example, the MEMS device may form an accelerometer, gyroscope, pressure sensor, or other type of motion sensor.
In an embodiment, one or more other portions or components of the MEMS device may have one or more voltages that are independent from either of the sense electrode and the proof mass. For example, a portion of additional electrodes substrate, cap, CMOS layer, MEMS layer, anchor, frame, or other similar component may have such a voltage, and may also be located at a position relative to the proof mass such that a capacitor is formed with at least a portion of the proof mass. Although it will be understood that the present disclosure may apply to a variety of components located at a variety of locations relative to the proof mass, in an exemplary embodiment described herein an electrode shield is located on a substrate surface (e.g., of a CMOS layer) that is parallel to a planar surface of the proof mass and that surrounds the sense electrodes. Because the shield is located on a fixed surface while the proof mass is movable, the capacitor formed by the shield and the proof mass may exert a force on the proof mass, which may cause the proof mass to move relative to the shield and the sense electrode. As a result of this voltage-induced sensor offset, an error is induced in the measured response to an inertial force along the sense axis.
The voltage of the shield may be modified which may also result in a change in the voltage-induced sensor offset. In an embodiment, this shield voltage may be set to an initial value based on an expected voltage at which the voltage-induced sensor offset is at a minimum. The shield voltage may then be varied in order to determine whether the initial offset is correct, and if not, to modify the initial offset. Such testing may be performed in a variety of manners, such as by performing a sweep of possible shield voltages or iterative searching based on measured absolute and/or derivative (slope) values for the offset. In this manner, a shield voltage that is associated with a minimum offset available value (e.g., based on applied shield voltage resolution) may be determined. In some instances, a sensor offset may exist even when the voltage-induced sensor offset is minimized. This mechanical sensor offset may be a result of mechanical factors (e.g., manufacturing tolerances, fabrication imperfections, stress-induced deformation), other system voltages that are not adjustable, or other similar factors such as electrical impacts of circuits within signal paths (e.g., ADC offset or differential capacitance mismatch within an output path).
Once the shield voltage that is associated with the minimum voltage-induced sensor offset is determined, processing may be performed in order improve the accuracy of the MEMS sensor. Compensation may be performed based on the determined mechanical sensor offset in order to remove the impact of the mechanical sensor offset from the signals that are sensed during normal operations. In some embodiments, the initial shield voltage may be retained but compensation may be performed based on the voltage-induced sensor offset at the initial voltage. In some embodiments, the shield voltage may be modified in order to remove some or all of the voltage-induced sensor offset. If the shield voltage is set to a revised voltage that corresponds to the minimum voltage for voltage-induced sensor offset, then it may be unnecessary to perform compensation for voltage-induced sensor offset. If another revised shield voltage is selected, compensation may be performed based on the voltage-induced sensor offset that is associated with the selected revised voltage.
Processing circuitry 14 may include one or more components providing necessary processing based on the requirements of the motion processing system 10. In some embodiments, processing circuitry 14 may include hardware control logic that may be integrated within a chip of a sensor (e.g., on a substrate or cap of an inertial sensor 12 or other sensor 18, or on an adjacent portion of a chip to the inertial sensor 12 or other sensor 18) to control the operation of the inertial sensor 12 or other sensor 18 and perform aspects of processing for the inertial sensor 12 or other sensor 18. In some embodiments, the inertial sensor 12 and other sensors 18 may include one or more registers that allow aspects of the operation of hardware control logic to be modified (e.g., by modifying a value of a register). In some embodiments, processing circuitry 14 may also include a processor such as a microprocessor that executes software instructions, e.g., that are stored in memory 16. The microprocessor may control the operation of the inertial sensor 12 by interacting with the hardware control logic, and process signals received from inertial sensor 12. The microprocessor may interact with other sensors in a similar manner.
In an embodiment, processing circuitry 14 may perform steps to eliminate and/or compensate for voltage-induced sensor offset and mechanical sensor offset as described herein of any of the sensor 12 or sensors 18. At one or more stages of the life cycle of any such sensor (e.g., manufacturing, final inspection, initial startup in the field, upon each application of power, periodically, after extended periods without experiencing an inertial force, or other suitable times), the processing circuitry may perform testing of the sensor offsets by modifying the shield voltage (or in some embodiments, other voltages or multiple voltages) while measuring the response of the proof mass to the modified shield voltage. Based on the results, the minimum shield voltage that corresponds to a minimum proof mass response may be associated with a minimum (e.g., substantially zero) voltage-induced sensor offset. In some embodiments, a mechanical sensor offset may also be determined, based on any remaining offset at the minimum shield voltage. The processing circuitry may compensate for the mechanical sensor offset, and in some embodiments, compensate for the voltage-induced sensor offset at a voltage other than the minimum shield voltage. In some embodiments, the operational shield voltage may be modified (e.g., to the minimum shield voltage) to eliminate or reduce the voltage-induced sensor offset.
Although in some embodiments (not depicted in
In some embodiments, certain types of information may be determined based on data from multiple inertial sensors 12 and sensors 18, in a process that may be referred to as sensor fusion. By combining information from a variety of sensors it may be possible to accurately determine information that is useful in a variety of applications, such as image stabilization, navigation systems, automotive controls and safety, dead reckoning, remote control and gaming devices, activity sensors, 3-dimensional cameras, industrial automation, and numerous other applications.
An exemplary MEMS inertial sensor (e.g., inertial sensor 12) may include one or more movable proof masses that are configured in a manner that permits the MEMS inertial sensor (e.g., a MEMS accelerometer or MEMS gyroscope) to measure a desired force (e.g., linear acceleration or angular velocity) along an axis. In some embodiments, the one or more movable proof masses may be suspended from anchoring points, which may refer to any portion of the MEMS sensor which is fixed, such as an anchor that extends from a layer (e.g., a CMOS layer) that is parallel to the MEMS layer of the device, a frame of the MEMS layer of the device, or any other suitable portion of the MEMS device that is fixed relative to the movable proof masses. The proof masses may be arranged in a manner such that they move in response to measured force. The movement of the proof masses relative to a fixed surface (e.g., a fixed sense electrode extending in to the MEMS layer or located parallel to the movable mass on the substrate) in response to the measured force is measured and scaled to determine the desired inertial parameter.
In the embodiment of
In an exemplary embodiment, the MEMS layer 210 may include at least one anchoring point 208 and at least one movable proof mass 201 that is attached to the anchoring point 208 and suspended above the substrate 220. The anchoring point 208 may be fixedly attached (e.g., bonded) to and extend from a planar surface of the substrate 220. The anchoring point 208 and the movable proof mass 201 may be composed of conductive material, and the movable proof mass 201 may be arranged to pivot about the anchoring point 208 such that one end of the proof mass 201 tilts up while the other end tilts down in response to a sensed inertial force. Thus, when one side of the proof mass surface moves away from the substrate 220 the other side of the proof mass surface on the opposite end moves toward the substrate 220. Although not depicted in
The proof mass 201 may define a plurality of planar surfaces, including an upper planar surface (top of proof mass 201, in the x/y plane) and a lower planar surface (bottom of proof mass 201, in the x/y plane). Although in different embodiments a proof mass may have a plurality of different shapes within the MEMS device plane, in the exemplary embodiment of
The inertial sensor 200 may also comprise at least one sense electrode that, in conjunction with the proof mass 201, forms a capacitor. The exemplary embodiment of
Each sense electrode 203 and 204 faces an opposite portion of the lower planar surface of the proof mass 201 that is suspended above the substrate 220. Using these sense electrodes 203 and 204, the position of the proof mass 201 is capacitively sensed. In this regard, the value of the capacitance between sense electrode 203 and the proof mass 201 changes based upon the distance between the upper planar surface of sense electrode 203 and the lower planar surface of proof mass 201. The capacitance between sense electrode 204 and the proof mass 201 changes based upon the distance between the upper planar surface of sense electrode 204 and the lower planar surface of proof mass 201.
The capacitance formed by each capacitor may be sensed, and the capacitance signals may be processed (e.g., by filtering, amplification, scaling, etc.) to determine information about the sensed inertial force. In an exemplary embodiment, the memory 16 (
The symmetric guided mass system 400a rotates out-of-plane about a first roll-sense axis. The symmetric guided mass system 400b rotates out-of-plane about a second roll-sense axis in-plane and parallel to the first roll-sense axis. In an embodiment, pitch proof-masses 450a and 450b are each flexibly connected to their respective four roll proof-masses 402a-402d via springs. The springs are torsionally compliant such that pitch proof-mass 450a can rotate out-of-plane about a first pitch sense axis in the y-direction relative to sense electrodes 460a and 460b, and such that pitch proof-mass 450b can rotate out-of-plane about a second pitch sense axis in the y-direction relative to sense electrodes 460c-460d.
Angular velocity about the pitch-input axis in the x-direction will cause Coriolis forces to act on the pitch proof-masses 450a and 450b about the first and second pitch-sense axes respectively. The Coriolis forces cause the pitch proof masses 450a and 450b to rotate anti-phase out-of-plane about the first and the second pitch-sense axes. The amplitudes of the rotations of the pitch proof-masses 450a and 450b about the first and the second pitch-sense axes are proportional to the angular velocity about the pitch-input axis.
In an embodiment, sense electrodes 460a-460d located on the substrate and under the pitch proof masses 450a and 450b are used to detect the anti-phase rotations about the first and the second pitch-sense axes. An electrode shield 414 may also be formed on the substrate (e.g., surrounding the sense electrodes), and in some embodiments may be of a same or similar material as the sense electrode. Externally applied angular acceleration about the roll-input axis will generate inertial torques in-phase on the pitch proof masses 450a and 450b causing them to rotate in-phase about the first and the second pitch-sense axes. To the extent that a gyroscope offset is imparted on the proof masses, the sense electrodes may sense an acceleration that is proportionally incorrect based on the relative size of the offset vis-à-vis the movement in response to the sensed inertial force.
Initial voltage 508 corresponds to an initial shield voltage, which may correspond to an arbitrary value or may be a selected value (e.g., a standard initial value provided during manufacturing or an updated value applied during sensor operation). Offset curve 502 represents an offset that is experienced by the proof mass in response to certain shield voltages. In exemplary embodiments, an offset curve or a portion thereof may be established by applying a number of shield voltages and determining offset responses to those applied shield voltages. A variety of search techniques may be applied, for example, based on known characteristics of an offset curve. By testing shield voltages that result in an increase or decrease in offset value, a change in slope (i.e., derivative) of offset values, or other suitable measurements, an offset curve 502 may be at least partially interpolated.
In some embodiments it may be possible to determine the offset curve 502 without modifying the shield voltage (e.g., for a sensor that does not have a variable shield voltage) or to use other information to assist in generating the offset curve 502 (e.g., with a shield voltage having limited resolution. Other voltages such as the proof mass voltage may be modified (e.g., to change the voltage difference between the shield voltage and the proof mass) or forces may be applied to the proof mass (e.g., to determine the response to particular forces, which may be based at least in part on the capacitance formed between the proof mass and the shield.
By establishing the offset curve 502, it may be possible to determine a mechanical sensor offset 504 and a voltage-induced sensor offset 506. A mechanical sensor offset 504 may be an offset that is not attributable to the component under analysis (e.g., the shield). Other components may create independent voltage-induced sensor offset s of their own (e.g., additional voltage-induced sensor offsets), and an offset error may be the result of manufacturing tolerances or changes in sensor function over time (e.g., mechanical sensor offsets). In some embodiments, it may be possible to optimize the voltage of multiple components, thereby reducing at least the non-mechanical portion of any mechanical sensor offset.
As is depicted in
In the exemplary embodiment of
Offset curve 608 depicts an exemplary shifted offset curve 608. Shifts in the offset curve may result in changes to the mechanical sensor offset, changes in the shape of the offset curve, and changes in the voltage at which the minimum offset voltage of the offset curve occurs. Offset curve 608 may have experienced an increase in the mechanical sensor offset (e.g., in addition to any compensation originally performed for offset curve 606), as is depicted by mechanical sensor offset 610. The minimum offset of the offset curve 608 has also shifted from the minimum offset of offset curve 606, such that the minimum offset of offset curve 608 occurs at a higher shield voltage than operational shield voltage 604. If the shield voltage of the exemplary sensor of
At block 702, sensing may be performed for the sensor (e.g., an inertial sensor such as a MEMS gyroscope or MEMS accelerometer) which may result in an output signal (e.g., a signal corresponding to an output from differential sense electrodes of the inertial sensor) that may be processed to determine a signal that is related to (e.g., is proportional to) the motion being sensed (e.g., linear acceleration or angular velocity). This processed output may be provided to the summer block 704.
At block 706, the voltage-induced sensor offset may be determined as described herein. In an exemplary embodiment, a set of shield voltage values may have been tested prior to the sensing of block 702 to generate an offset curve or related values. Based on this information and the operational shield voltage used at block 702, a voltage-induced sensor offset may be calculated and output from block 706. Block 708 may provide scaling for the determined voltage-induced sensor offset so that a value output from block 708 is in the same units and scaling as the output from block 702. The output of block 708 may be provided to summer 704 as a subtraction input to be removed from the output of the measured value from block 702.
Block 710 may access a mechanical sensor offset. In some embodiments the mechanical sensor offset may be a fixed value. In other embodiments, the mechanical sensor offset may be updated, for example, based on the same offset curve used to determine the voltage-induced sensor offset at block 706. If the offset curve is determined during operation, it may be desirable to have a zero or known input of the measured characteristic (e.g., linear acceleration). This mechanical sensor offset may be scaled in the same manner as the outputs from blocks 702 and block 708, and provided to summer 704 to be subtracted from the measured output of block 702. The output of block 704 may therefore correspond to the raw measured output from block 702 at the operational shield voltage, corrected based on the voltage-induced sensor offset and the mechanical sensor offset. The output of summer 704 may then be used to accurately determine the desired sensor output (e.g., linear acceleration or angular velocity).
At block 802, sensing may be performed for the sensor (e.g., an inertial sensor such as a MEMS gyroscope or MEMS accelerometer) which may result in an output signal (e.g., a signal corresponding to an output from differential sense electrodes of the inertial sensor) that may be processed to determine a signal that is related to (e.g., is proportional to) the motion being sensed (e.g., linear acceleration or angular velocity). This processed output may be provided to the summer block 704.
At block 806, the voltage-induced sensor offset may be determined as described herein. In an exemplary embodiment, a set of shield voltage values may have been tested prior to the sensing of block 802 to generate an offset curve or related values. Based on this information a minimum offset voltage for the offset curve may be determined and output from block 806. Block 808 may modify the operational shield voltage of the sensor to correspond to the minimum offset voltage, which modifies the operation and sensing of block 802. In this manner, the minimum offset voltage is repeatedly determined, the shield voltage is repeatedly updated to the minimum offset voltage, and the operation value of block 802 is repeatedly modified to eliminate the voltage-induced sensor offset from the signal that is output from block 802.
Block 810 may access a mechanical sensor offset. In some embodiments the mechanical sensor offset may be a fixed value. In other embodiments, the mechanical sensor offset may be updated, for example, based on the same offset curve used to determine the voltage-induced sensor offset at block 806. This mechanical sensor offset may be scaled in the same manner as the outputs from block 802, and provided to summer 704 to be subtracted from the measured output of block 802. The output of block 804 may therefore correspond to the measured output from block 802 with the shield voltage set to the minimum offset voltage, corrected based on the mechanical sensor offset. The output of summer 804 may then be used to accurately determine the desired sensor output (e.g., linear acceleration or angular velocity).
At step 904, the voltage of the shield may be modified to match the voltage that corresponds to the minimum offset. In addition, a mechanical sensor offset associated with other factors (e.g., a mechanical offset or offset due to other devices) may be identified. Processing may then continue to step 906, at which the modified shield voltage and other values such as residual voltage may be stored.
At step 1004, it may be determined whether correction of any sensor offset will be performed using closed loop methodology (e.g., modifying the shield voltage to reduce the offset) or an open loop methodology (e.g., compensating for the voltage-induced sensor offset by modifying the operation of circuitry and/or scaling factors). If closed loop correction is to be performed, processing may continue to step 1006 at which the shield voltage may be set to the voltage that is associated with the minimum sensor offset. Processing may then continue from step 1006 to step 1010. If open loop correction is to be performed, processing may continue to step 1008 at which the voltage-induced sensor offset is determined and compensation is performed in the measurement circuitry and/or scaling to factor in the known offset. Processing may then continue from step 1008 to step 1010.
At step 1010, the mechanical sensor offset may be determined based on the offset that remains in the sensor even at the minimum offset voltage. If the offset curve is determined during operation, it may be desirable to have a zero or known input of the measured characteristic (e.g., linear acceleration). Compensation may then be performed in the measurement circuitry and/or scaling to remove this mechanical sensor offset from the determination of the measured values at step 1012. Once correction and compensation have been performed for both of the voltage-induced sensor offset and the mechanical sensor offset, the processing of
The foregoing description includes exemplary embodiments in accordance with the present disclosure. These examples are provided for purposes of illustration only, and not for purposes of limitation. It will be understood that the present disclosure may be implemented in forms different from those explicitly described and depicted herein and that various modifications, optimizations, and variations may be implemented by a person of ordinary skill in the present art, consistent with the following claims.