Signal processing method, signal processing device, physical quantity measuring device, and sensor module

Abstract

To provide a signal processing method that can detect a signal component with periodicity without performing envelope processing.SOLUTION: A signal processing method includes: a processing target signal generation step of generating a processing target signal being a time-series signal based on a source signal being a time-series signal output from an object; and a vibration rectification error calculation step of calculating a plurality of vibration rectification errors by performing, a plurality of times, product-sum operation processing of a first signal based on the processing target signal and a second signal based on a phase-shifted signal of the processing target signal while changing a shift amount.SELECTED DRAWING: Figure 19

Description

[Technical field]

[0001] The present invention relates to a signal processing method, a signal processing device, a physical quantity measuring device, and a sensor module. [Background technology]

[0002] Synchronous addition is known as a method for reducing components that are asynchronous with the target stationary repetitive waveform. However, this method has the problem that it also reduces waveform components that are correlated with the repetitive waveform but are not synchronized with the synchronous addition timing. As a method for addressing this problem, Non-Patent Document 1 proposes a method in which envelope processing is performed on the target stationary repetitive time series waveform, and the obtained waveform is subjected to spectrum analysis. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Pete Sopcik and Dara O'Sullivan, "How Sensor Performance Enables Condition-Based Monitoring Solutions", Analog Dialogue 53-06, June 2019. Summary of the Invention [Problem to be solved by the invention]

[0004] However, in envelope processing, it is necessary to perform smoothing processing after rectifying the time series waveform, and it is necessary to appropriately select the cutoff frequency of the smoothing filter so that the desired signal component is appropriately extracted, so that the method described in Non-Patent Document 1 involves complex calculations. [Means for solving the problem]

[0005] One aspect of the signal processing method according to the present invention includes: a process to generate a process target signal which is a time-series signal based on a source signal which is a time-series signal output from an object; The method includes a vibration rectification error calculation process for calculating a plurality of vibration rectification errors by performing a product-sum operation a plurality of times with changing the shift amount of a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed.

[0006] One aspect of the signal processing device according to the present invention is a processing target signal generating circuit that generates a processing target signal, which is a time-series signal, based on a source signal, which is a time-series signal output from an object; The vibration rectification error calculation circuit includes a vibration rectification error calculation circuit that performs a product-sum operation on a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed multiple times while changing the shift amount, thereby generating multiple vibration rectification errors.

[0007] One aspect of the physical quantity measuring device according to the present invention is a reference signal generating circuit that outputs a reference signal; a frequency delta-sigma modulation circuit that uses a signal under test to frequency delta-sigma modulate the reference signal to generate a frequency delta-sigma modulated signal; a first filter that operates in synchronization with the signal under test and has a variable group delay; a second filter that operates in synchronization with the reference signal; the first filter is provided on a signal path from an output of the frequency delta-sigma modulation circuit to an input of the second filter; a first operation mode for measuring a frequency ratio between the signal under test and the reference signal; and a second mode of operation in which the cutoff frequency of the filter is lower than that of the first mode of operation.

[0008] One aspect of the sensor module according to the present invention is One aspect of the physical quantity measuring device; A physical quantity sensor, The measured signal is a signal based on an output signal of the physical quantity sensor. [Brief description of the drawings]

[0009] [Figure 1] FIG. [Diagram 2] FIG. [Diagram 3] FIG. [Figure 4] FIG. [Diagram 5] 5 is a cross-sectional view taken along line P1-P1 in FIG. 4 . [Figure 6] FIG. 4 is a diagram illustrating the operation of a physical quantity sensor. [Figure 7] FIG. 4 is a diagram illustrating the operation of a physical quantity sensor. [Figure 8] Functional block diagram of the sensor module. [Figure 9] 4A and 4B are diagrams for explaining the principle of how vibration rectification error occurs due to output waveform distortion. [Figure 10] FIG. 13 is a graph showing the nonlinearity between the applied acceleration and the reciprocal count value. [Figure 11] 5 is a diagram showing nonlinearity between an applied acceleration and an oscillation frequency of a physical quantity sensor. [Figure 12] 5 is a diagram showing nonlinearity between the oscillation frequency and the reciprocal count value of a physical quantity sensor. [Figure 13] FIG. 4 is a diagram showing a configuration example of a frequency ratio measuring circuit. [Figure 14] FIG. 4 is a diagram showing an example of the configuration of a first low-pass filter. [Figure 15] FIG. 4 is a diagram showing an example of the configuration of a second low-pass filter. [Figure 16] 13 is a diagram for explaining that it is possible to adjust vibration rectification errors caused by nonlinearity of the input / output of a frequency ratio measurement circuit. FIG. [Figure 17] FIG. 13 is a graph showing the dependence of vibration rectification error contained in a measured value on the number of taps. [Figure 18] FIG. 1 is a diagram showing an example of the configuration of a signal processing device according to a first embodiment. [Figure 19]FIG. 4 is a flowchart showing the procedure of a signal processing method according to the first embodiment. [Figure 20] 20 is a flowchart showing an example of a procedure of a vibration rectification error calculation step S3 in FIG. 19. [Figure 21] FIG. 4 is a diagram showing a frequency spectrum obtained by performing FFT on a source signal. [Figure 22] FIG. 4 is a diagram showing a frequency spectrum obtained by performing FFT on a source signal. [Diagram 23] FIG. 4 is a diagram showing a frequency spectrum obtained by performing FFT on a source signal. [Figure 24] FIG. 4 is a diagram showing a frequency spectrum obtained by performing FFT on a source signal. [Diagram 25] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Figure 26] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Figure 27] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Figure 28] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Figure 29] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Diagram 30] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Diagram 31] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Diagram 32] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Diagram 33] FIG. 4 is a plot of vibration rectification error in the first embodiment. [Diagram 34] FIG. 13 is a plot of vibration rectification error in the second embodiment. [Diagram 35] FIG. 13 is a plot of vibration rectification error in the second embodiment. [Diagram 36] FIG. 13 is a plot of vibration rectification error in the second embodiment. [Figure 37] FIG. 13 is a plot of vibration rectification error in the second embodiment. [Figure 38]FIG. 13 is a plot of vibration rectification error in the second embodiment. [Figure 39] FIG. 13 is a plot of vibration rectification error in the second embodiment. [Diagram 40] FIG. 13 is a plot of vibration rectification error in the third embodiment. [Diagram 41] FIG. 13 is a diagram showing an example of the configuration of a signal processing device according to a fourth embodiment. [Diagram 42] FIG. 13 is a flowchart showing the procedure of a signal processing method according to a fourth embodiment. [Diagram 43] FIG. 13 is a plot of vibration rectification error in the fourth embodiment. [Diagram 44] FIG. 13 is a plot of vibration rectification error in the fourth embodiment. [Diagram 45] FIG. 13 is a diagram showing an example of the configuration of a signal processing device according to a fifth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Preferred embodiments of the present invention will be described in detail below with reference to the drawings. Note that the embodiments described below do not unduly limit the contents of the present invention described in the claims. Furthermore, not all of the configurations described below are necessarily essential components of the present invention.

[0011] In the following, the signal processing method according to the present invention will be described assuming that the target object to be processed is a sensor module. The type of the target object is not particularly limited as long as it generates a periodic signal, and may be, in addition to the sensor module, various devices such as a motor, structures such as a bridge or a building, or an electric circuit.

[0012] 1. First embodiment 1-1.Sensor module structure First, an example of the structure of a sensor module, which is an example of an object to which the signal processing method of the present embodiment is applied, will be described.

[0013] 1 is a perspective view of the sensor module 1 as viewed from the mounting surface side to which the sensor module 1 is fixed. In the following explanation, the direction along the long side of the rectangular sensor module 1 in a plan view is defined as the X-axis direction, the direction perpendicular to the X-axis direction in a plan view is defined as the Y-axis direction, and the thickness direction of the sensor module 1 is defined as the Z-axis direction.

[0014] The sensor module 1 has a rectangular parallelepiped shape in plan view, with long sides along the X-axis direction and short sides along the Y-axis direction perpendicular to the X-axis direction. Screw holes 103 are formed in two locations near the ends of one long side and in one location in the center of the other long side. Fixing screws are passed through each of the three screw holes 103, and the sensor module is used in a fixed state to the mounting surface of a mounting object, such as a building, a bulletin board, or various types of equipment.

[0015] 1, an opening 121 is provided on the surface of the sensor module 1 as viewed from the side of the surface on which the sensor module 1 is to be attached. A plug-type connector 116 is disposed inside the opening 121. The connector 116 has a plurality of pins arranged in two rows, and in each row, the plurality of pins are aligned in the Y-axis direction. A socket-type connector (not shown) is connected to the body to which the sensor module 1 is to be attached, and electrical signals such as a drive voltage for the sensor module 1 and detection data are transmitted and received.

[0016] Fig. 2 is an exploded perspective view of the sensor module 1. As shown in Fig. 2, the sensor module 1 is composed of a container 101, a lid 102, a sealing member 141, a circuit board 115, etc. In detail, the sensor module 1 is configured such that the circuit board 115 is attached inside the container 101 with a fixing member 130 interposed therebetween, and the opening of the container 101 is covered with the lid 102 via the sealing member 141 having cushioning properties.

[0017] The container 101 is made of, for example, aluminum and is a container for housing the circuit board 115, which is formed in a box shape having an internal space. Similarly, the planar shape is a rectangular parallelepiped with a substantially rectangular shape, and fixing protrusions 104 are provided at two locations near both ends of one long side and at one location in the center of the other long side. Each of these fixing protrusions 104 has a screw hole 103 formed therein.

[0018] The container 101 has a box-like shape with a rectangular parallelepiped outer shape and an opening on one side. The inside of the container 101 is an internal space surrounded by a bottom wall 112 and a side wall 111. In other words, the container 101 is box-like with one side facing the bottom wall 112 as an opening surface 123, and the outer edge of the circuit board 115 is arranged along the inner surface 122 of the side wall 111, and the lid 102 is fixed to cover the opening. Fixing protrusions 104 are erected on the opening surface 123 at two locations near both ends of one long side of the container 101 and at one location in the center of the other long side. The upper surface of the fixing protrusion 104, i.e., the surface exposed in the -Z direction, protrudes from the upper surface of the container 101.

[0019] Furthermore, in the internal space of container 101, a protrusion 129 is provided at the center of one long side opposite fixing protrusion 104 provided at the center of the other long side, protruding from side wall 111 toward the internal space from bottom wall 112 to opening surface 123. A female screw 174 is provided on the upper surface of protrusion 129. Lid 102 is fixed to container 101 via seal member 141 by screw 172 inserted into through hole 176 and female screw 174. Protrusion 129 and fixing protrusion 104 are provided at positions facing constricted portions 133, 134 of circuit board 115 described later.

[0020] In the internal space of container 101, first pedestal 127 and second pedestal 125 are provided, which protrude from bottom wall 112 toward opening surface 123 in a stepped manner. First pedestal 127 is provided at a position facing an arrangement area of ​​plug-type connector 116 attached to circuit board 115. First pedestal 127 is provided with opening 121 shown in FIG. 1, into which plug-type connector 116 is inserted. First pedestal 127 functions as a pedestal for fixing circuit board 115 to container 101.

[0021] Second base 125 is located on the opposite side of first base 127 with respect to fixing protrusion 104 and protrusion 129 located in the center of the long side, and is provided in the vicinity of fixing protrusion 104 and protrusion 129. Second base 125 functions as a base for fixing circuit board 115 to container 101 on the opposite side of first base 127 with respect to fixing protrusion 104 and protrusion 129.

[0022] In the above description, the outer shape of the container 101 is a box shape with a substantially rectangular parallelepiped planar shape and no lid, but the planar shape of the outer shape of the container 101 is not limited to this, and the planar shape of the outer shape of the container 101 may be a square, a hexagon, an octagon, or the like. In addition, in the planar shape of the outer shape of the container 101, the corners of the vertices of a polygon may be chamfered, and further, any of the sides may be a planar shape formed by a curve. In addition, the planar shape of the inside of the container 101 is not limited to the above-mentioned shape, and may be another shape. Furthermore, the planar shapes of the outer shape and the inside of the container 101 may or may not be similar to each other.

[0023] The circuit board 115 is a multi-layer board in which a plurality of through holes and the like are formed, and for example, a glass epoxy board, a composite board, a ceramic board, or the like is used.

[0024] The circuit board 115 has a second surface 115r on the bottom wall 112 side, and a first surface 115f which is the front and back of the second surface 115r. The physical quantity measuring device 2, the three physical quantity sensors 200, and other electronic components (not shown) are mounted on the first surface 115f of the circuit board 115. In addition, a connector 116 is mounted on the second surface 115r of the circuit board 115. Although not shown and their description are omitted, the circuit board 115 may be provided with other wiring, terminal electrodes, and the like.

[0025] Circuit board 115 includes constricted portions 133, 134 where the outer edge of circuit board 115 is constricted, in a central portion in the X-axis direction along the long side of container 101 in a plan view. Constricted portions 133, 134 are provided on both sides of circuit board 115 in the Y-axis direction in a plan view, and are constricted from the outer edge of circuit board 115 toward the center. In addition, constricted portions 133, 134 are provided opposite protrusion 129 and fixed protrusion 104 of container 101.

[0026] The circuit board 115 is inserted into the internal space of the container 101 with the second surface 115r facing the first pedestal 127 and the second pedestal 125. The circuit board 115 is supported in the container 101 by the first pedestal 127 and the second pedestal 125.

[0027] Each of the three physical quantity sensors 200 is a frequency change type sensor in which the frequency of the output signal changes according to the applied physical quantity. Of the three physical quantity sensors 200, the physical quantity sensor 200X detects a physical quantity in the X-axis direction, the physical quantity sensor 200Y detects a physical quantity in the Y-axis direction, and the physical quantity sensor 200Z detects a physical quantity in the Z-axis direction. Specifically, the physical quantity sensor 200X is erected so that the front and back surfaces of the package face the X-axis direction and its side faces the first surface 115f of the circuit board 115. The physical quantity sensor 200X outputs a signal according to the detected physical quantity in the X-axis direction. The physical quantity sensor 200Y is erected so that the front and back surfaces of the package face the Y-axis direction and its side faces the first surface 115f of the circuit board 115. The physical quantity sensor 200Y outputs a signal according to the detected physical quantity in the Y-axis direction. The physical quantity sensor 200Z is provided so that the front and back surfaces of the package face the Z-axis direction, that is, so that the front and back surfaces of the package face the first surface 115f of the circuit board 115. The physical quantity sensor 200Z outputs a signal according to the detected physical quantity in the Z-axis direction.

[0028] The physical quantity measuring device 2 is electrically connected to the physical quantity sensors 200X, 200Y, and 200Z via wiring and electronic components (not shown). In addition, the physical quantity measuring device 2 generates physical quantity data in which vibration rectification errors have been reduced, based on output signals from the physical quantity sensors 200X, 200Y, and 200Z.

[0029] 1-2. Structure of physical quantity sensor Next, an example of the structure of the physical quantity sensor 200 will be described, taking as an example a case where the physical quantity sensor 200 is an acceleration sensor. The three physical quantity sensors 200 shown in Fig. 2, that is, the physical quantity sensors 200X, 200Y, and 200Z, may have the same structure.

[0030] FIG. 3 is a perspective view of the physical quantity sensor 200, FIG. 4 is a plan view of the physical quantity sensor 200, and FIG. 5 is a cross-sectional view taken along line P1-P1 in FIG. 4. Note that FIGS. 3 to 5 only show the inside of the package of the physical quantity sensor 200. For convenience of explanation, each of the following figures shows an x-axis, a y-axis, and a z-axis as three mutually orthogonal axes. In the following explanation, for convenience of explanation, a plan view as seen from the z-axis direction, which is the thickness direction of the extensions 38a and 38b, is also simply referred to as a "plan view."

[0031] The physical quantity sensor 200 has a substrate portion 5 and four weights 50, 52, 54, and 56, as shown in FIGS.

[0032] The substrate unit 5 includes a plate-shaped base 10 having main surfaces 10a, 10b extending in the x-axis direction and facing opposite to each other, a joint portion 12 extending from the base 10 in the y-axis direction, a movable portion 13 extending in a rectangular shape from the joint portion 12 in the opposite direction to the base 10, two support portions 30a, 30b extending from both ends of the base 10 in the x-axis direction along the outer edge of the movable portion 13, and a physical quantity detection element 4 extending from the base 10 to the movable portion 13 and joined to the base 10 and the movable portion 13. 0 and.

[0033] In the two support parts 30a, 30b, the support part 30a extends along the y-axis across a gap 32a from the movable part 13 and is provided with a joint part 36a that fixes the support part 30a and an extension part 38a that extends along the x-axis across a gap 32c from the movable part 13. In other words, the support part 30a extends along the y-axis across a gap 32a from the movable part 13 and is provided with the extension part 38a that extends along the x-axis across a gap 32c from the movable part 13, and the joint part 36a is provided at the part of the support part 30a where the extension part 38a extends. Furthermore, the support portion 30b extends along the y-axis across a gap 32b from the movable portion 13, and is provided with a joint portion 36b that fixes the support portion 30b, and an extension portion 38b that extends along the x-axis across a gap 32c from the movable portion 13. In other words, the support portion 30b extends along the y-axis across a gap 32b from the movable portion 13, and is provided with an extension portion 38b that extends along the x-axis across a gap 32c from the movable portion 13, and the joint portion 36b is provided at the portion of the support portion 30b where the extension portion 38b extends.

[0034] The joints 36a, 36b provided on the supports 30a, 30b are for mounting the substrate 5 of the physical quantity sensor 200 on an external member such as a package. The base 10, the joint 12, the movable part 13, the supports 30a, 30b, and the extensions 38a, 38b may be integrally formed.

[0035] Movable part 13 is surrounded by support parts 30a, 30b and base part 10, and is connected to base part 10 via joint part 12 and supported in a cantilevered state. Movable part 13 has main surfaces 13a, 13b facing opposite each other, a side surface 13c along support part 30a, and a side surface 13d along support part 30b. Main surface 13a faces the same side as main surface 10a of base part 10, and main surface 13b faces the same side as main surface 10b of base part 10.

[0036] The joint part 12 is provided between the base part 10 and the movable part 13, and connects the base part 10 and the movable part 13. The thickness of the joint part 12 is formed to be thinner than the thicknesses of the base part 10 and the movable part 13. The joint part 12 has grooves 12a and 12b. These grooves 12a and 12b are formed along the X-axis, and when the movable part 13 is displaced relative to the base part 10, the grooves 12a and 12b of the joint part 12 function as fulcrums, i.e., intermediate hinges. Such joint part 12 and movable part 13 function as a cantilever.

[0037] Further, a physical amount detection element 40 is fixed to a surface extending from the principal surface 10a of the base 10 to the principal surface 13a of the movable part 13 by a bonding agent 60. The physical amount detection element 40 is fixed at two positions, namely, the central positions in the x-axis direction of each of the principal surfaces 10a and 13a.

[0038] The physical quantity detection element 40 has a base part 42a fixed to the main surface 10a of the base 10 with a bonding agent 60, a base part 42b fixed to the main surface 13a of the movable part 13 with a bonding agent 60, and vibrating beams 41a, 41b located between the base parts 42a and 42b for detecting a physical quantity. In this case, the vibrating beams 41a, 41b are shaped like rectangular columns, and when an AC voltage drive signal is applied to excitation electrodes (not shown) provided on the vibrating beams 41a, 41b, they vibrate in a bending manner along the x-axis so as to move away from or close to each other. In other words, the physical quantity detection element 40 is a tuning fork-type vibrator element.

[0039] Lead electrodes 44a and 44b are provided on the base portion 42a of the physical quantity detection element 40. These lead electrodes 44a and 44b are electrically connected to excitation electrodes (not shown) provided on the vibrating beams 41a and 41b. The lead electrodes 44a and 44b are electrically connected to connection terminals 46a and 46b provided on the main surface 10a of the base portion 10 by metal wires 48. The connection terminals 46a and 46b are connected to an external connection terminal 49 by wiring (not shown). a, 49b. The external connection terminals 49a, 49b are provided on the main surface 10b side of the base 10, which is the surface on which the physical quantity sensor 200 is mounted in a package or the like, so as to overlap with the package joint 34 in a plan view. The package joint 34 is for mounting the substrate 5 of the physical quantity sensor 200 on an external member such as a package, and is provided in two places at the ends on both ends in the x-axis direction of the base 10.

[0040] The physical quantity detection element 40 is formed by patterning a quartz substrate cut at a predetermined angle from a raw quartz stone or the like, using photolithography and etching techniques. In this case, it is desirable that the physical quantity detection element 40 is made of the same material as the base 10 and the movable part 13, in order to reduce the difference in linear expansion coefficient between the physical quantity detection element 40 and the base 10 and the movable part 13.

[0041] Weights 50, 52, 54, and 56 are rectangular in plan view and are provided on movable part 13. Weights 50 and 52 are fixed to main surface 13a of movable part 13 with bonding members 62, and weights 54 and 56 are fixed to main surface 13b of movable part 13 with bonding members 62. Here, weight 50 fixed to main surface 13a has one side, which is an edge of the rectangle, aligned with side surface 13c of movable part 13 in plan view, and the other side aligned with side surface 31d of extension part 38a. By aligning the directions in this way, weight 50 is disposed on the side of side surface 13c of movable part 13, and is disposed so that weight 50 and extension part 38a overlap in plan view. Similarly, weight 52 fixed to main surface 13a is arranged such that one side of the rectangle is oriented in the same direction as side surface 13d of movable part 13, and the other side is oriented in the same direction as side surface 31e of extension part 38b, so that weight 52 is arranged on the side of side surface 13d of movable part 13 and weight 52 and extension part 38b overlap in plan view. Weight 54 fixed to main surface 13b is arranged such that one side of the rectangle is oriented in the same direction as side surface 13c of movable part 13, and the other side is oriented in the same direction as side surface 31d of extension part 38a, so that weight 54 is arranged on the side of side surface 13c of movable part 13 and weight 54 and extension part 38a overlap in plan view. Similarly, weight 56 ​​fixed to main surface 13b is arranged so that, in a planar view, one side of the rectangle is oriented in the same direction as side surface 13d of movable part 13, and the other side is oriented in the same direction as side surface 31e of extension part 38b, such that weight 56 ​​is positioned on the side of side surface 13d of movable part 13 and weight 56 ​​and extension part 38b overlap in a planar view.

[0042] The weights 50, 52, 54, and 56 thus arranged are arranged symmetrically with respect to the physical quantity detection element 40, and the weights 54, 56 are arranged so as to overlap the weights 50, 52, respectively, in a plan view. These weights 50, 52, 54, and 56 are fixed to the movable part 13 by joint members 62 provided at the center of gravity positions of the weights 50, 52, 54, and 56, respectively. Furthermore, since the weights 50, 54 and the extension part 38a and the weights 52, 56 and the extension part 38b overlap each other in a plan view, when an excessive physical quantity is applied, the weights 50, 52, 54, and 56 come into contact with the extension parts 38a and 38b, and the amount of displacement of the weights 50, 52, 54, and 56 can be suppressed.

[0043] The joining member 62 is made of a silicone resin-based thermosetting adhesive or the like. It is applied to two locations on each of the main surface 13a and the main surface 13b of the movable part 13, and after the weights 50, 52, 54, and 56 are placed thereon, it is cured by heating to fix the weights 50, 52, 54, and 56 to the movable part 13. The joining surfaces of the weights 50, 52, 54, and 56 facing the main surface 13a and the main surface 13b of the movable part 13 are rough surfaces. As a result, when the weights 50, 52, 54, and 56 are fixed to the movable part 13, the joining area of ​​the joining surfaces is increased, and the joining strength can be improved.

[0044] When acceleration in the +Z direction represented by the arrow α1 is applied to the physical quantity sensor 200 configured as above, as shown in FIG. 6, a force acts on the movable part 13 in the -Z direction, and the movable The portion 13 is displaced in the -Z direction with the joint portion 12 as a fulcrum. As a result, a force is applied to the physical quantity detection element 40 in a direction in which the base portion 42a and the base portion 42b move away from each other along the Y axis, and tensile stress is generated in the vibrating beams 41a and 41b. As a result, the frequency at which the vibrating beams 41a and 41b vibrate becomes higher.

[0045] 7, when acceleration in the -Z direction indicated by the arrow α2 is applied to the physical quantity sensor 200, a force acts on the movable part 13 in the +Z direction, and the movable part 13 is displaced in the +Z direction with the joint part 12 as a fulcrum. As a result, a force is applied to the physical quantity detection element 40 in a direction in which the base part 42a and the base part 42b approach each other along the Y axis, and compressive stress is generated in the vibrating beams 41a and 41b. Therefore, the frequency at which the vibrating beams 41a and 41b vibrate decreases.

[0046] When the frequency at which the vibrating beams 41a, 41b vibrate changes in response to the acceleration, the frequency of the signal output from the external connection terminals 49a, 49b of the physical quantity sensor 200 changes. The sensor module 1 can calculate the value of the acceleration applied to the physical quantity sensor 200 based on the change in frequency of the output signal from the physical quantity sensor 200.

[0047] In order to improve the accuracy of detecting acceleration, which is a physical quantity, it is desirable that joint 12 connecting base 10, which is the fixed part, and movable part 13 be made of quartz, which is a material with a high Q value. For example, base 10, supports 30a and 30b, and movable part 13 may be formed from a quartz plate, and grooves 12a and 12b of joint 12 may be formed by half-etching from both sides of the quartz plate.

[0048] 1-3. Functional configuration of the sensor module 8 is a functional block diagram of the sensor module 1. As described above, the sensor module 1 includes the physical quantity sensors 200X, 200Y, and 200Z, and the physical quantity measuring device 2.

[0049] The physical quantity measuring device 2 includes oscillator circuits 201X, 201Y, and 201Z, frequency ratio measuring circuits 202X, 202Y, and 202Z, a microcontrol unit 210, a storage unit 220, and an interface circuit 230.

[0050] The oscillator circuit 201X amplifies the output signal of the physical quantity sensor 200X to generate a drive signal, and applies the drive signal to the physical quantity sensor 200X. The drive signal causes the vibrating beams 41a and 41b of the physical quantity sensor 200X to vibrate at a frequency corresponding to the acceleration in the X-axis direction, and a signal of the frequency is output from the physical quantity sensor 200X. The oscillator circuit 201X also outputs a measured signal SIN_X, which is a square wave signal obtained by amplifying the output signal of the physical quantity sensor 200X, to the frequency ratio measurement circuit 202X. The measured signal SIN_X is a signal based on the output signal of the physical quantity sensor 200X.

[0051] Similarly, the oscillator circuit 201Y amplifies the output signal of the physical quantity sensor 200Y to generate a drive signal, and applies the drive signal to the physical quantity sensor 200Y. The drive signal causes the vibrating beams 41a and 41b of the physical quantity sensor 200Y to vibrate at a frequency corresponding to the acceleration in the Y-axis direction, and a signal of the frequency is output from the physical quantity sensor 200Y. In addition, the oscillator circuit 201Y outputs a measured signal SIN_Y, which is a square wave signal obtained by amplifying the output signal of the physical quantity sensor 200Y, to the frequency ratio measurement circuit 202Y. The measured signal SIN_Y is a signal based on the output signal of the physical quantity sensor 200Y.

[0052] Similarly, the oscillator circuit 201Z amplifies the output signal of the physical quantity sensor 200Z to generate a drive signal, and applies the drive signal to the physical quantity sensor 200Z. The vibrating beams 41a and 41b of the physical quantity sensor 200Z vibrate at a frequency according to the acceleration in the Z-axis direction, and a signal of the frequency is output from the physical quantity sensor 200Z. The oscillator circuit 201Z outputs a measured signal SIN_Z, which is a square wave signal obtained by amplifying the output signal of the physical quantity sensor 200Z, to the frequency ratio measuring circuit 202Z. The measured signal SIN_Z is a signal based on the output signal of the physical quantity sensor 200Z.

[0053] The reference signal generating circuit 203 generates and outputs a reference signal CLK having a constant frequency. In this embodiment, the frequency of the reference signal CLK is higher than the frequencies of the signals under measurement SIN_X, SIN_Y, and SIN_Z. It is preferable that the reference signal CLK has high frequency accuracy, and the reference signal generating circuit 203 may be, for example, a temperature compensated crystal oscillator.

[0054] The frequency ratio measurement circuit 202X counts the number of pulses of the reference signal CLK included in a predetermined period of the signal under measurement SIN_X, which is a signal based on the signal output from the oscillation circuit 201X, and outputs a count value CNT_X. The count value CNT_X is a reciprocal count value corresponding to the frequency ratio between the signal under measurement SIN_X and the reference signal CLK.

[0055] The frequency ratio measuring circuit 202Y counts the number of pulses of the reference signal CLK included in a predetermined period of the signal under measurement SIN_Y output from the oscillation circuit 201Y, and outputs a count value CNT_Y. The count value CNT_Y is a reciprocal count value corresponding to the frequency ratio between the signal under measurement SIN_Y and the reference signal CLK.

[0056] The frequency ratio measuring circuit 202Z counts the number of pulses of the reference signal CLK included in a predetermined period of the signal under measurement SIN_Z output from the oscillation circuit 201Z, and outputs a count value CNT_Z. The count value CNT_Z is a reciprocal count value corresponding to the frequency ratio between the signal under measurement SIN_Z and the reference signal CLK.

[0057] The storage unit 220 stores programs and data, and may include volatile memory such as SRAM, DRAM, etc. SRAM is an abbreviation for Static Random Access Memory, and DRAM is an abbreviation for Dynamic Random Access Memory.

[0058] The storage unit 220 may also include non-volatile memory such as semiconductor memory such as EEPROM or flash memory, magnetic storage device such as hard disk device, optical storage device such as optical disk device, etc. EEPROM is an abbreviation for Electrically Erasable Programmable Read Only Memory.

[0059] The micro control unit 210 operates in synchronization with the reference signal CLK, and performs a predetermined calculation process and control process by executing a program (not shown) stored in the storage unit 220. For example, the micro control unit 210 measures the physical quantities detected by the physical quantity sensors 200X, 200Y, and 200Z based on the count value CNT_X output from the frequency ratio measurement circuit 202X, the count value CNT_Y output from the frequency ratio measurement circuit 202Y, and the count value CNT_Z output from the frequency ratio measurement circuit 202Z. Specifically, the micro control unit 210 converts the count value CNT_X, the count value CNT_Y, and the count value CNT_Z into a measurement value of the physical quantity in the X-axis direction, a measurement value of the physical quantity in the Y-axis direction, and a measurement value of the physical quantity in the Z-axis direction, respectively. For example, the storage unit 220 may store table information that specifies the correspondence between the count value and the measurement value of the physical quantity, or information on a relational expression between the count value and the measurement value of the physical quantity, and the micro control unit 210 may convert each count value into a measurement value of the physical quantity by referring to the information.

[0060] The microcontrol unit 210 receives the measured values ​​of the physical quantity in the X-axis direction, the measured values ​​of the physical quantity in the Y-axis direction, and the measured values ​​of the physical quantity in the Z-axis direction via an interface circuit 230. Alternatively, the micro control unit 210 may write the measured values ​​of the physical quantity in the X-axis direction, the measured values ​​of the physical quantity in the Y-axis direction, and the measured values ​​of the physical quantity in the Z-axis direction into the storage unit 220, and the signal processing device 400 may read out each measured value via the interface circuit 230.

[0061] Since the frequency ratio measurement circuits 202X, 202Y, and 202Z have the same configuration and operation, hereinafter, any one of the frequency ratio measurement circuits 202X, 202Y, and 202Z will be referred to as the frequency ratio measurement circuit 202. Also, any one of the signals under measurement SIN_X, SIN_Y, and SIN_Z input to the frequency ratio measurement circuit 202 will be referred to as the signal under measurement SIN, and any one of the count values ​​CNT_X, CNT_Y, and CNT_Z output from the frequency ratio measurement circuit 202 will be referred to as the count value CNT.

[0062] 1-4. Vibration rectification error The vibration rectification error corresponds to a DC offset that occurs during rectification due to the nonlinearity of the response of the sensor module 1 to vibration, and is observed as an abnormal shift in the output offset of the sensor module 1. In applications where the DC output of the sensor module 1 is directly used as the measurement target, such as an inclinometer using the sensor module 1, this can cause serious measurement errors. There are three main mechanisms that cause vibration rectification errors: [1] asymmetric rails, [2] nonlinearity of the scale factor, and [3] structural resonance of the physical quantity sensor 200.

[0063] [1] Vibration rectification error caused by asymmetric rails When the sensitivity axis of the physical quantity sensor 200 is in the direction of gravitational acceleration, the measurement value of the sensor module 1 includes a gravitational acceleration of 1g=9.8m / s 2 For example, if the dynamic range of the physical quantity sensor 200 is 2 g, then vibrations of up to 1 g can be measured without clipping. If vibrations exceeding 1 g are applied in this state, clipping occurs asymmetrically, and the measured value will contain a vibration rectification error.

[0064] For example, when the dynamic range is as wide as 15 g, clipping rarely becomes a problem in normal usage environments. On the other hand, the physical quantity sensor 200 has a built-in physical protection mechanism for the purpose of preventing damage to the physical quantity detection element 40, and when the vibration level exceeds a certain threshold, the protection mechanism is activated, causing clipping. To prevent this, it is necessary to devise an attachment for installing the sensor module 1 and take measures such as damping vibrations in the resonant frequency band.

[0065] [2] Vibration rectification error caused by nonlinearity of the scale factor Fig. 9 is a diagram for explaining the principle of vibration rectification error caused by output waveform distortion. In Fig. 9, the solid line shows a sine wave vibration waveform and a waveform obtained by smoothing the vibration waveform, and the dashed line shows a vibration waveform that is asymmetric above and below the vibration center and a waveform obtained by smoothing the vibration waveform. The smoothed waveform shown by the solid line is 0, while the smoothed waveform shown by the dashed line is a negative value, and an offset occurs during smoothing.

[0066] The physical quantity sensor 200 is a frequency change type sensor, and the count value CNT corresponding to the frequency ratio between the measured signal SIN and the reference signal CLK is a reciprocal count value. The relationship between the acceleration applied to the physical quantity sensor 200 and the reciprocal count value is nonlinear. The dashed line in FIG. 10 indicates the nonlinearity between the applied acceleration and the reciprocal count value. The dashed line in FIG. 11 indicates the nonlinearity between the applied acceleration and the oscillation frequency of the physical quantity sensor 200. The dashed line in FIG. 12 indicates the nonlinearity between the oscillation frequency of the physical quantity sensor 200 and the reciprocal count value. The dashed line in FIG. 10 is obtained by combining the dashed line in FIG. 11 and the dashed line in FIG. 12.

[0067] Here, by correcting the relationship between the oscillation frequency and the reciprocal count value as shown by the solid line in Fig. 12, the relationship between the acceleration and the reciprocal count value can be made closer to linear as shown by the solid line in Fig. 10. Specifically, the above-mentioned micro control unit 210 can correct the count value CNT using the correction function expressed by equation (1).

[0068]

number

[0069] In formula (1), c is the count value before correction corresponding to the dashed line in Fig. 10, Y is the count value after correction corresponding to the solid line in Fig. 10, and d is a coefficient that determines the degree of correction shown in Fig. 12. For example, the coefficient d is stored in the storage unit 220 or is set by the signal processing device 400.

[0070] [3] Vibration rectification error caused by cantilever resonance The principle of acceleration detection in the physical quantity sensor 200 is to transmit the deflection of a weighted cantilever due to acceleration to the physical quantity detection element 40, which is a double-ended tuning fork vibrator, to change the tension acting on the physical quantity detection element 40, thereby changing the oscillation frequency. Therefore, the physical quantity detection element 40 has a resonance frequency due to the structure of the cantilever, and when the cantilever resonance is excited, an inherent vibration rectification error occurs. The cantilever resonance has a frequency higher than the frequency band corresponding to the range of detectable acceleration, and the vibration component is removed by the low-pass filter inside the physical quantity measuring device 2, but a vibration rectification error occurs as a bias offset reflecting the asymmetry of the vibration. As the amplitude of the cantilever resonance increases, the asymmetry of the output waveform of the physical quantity sensor 200 increases, and the vibration rectification error also increases. Therefore, it is an important issue to reduce the vibration rectification error caused by the cantilever resonance.

[0071] In this embodiment, the frequency ratio measurement circuit 202 is a reciprocal counting circuit that counts the number of pulses of the reference signal CLK included in a predetermined period of the signal SIN under test, and the timing of acquiring this count value is synchronized with the signal SIN under test. On the other hand, the count value CNT output from the frequency ratio measurement circuit 202 needs to be synchronized with a divided signal of the reference signal CLK, and resampling is required because the timing of acquiring the count value of the number of pulses of the reference signal CLK is not synchronized with the divided signal of the reference signal CLK. In the frequency ratio measurement circuit 202, by devising a configuration required for resampling, it is possible to generate a count value CNT in which the vibration rectification error caused by the cantilever resonance is corrected.

[0072] 1-5. Configuration of frequency ratio measurement circuit The frequency ratio measurement circuit 202 measures the frequency ratio between the signal under test SIN and the reference signal CLK by a reciprocal counting method. Fig. 13 is a diagram showing an example of the configuration of the frequency ratio measurement circuit 202. As shown in Fig. 13, the frequency ratio measurement circuit 202 includes a frequency delta-sigma modulation circuit 300, a first low-pass filter 310, a latch circuit 320, and a second low-pass filter 330.

[0073] The frequency delta-sigma modulation circuit 300 uses the signal under test SIN to frequency delta-sigma modulate the reference signal CLK to generate a frequency delta-sigma modulated signal. The frequency delta-sigma modulation circuit 300 includes a counter 301, a latch circuit 302, a latch circuit 303, and a subtractor 304. The counter 301 counts the rising edges of the reference signal CLK and outputs a count value CT0. The latch circuit 302 latches and holds the count value CT0 in synchronization with the rising edges of the signal under test SIN. The latch circuit 303 outputs the counter value held by the latch circuit 302 in synchronization with the rising edges of the signal under test SIN. The subtractor 304 subtracts the count value held by the latch circuit 303 from the count value held by the latch circuit 302 to generate and output a count value CT1. This count value CT1 is the frequency delta-sigma modulated signal generated by the frequency delta-sigma modulation circuit 300.

[0074] This frequency delta-sigma modulation circuit 300 is also called a first-order frequency delta-sigma modulator, and latches the count value of the number of pulses of the reference signal CLK twice by the signal under test SIN, and sequentially holds the count value of the number of pulses of the reference signal CLK using the rising edge of the signal under test SIN as a trigger. Here, the frequency delta-sigma modulation circuit 300 has been described as performing the latch operation at the rising edge of the signal under test SIN, but the latch operation may be performed at the falling edge, or at both the rising edge and the falling edge. In addition, the subtractor 304 calculates the difference between the two count values ​​held in the latch circuits 302 and 303, and outputs the increment of the count value of the number of pulses of the reference signal CLK observed during one period of the signal under test SIN over time without any dead period. When the frequency of the signal under test SIN is fx and the frequency of the reference signal CLK is fc, the frequency ratio is fc / fx. The frequency delta-sigma modulation circuit 300 outputs a frequency delta-sigma modulated signal indicating the frequency ratio as a digital signal sequence.

[0075] The first low-pass filter 310 operates in synchronization with the signal under test SIN, and outputs a count value CT2 in which noise components contained in the count value CT1, which is the frequency delta-sigma modulated signal output from the frequency delta-sigma modulation circuit 300, have been removed or reduced. In Fig. 13, the first low-pass filter 310 is provided immediately after the frequency delta-sigma modulation circuit 300, but it is sufficient that the first low-pass filter 310 is provided on the signal path from the output of the frequency delta-sigma modulation circuit 300 to the input of the second low-pass filter 330.

[0076] The latch circuit 320 latches the count value CT2 output from the first low-pass filter 310 in synchronization with the rising edge of the reference signal CLK, and holds it as a count value CT3.

[0077] The second low-pass filter 330 operates in synchronization with the reference signal CLK, and outputs a count value from which noise components contained in the count value CT3 held by the latch circuit 320 have been removed or reduced. The count value output from the second low-pass filter 330 is output to the micro control unit 210 as the count value CNT.

[0078] Fig. 14 is a diagram showing an example of the configuration of first low-pass filter 310. In the example of Fig. 14, first low-pass filter 310 has delay element 311, integrator 312, integrator 313, decimator 314, delay element 315, differentiator 316, delay element 317, and differentiator 318. Each section of first low-pass filter 310 operates in synchronization with signal under measurement SIN.

[0079] The delay element 311 outputs a count value obtained by delaying the count value CT1 in synchronization with the signal under test SIN. The number of taps of the delay element 311 is na. For example, the delay element 311 is realized by a shift register in which na registers are connected in series.

[0080] The integrator 312 outputs a count value obtained by integrating the count values ​​output from the delay element 311 in synchronization with the signal under test SIN.

[0081] The integrator 313 outputs a count value obtained by integrating the count value output from the integrator 312 in synchronization with the signal under test SIN.

[0082] Decimator 314 outputs a count value obtained by decimating the count value output from integrator 313 to a rate of 1 / R in synchronization with signal under test SIN.

[0083] The delay element 315 outputs a count value obtained by delaying the count value output from the decimator 314 in synchronization with the signal under test SIN. The number of taps of the delay element 315 is n1. For example, the delay element 315 is realized by a shift register in which n1 registers are connected in series.

[0084] The differentiator 316 outputs a count value obtained by subtracting the count value output from the delay element 315 from the count value output from the decimator 314 .

[0085] The delay element 317 outputs a count value obtained by delaying the count value output from the differentiator 316 in synchronization with the signal under test SIN. The number of taps of the delay element 317 is n2. For example, the delay element 317 is realized by a shift register in which n2 registers are connected in series.

[0086] The differentiator 318 subtracts the count value output from the delay element 317 from the count value output from the differentiator 316 to output a count value CT2.

[0087] The tap numbers n1 and n2 and the decimation ratio R are fixed, and the tap number na is variable. For example, the tap number na is stored in the storage unit 220 or is set by the signal processing device 400.

[0088] The first low-pass filter 310 thus configured functions as a CIC filter with a group delay amount that is variable according to the number of taps na. CIC is an abbreviation for Cascaded Integrator Comb.

[0089] Fig. 15 is a diagram showing an example of the configuration of the second low-pass filter 330. In the example of Fig. 15, the second low-pass filter 330 has an integrator 331, a delay element 332, a differentiator 333, and a decimator 334. Each component of the second low-pass filter 330 operates in synchronization with the reference signal CLK.

[0090] The integrator 331 outputs a count value obtained by integrating the count value CT3 in synchronization with the reference signal CLK.

[0091] The delay element 332 outputs a count value obtained by delaying the count value output from the integrator 331 in synchronization with the reference signal CLK. The number of taps of the delay element 332 is n3. For example, the delay element 332 is realized by a shift register in which n3 registers are connected in series.

[0092] The differentiator 333 outputs a count value obtained by subtracting the count value output from the delay element 332 from the count value output from the integrator 331 .

[0093] The decimator 334 outputs a count value CNT obtained by decimating the count value output from the differentiator 333 at a rate of 1 / n3 in synchronization with the reference signal CLK.

[0094] The number of taps and the decimation ratio, n3, are fixed.

[0095] The second low-pass filter 330 thus configured accumulates the count value CT3 while resampling it with the reference signal CLK. It acts as a weighted moving average filter.

[0096] In this way, the first low-pass filter 310 operates in synchronization with the signal under test SIN, and the second low-pass filter 330 performs resampling in synchronization with the reference signal CLK, so that nonlinearity occurs in the input and output of the frequency ratio measurement circuit 202. Therefore, the count value CNT output from the frequency ratio measurement circuit 202 contains a vibration rectification error caused by this nonlinearity. Then, by adjusting the number of taps na of the delay element 311 of the first low-pass filter 310, this vibration rectification error can be adjusted.

[0097] FIG. 16 is a diagram for explaining that the vibration rectification error caused by the nonlinearity of the input and output of the frequency ratio measurement circuit 202 can be adjusted. FIG. 16 shows an example in which the period of the signal under test SIN is longer than that of the reference signal CLK, and the update period of the count value CNT is longer than that of the signal under test SIN, and the horizontal axis corresponds to the passage of time. In FIG. 16, the timing of the rising edge of the reference signal CLK is indicated by a short vertical line. Also, the timing of the change of the count values ​​CT1 and CT2 is indicated by a short vertical line. Note that FIG. 16 uses simplified numerical values ​​for the purpose of explaining the mechanism for adjusting the vibration rectification error to facilitate understanding. Also, although the count value CT2 is not determined until after the count value CT1 is determined, it is described as if the count value CT2 is determined before the count value CT1 is determined, but the actual calculation of the count value CT2 is performed after the count value CT1 is determined.

[0098] In Fig. 16, (A) is an example in which the period of the signal SIN under test is constant, and (B), (C), and (D) are examples in which the signal SIN under test is frequency modulated. In (B), (C), and (D), the group delay of the first low-pass filter 310 is different from each other. For simplicity, the period of the reference signal CLK and the period of the signal SIN under test are assumed to be a simple integer ratio, and the count value CT1 input to the first low-pass filter 310 is output as it is with a constant group delay. The second low-pass filter 330 integrates the count value CT2 output from the first low-pass filter 310 with the latched count value CT3 in synchronization with the reference signal CLK, and outputs the integrated value for 16 times as the count value CNT.

[0099] In the example of (A), the count value CT2 is always 4, and the count value CNT is 4×16=64. In the example of (B), the measured signal SIN is frequency modulated, and the group delay of the first low-pass filter 310 is set to 0, so the count value CT2 repeats 5, 5, 3, 3. Since weighting is performed by time during the accumulation, the count value CNT becomes 5×10+3×6=68, which is larger than the count value CNT of (A). In the example of (C), the count value CT2 repeats 5, 5, 3, 3, as in the example of (B), but a group delay occurs in the first low-pass filter 310. As a result of weighting by time during the accumulation, the count value CNT becomes 5×8+3×8=64, which is the same value as the count value CNT of (A). In the example of (D), the count value CT2 repeats 5, 5, 3, 3, as in the examples of (B) and (C), but compared to the example of (C), this example shows a case in which the group delay caused by the first low-pass filter 310 is large. In the example of (D), the count value CNT is 5×6+3×10=60, which is smaller than the count value CNT of (A).

[0100] 16, it can be qualitatively understood that the vibration rectification error caused by the nonlinearity of the input / output of the frequency ratio measurement circuit 202 varies depending on the group delay of the first low-pass filter 310. By adjusting the group delay of the first low-pass filter 310 so that the vibration rectification error caused by the nonlinearity of the input / output of the frequency ratio measurement circuit 202 is in antiphase with the vibration rectification error caused by the cantilever resonance, it becomes possible to cancel each other's vibration rectification errors. The group delay of the first low-pass filter 310 can be adjusted by setting the number of taps na of the delay element 311.

[0101] Fig. 17 is a diagram showing the dependency of the vibration rectification error included in the measurement value by the physical quantity measuring device 2 on the number of taps na. In Fig. 17, the horizontal axis is the number of taps na, and the vertical axis is the vibration rectification error. Note that VRE on the vertical axis is an abbreviation for Vibration Rectification Error. Fig. 17 shows that by appropriately setting the number of taps na, it is possible to correct the vibration rectification error and bring it closer to zero.

[0102] 1-6. Configuration of signal processing device In this embodiment, the signal processing device 400 performs processing to detect a signal component having periodicity contained in a signal output from the sensor module 1. Fig. 18 is a diagram showing an example of the configuration of the signal processing device 400. As shown in Fig. 18, the signal processing device 400 includes a processing circuit 410, a memory circuit 420, an operation unit 430, a display unit 440, a sound output unit 450, and a communication unit 460. Note that the signal processing device 400 may be configured such that some of the components in Fig. 18 are omitted or changed, or other components are added.

[0103] The processing circuit 410 acquires a source signal, which is a digital time-series signal output from the sensor module 1, and performs signal processing on the source signal. Specifically, the processing circuit 410 executes a signal processing program 421 stored in the storage circuit 420, and performs various calculation processing on the source signal. In addition, the processing circuit 410 performs various processes according to an operation signal from the operation unit 430, a process of transmitting a display signal for displaying various information on the display unit 440, a process of transmitting a sound signal for generating various sounds to the sound output unit 450, a process of controlling the communication unit 460 for performing data communication with other devices, and the like. The processing circuit 410 is realized by, for example, a CPU or a DSP. CPU is an abbreviation for Central Processing Unit, and DSP is an abbreviation for Digital Signal Processor.

[0104] The processing circuit 410 executes the signal processing program 421 to function as a source signal acquisition circuit 411, a processed signal generation circuit 412, and a vibration rectification error calculation circuit 413. That is, the signal processing device 400 includes the source signal acquisition circuit 411, the processed signal generation circuit 412, and the vibration rectification error calculation circuit 413.

[0105] The source signal acquisition circuit 411 acquires a source signal which is a time-series signal output from the sensor module 1. The source signal is a signal including a signal component having periodicity. For example, the source signal may be a signal including a signal component of the structural resonance frequency of the sensor module 1, specifically, the cantilever resonance frequency of the physical quantity sensor 200. For example, the source signal acquisition circuit 411 may acquire, as the source signal, time-series data of a count value CT1 which is a delta-sigma modulated signal input to the first low-pass filter 310 in the physical quantity measuring device 2.

[0106] The processed signal generating circuit 412 generates a processed signal, which is a time-series signal, based on the source signal acquired by the source signal acquiring circuit 411. For example, the processed signal generating circuit 412 may generate the processed signal by extracting a part of the time-series signal included in the source signal. Alternatively, the processed signal may be the source signal itself. The processed signal generated by the processed signal generating circuit 412 is stored in the storage circuit 420 as a processed signal 422.

[0107] The vibration rectification error calculation circuit 413 performs a multiplication and accumulation operation of a first signal based on the processing target signal and a second signal based on a signal obtained by shifting the phase of the processing target signal multiple times while changing the shift amount, thereby calculating multiple vibration rectification errors. The multiple vibration rectification errors calculated by the vibration rectification error calculation circuit 413 are stored in the storage circuit 420 as vibration rectification error information 423.

[0108] The first signal may be the signal to be processed itself. Alternatively, the first signal may be a signal obtained by filtering the signal to be processed. For example, the filtering may be performed using a smoothing filter. The first signal may be a signal obtained by removing or reducing a DC component from the signal to be processed. The first signal may be a signal obtained by removing or reducing a DC component from the signal to be processed and filtering the signal.

[0109] The second signal may be the signal itself obtained by shifting the phase of the signal to be processed. The second signal may also be a signal obtained by filtering the signal obtained by shifting the phase of the signal to be processed. For example, the filtering may be a smoothing filter process. The second signal may also be a signal obtained by removing or reducing a direct current component from a signal obtained by shifting the phase of the signal to be processed. The second signal may also be a signal obtained by removing or reducing a direct current component from a signal obtained by shifting the phase of the signal to be processed and filtering the signal.

[0110] If the i-th sample value of a signal to be processed having N samples is S(i), the i-th sample value of a signal obtained by shifting the phase of the signal to be processed is S(i+k). N is an integer of 2 or more, and i is an integer of 1 to N. For example, when the first signal is the signal to be processed itself, and the second signal is a signal obtained by shifting the phase of the signal to be processed itself, the k-th vibration rectification error VRE(k) of the M vibration rectification errors is calculated by Equation (2). M is an integer of 2 or more, and k is an integer of 1 to M. In Equation (2), S(i) is the i-th sample value of the first signal, and S(i+k) is the i-th sample value of the second signal.

[0111]

number

[0112] In addition, for example, when the first signal is a signal obtained by removing or reducing the DC component from the signal to be processed, and the second signal is a signal obtained by shifting the phase of the signal to be processed, the k-th vibration rectification error VRE(k) is calculated by the formula (3). HPF (S(i)) is the i-th sample value of the first signal and S(i+k) is the i-th sample value of the second signal.

[0113]

number

[0114] In addition, for example, when the first signal is a signal obtained by removing or reducing the DC component from the signal to be processed and performing smoothing filter processing, and the second signal is a signal obtained by shifting the phase of the signal to be processed, the k-th vibration rectification error VRE(k) is calculated by the formula (4). LPF (f HPF (S(i)) is the i-th sample value of the first signal and S(i+k) is the i-th sample value of the second signal.

[0115]

number

[0116] In addition, for example, when the first signal is a signal obtained by removing or reducing the DC component from the signal to be processed and performing smoothing filtering, and the second signal is a signal obtained by performing smoothing filtering on a signal obtained by shifting the phase of the signal to be processed, the k-th vibration rectification error VRE(k) is expressed by the formula: (5). In formula (5), f LPF (f HPF (S(i)) is the i-th sample value of the first signal, and f LPF (S(i+k)) is the i-th sample value of the second signal.

[0117]

number

[0118] In addition, in formulas (2), (3), (4), and (5), the division by N may be omitted.

[0119] By such a product-sum operation process, signal components having ergodicity such as noise are attenuated, and a signal having periodicity included in the signal to be processed appears as a vibration rectification error corresponding to the phase difference between the first signal and the second signal. Specifically, if the period of the signal having periodicity included in the signal to be processed is T, when the first signal and the second signal have a phase difference corresponding to an even multiple of T / 2, the vibration rectification error obtained by the product-sum operation process is maximized. When the first signal and the second signal have a phase difference corresponding to an odd multiple of T / 4, the vibration rectification error obtained by the product-sum operation process is 0. When the first signal and the second signal have a phase difference corresponding to an odd multiple of T / 2, the vibration rectification error obtained by the product-sum operation process is minimized. Therefore, if a plurality of vibration rectification errors calculated while changing the phase difference between the first signal and the second signal are plotted, the value of the vibration rectification error changes with the period T, so that a signal component having periodicity included in the signal to be processed can be detected.

[0120] In addition, when it is desired to detect the signal component of the structural resonance of the sensor module 1, specifically the signal component of the cantilever resonance, as a signal having periodicity, it is preferable that the number of additions N in the product-sum calculation process of the vibration rectification error calculation circuit 413 is greater than the value obtained by dividing the sampling frequency of the source signal by the resonance frequency. The sampling frequency of the source signal is, for example, the frequency of the measured signal SIN, which is the sampling signal of the count value CT1 input to the first low-pass filter 310. In this way, the signal component of the resonance frequency is integrated for one or more periods in the product-sum calculation process, and the signal component of the resonance frequency is effectively detected. In addition, the cantilever resonance is excited even in a general environment, but if the excitation level of the cantilever resonance changes, the calculated vibration rectification error also changes, so it is preferable that the signal processing device 400 performs the product-sum calculation process using the source signal acquired in a stable environment.

[0121] The memory circuit 420 has a ROM and a RAM, not shown. ROM stands for Read Only Memory, and RAM stands for Random Access Memory. The ROM stores various programs such as a signal processing program 421 and predetermined data, and the RAM stores signals generated by the processing circuit 410 and calculated information such as a processing target signal 422 and vibration rectification error information 423. The RAM is also used as a working area for the processing circuit 410, and stores programs and data read from the ROM, data input from the operation unit 430, and signals and data temporarily generated by the processing circuit 410.

[0122] The operation unit 430 is an input device including operation keys, button switches, and the like, and outputs to the processing circuit 410 an operation signal in response to an operation by a user.

[0123] The display unit 440 is a display device configured with an LCD or the like, and displays various information based on a display signal output from the processing circuit 410. LCD is an abbreviation for Liquid Crystal Display. The display unit 440 may be provided with a touch panel that functions as the operation unit 430. For example, the display unit 440 displays the following based on the display signal output from the processing circuit 410: An image in which the vibration rectification error information 423 is plotted may be displayed.

[0124] The sound output unit 450 is configured with a speaker or the like, and generates various sounds based on the sound signal output from the processing circuit 410. For example, the sound output unit 450 may generate sounds indicating the start or end of signal processing based on the sound signal output from the processing circuit 410.

[0125] The communication unit 460 performs various controls for establishing data communication between the processing circuit 410 and other devices. For example, the communication unit 460 may transmit the vibration rectification error information 423 to other devices.

[0126] At least a part of the source signal acquisition circuit 411, the processing target signal generation circuit 412, and the vibration rectification error calculation circuit 413 may be realized by dedicated hardware. The signal processing device 400 may be a single device or may be configured by multiple devices. For example, the processing circuit 410 and the storage circuit 420 may be realized by a device such as a cloud server, which calculates the vibration rectification error information 423 and transmits the calculated vibration rectification error information 423 to a terminal including the operation unit 430, the display unit 440, the sound output unit 450, and the communication unit 460 via a communication line.

[0127] 1-7.Signal processing method Fig. 19 is a flow chart showing the procedure of the signal processing method of the first embodiment. As shown in Fig. 19, the signal processing method of the first embodiment includes a source signal acquisition step S1, a processing target signal generation step S2, and a vibration rectification error calculation step S3. The signal processing method of the present embodiment is performed by, for example, a signal processing device 400.

[0128] First, in a source signal acquisition step S1, the signal processing device 400 acquires a source signal, which is a time-series signal output from the sensor module 1, which is an object.

[0129] Next, in a process target signal generating step S2, the signal processing device 400 generates a process target signal, which is a time-series signal, based on the source signal acquired in step S1.

[0130] Finally, in a vibration rectification error calculation process S3, the signal processing device 400 performs a product-sum operation multiple times on a first signal based on the signal to be processed generated in process S2 and a second signal based on a signal obtained by shifting the phase of the signal to be processed while changing the shift amount, to calculate multiple vibration rectification errors.

[0131] FIG. 20 is a flowchart showing an example of the procedure of the vibration rectification error calculation step S3 of FIG.

[0132] As shown in FIG. 20, first, in step S31, the signal processing device 400 generates a first signal based on the processing target signal generated in step S2.

[0133] Next, in step S32, the signal processing device 400 generates a second signal based on a signal obtained by shifting the phase of the processing target signal generated in step S2.

[0134] Next, in step S33, the signal processing device 400 performs a product-sum operation on the first signal generated in step S31 and the second signal generated in step S32 to calculate a vibration rectification error.

[0135] Next, in step S33, the signal processing device 400 determines whether or not the calculation of the required number of vibration rectification errors has been completed.

[0136] Then, if the calculation of the required number of vibration rectification errors has not been completed, the signal processing device 400 changes the phase shift amount in step S34, and repeats the processing from step S32 onwards until the calculation of the required number of vibration rectification errors has been completed.

[0137] 1-8. Specific examples of calculated vibration rectification error information Below is a specific example in which the signal processing device 400 acquires, as the source signal, the count value CT1 input to the first low-pass filter 310 in the physical quantity measuring device 2 of the sensor module 1, and plots a plurality of calculated vibration errors.

[0138] 21 to 24 are diagrams showing frequency spectra obtained by performing FFT on source signals acquired under four measurement conditions. The cantilever resonance frequency of the physical quantity sensor 200 is 850 Hz, and as shown in FIG. 21, the signal component due to cantilever resonance included in the source signal acquired under the first measurement condition has a large intensity. Also, as shown in FIG. 22, the signal component due to cantilever resonance included in the source signal acquired under the second measurement condition has a slightly smaller intensity than the signal component due to cantilever resonance included in the source signal acquired under the first measurement condition. Also, as shown in FIG. 23, the signal component due to cantilever resonance included in the source signal acquired under the third measurement condition is even smaller than the signal component due to cantilever resonance included in the source signal acquired under the second measurement condition. Also, as shown in FIG. 24, the signal component due to cantilever resonance included in the source signal acquired under the fourth measurement condition is even smaller than the signal component due to cantilever resonance included in the source signal acquired under the third measurement condition.

[0139] 25 to 27 are plots of the vibration rectification error VRE(k) obtained by a product-sum operation with k=1 to 2048 and N=2048 using source signals acquired under each of four measurement conditions. In Figs. 25 to 27, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum and minimum values ​​is a constant value. In Figs. 25 to 27, the product-sum operation formula used to calculate the vibration rectification error VRE(k) is different.

[0140] FIG. 25 is a plot of the vibration rectification error VRE(k) obtained by the product-sum calculation of the above-mentioned formula (3). In FIG. 25, A1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. A2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. A3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. A4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition. In any of A1 to A4, the periodicity of the vibration rectification error VRE(k) cannot be clearly confirmed.

[0141] FIG. 26 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (4). In FIG. 26, B1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. B2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. B3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. B4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition. The noise components contained in the first signal are reduced by performing smoothing filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in B1 to B3. The interval between two adjacent maximum values ​​of the vibration rectification error VRE(k) corresponds to the period of the cantilever resonance. In B4, the periodicity of the vibration rectification error VRE(k) is unclear. Also, from B1 to B4, it can be seen that the larger the signal component due to cantilever resonance contained in the source signal, the clearer the periodicity of the vibration rectification error VRE(k). Note that in Figures 21 and 22, the source signal contains a signal component with a frequency that is half the cantilever resonance frequency, and due to the influence of this signal component, the maximum value of the vibration rectification error VRE(k) increases and decreases in B1 and B2.

[0142] FIG. 27 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (5). In FIG. 27, C1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. C2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. C3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. C4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition. The noise components contained in the first signal are reduced by performing smoothing filter processing on the processing target signal, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the processing target signal. In any of C1 to C4, the periodicity of the vibration rectification error VRE(k) can be confirmed. Moreover, the periodicity of the vibration rectification error VRE(k) is clearer in C1 to C4 compared to B1 to B4 in Fig. 26. Moreover, it can be seen from C1 to C4 that the periodicity of the vibration rectification error VRE(k) becomes clearer as the signal component due to cantilever resonance contained in the source signal becomes larger.

[0143] 25 to 27, the periodicity of the vibration rectification error VRE(k) is clearer in the order of the first measurement condition, the second measurement condition, the third measurement condition, and the fourth measurement condition, and it can be seen that the larger the signal component due to cantilever resonance, the higher the detection accuracy of that signal component.

[0144] 28 to 30 are plots of vibration rectification error VRE(k) obtained by product-sum calculation for four cases of N=2048, 512, 128, and 32, where k=1 to 2048, using a source signal acquired under the second measurement condition. In Figs. 28 to 30, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum and minimum values ​​is a constant value. In Figs. 28 to 30, the formula for product-sum calculation used to calculate the vibration rectification error VRE(k) is different.

[0145] Fig. 28 is a plot of the vibration rectification error VRE(k) obtained by the product-sum calculation of the above-mentioned formula (3). In Fig. 28, D1 is the vibration rectification error VRE(k) obtained when N=2048. D2 is the vibration rectification error VRE(k) obtained when N=512. D3 is the vibration rectification error VRE(k) obtained when N=128. D4 is the vibration rectification error VRE(k) obtained when N=32. In any of D1 to D4, the periodicity of the vibration rectification error VRE(k) cannot be clearly confirmed.

[0146] FIG. 29 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (4). In FIG. 29, E1 is the vibration rectification error VRE(k) obtained when N=2048. E2 is the vibration rectification error VRE(k) obtained when N=512. E3 is the vibration rectification error VRE(k) obtained when N=128. E4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing smoothing filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in E1. In E2 to E4, the number of additions N in the product-sum operation is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear.

[0147] FIG. 30 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (5). In FIG. 30, F1 is the vibration rectification error VRE(k) obtained when N=2048. F2 is the vibration rectification error VRE(k) obtained when N=512. F3 is the vibration rectification error VRE(k) obtained when N=128. F4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing smoothing filter processing on the signal to be processed, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the signal to be processed. Compared to E1 in FIG. 29, the periodicity of the vibration rectification error VRE(k) can be clearly confirmed in F1. In F2 to F4, the product-sum operation The number of additions N is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear.

[0148] 31 to 33 are plots of vibration rectification error VRE(k) obtained by product-sum calculation for four cases of N=2048, 512, 128, and 32, where k=1 to 2048, using a source signal acquired under the first measurement condition. In Figs. 31 to 33, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum and minimum values ​​is a constant value. In Figs. 31 to 33, the formula for product-sum calculation used to calculate the vibration rectification error VRE(k) is different.

[0149] FIG. 31 is a plot of the vibration rectification error VRE(k) obtained by the product-sum calculation of the above-mentioned formula (3). In FIG. 31, G1 is the vibration rectification error VRE(k) obtained when N=2048. G2 is the vibration rectification error VRE(k) obtained when N=512. G3 is the vibration rectification error VRE(k) obtained when N=128. G4 is the vibration rectification error VRE(k) obtained when N=32. In any of G1 to G4, the periodicity of the vibration rectification error VRE(k) cannot be clearly confirmed.

[0150] FIG. 32 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (4). In FIG. 32, H1 is the vibration rectification error VRE(k) obtained when N=2048. H2 is the vibration rectification error VRE(k) obtained when N=512. H3 is the vibration rectification error VRE(k) obtained when N=128. H4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing smoothing filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in H1 to H3. In H4, the number of additions N of the product-sum operation is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear. In addition, since the signal component due to cantilever resonance contained in the source signal is larger than in the case of Figure 29, it can be seen that the periodicity of the vibration rectification error VRE(k) is clearer in H1 to H3 than in E1 to E3 in Figure 29.

[0151] FIG. 33 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the above-mentioned formula (5). In FIG. 33, I1 is the vibration rectification error VRE(k) obtained when N=2048. I2 is the vibration rectification error VRE(k) obtained when N=512. I3 is the vibration rectification error VRE(k) obtained when N=128. I4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing smoothing filter processing on the signal to be processed, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the signal to be processed. In any of I1 to I4, the periodicity of the vibration rectification error VRE(k) can be confirmed. In addition, compared with H1 to H4 in FIG. 32, the periodicity of the vibration rectification error VRE(k) can be clearly confirmed in I1 to I4. In addition, since the signal component due to cantilever resonance contained in the source signal is larger than in the case of Figure 30, it can be seen that the periodicity of the vibration rectification error VRE(k) is clearer in I1 to I4 than in E1 to E4 in Figure 30.

[0152] 28 to 33, the periodicity of the vibration rectification error VRE(k) is clear in the order of N=2048, 512, 128, and 32, and it can be seen that the detection accuracy of the signal component due to cantilever resonance increases as the number of additions N in the product-sum calculation process increases.

[0153] 1-9.Effects As described above, the first signal based on the processing target signal generated based on the source signal output from the sensor module 1 and the second signal based on the phase-shifted signal of the processing target signal contain a periodic signal component generated by the sensor module 1, specifically, a color filter. The first and second signals include a signal component due to cantilever resonance in common. Therefore, by the product-sum operation process of the first signal and the second signal, the signal component having ergodicity such as noise is attenuated, while the signal component due to cantilever resonance strengthens or weakens each other depending on the phase difference between the first signal and the second signal. As a result, the vibration rectification errors obtained by performing the product-sum operation process multiple times with changing the shift amount differ in magnitude depending on the phase difference between the first signal and the second signal and the period of the signal component due to the cantilever resonance. Therefore, according to the first embodiment, the signal processing device 400 can detect the signal component due to the cantilever resonance included in the signal output from the sensor module 1 without performing envelope processing. Note that the signal processing device 400 may reduce the measurement error when detecting the periodicity by performing the product-sum operation process multiple times with the same shift amount for a part of the signals.

[0154] Furthermore, according to the first embodiment, the signal processing device 400 only needs to acquire the source signal for a predetermined time once to calculate multiple vibration rectification errors, enabling high-speed calculation processing, and since environmental changes such as temperature changes during the short period of time in which the source signal is acquired are extremely small, calculation errors caused by environmental changes are reduced.

[0155] Furthermore, according to the first embodiment, if the first signal is a signal obtained by smoothing filter processing of the signal to be processed, high-frequency noise components contained in the first signal are reduced, and if the second signal is a signal obtained by smoothing filter processing of a phase-shifted signal of the signal to be processed, high-frequency noise components contained in the second signal are reduced, thereby improving the detection accuracy of the signal component due to cantilever resonance.

[0156] Furthermore, according to the first embodiment, if the first signal is a signal obtained by removing or reducing the DC component from the signal to be processed, each sample value of the first signal becomes small, and if the second signal is a signal obtained by shifting the phase of the signal to be processed and removing or reducing the DC component, each sample value of the second signal becomes small, thereby reducing the load of the product-sum calculation between the first signal and the second signal and improving the calculation accuracy.

[0157] Furthermore, according to the first embodiment, by making the number of additions N in the product-sum calculation process larger than the value obtained by dividing the sampling frequency of the source signal by the cantilever resonance frequency, the signal components due to the cantilever resonance contained in the first signal and the second signal are accumulated for one period or more, so that the signal components are effectively detected.

[0158] 2. Second embodiment In the following, in the second embodiment, the same components as those in the first embodiment are given the same reference numerals, explanations that overlap with the first embodiment are omitted or simplified, and differences from the first embodiment are mainly described.

[0159] In the first embodiment, smoothing filter processing is used as filter processing for the signal to be processed or the signal obtained by shifting the phase of the processed signal. When the source signal contains a signal component with a frequency half the cantilever resonance frequency, the signal component is not reduced by smoothing filter processing, so that the maximum value of the vibration rectification error VRE(k) such as C1 in Fig. 27 increases and decreases and is not constant. Therefore, in the second embodiment, band-limiting filter processing with only the vicinity of the cantilever resonance frequency as the filter processing for the signal to be processed or the signal obtained by shifting the phase of the processed signal is used to reduce the influence of signal components with a frequency such as half the cantilever resonance frequency on the calculated vibration rectification error VRE(k).

[0160] For example, when the first signal is a signal obtained by band-limiting filtering of the signal to be processed, and the second signal is a signal obtained by shifting the phase of the signal to be processed, the k-th vibration rectification error VRE(k) is calculated by the following formula (6). BPF (S(i)) is , is the i-th sample value of the first signal, and S(i+k) is the i-th sample value of the second signal.

[0161]

number

[0162] For example, when the first signal is a signal obtained by performing band-limiting filtering on the signal to be processed, and the second signal is a signal obtained by performing smoothing filtering on a signal obtained by shifting the phase of the signal to be processed, the k-th vibration rectification error VRE(k) is calculated by the formula (7). BPF (S(i)) is the i-th sample value of the first signal, and f LPF (S(i+k)) is the i-th sample value of the second signal.

[0163]

number

[0164] In addition, in equations (6) and (7), the division by N may be omitted. Furthermore, in equations (6) and (7), a first signal obtained by subjecting the target signal to band-limiting filtering is used, but a second signal obtained by subjecting a phase-shifted signal obtained by subjecting the target signal to band-limiting filtering may be used.

[0165] The procedure of the signal processing method of the second embodiment is the same as that of Fig. 19, and therefore illustration and description thereof will be omitted. Also, the configuration of the signal processing device 400 of the second embodiment is the same as that of Fig. 18, and therefore illustration and description thereof will be omitted.

[0166] Fig. 34 and Fig. 35 are plots of the vibration rectification error VRE(k) obtained by a product-sum operation with k = 1 to 2048 and N = 2048 using source signals acquired under each of the four measurement conditions. In Fig. 34 and Fig. 35, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum value and the minimum value is a constant value. In Fig. 34 and Fig. 35, the product-sum operation formula used to calculate the vibration rectification error VRE(k) is different.

[0167] FIG. 34 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of Equation (6). In FIG. 34, J1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition described above. J2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition described above. J3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition described above. J4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition described above. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in J1 to J3. The interval between two adjacent maximum values ​​of the vibration rectification error VRE(k) corresponds to the period of the cantilever resonance. In J4, the periodicity of the vibration rectification error VRE(k) is unclear. Also, from J1 to J4, it can be seen that the larger the signal component due to cantilever resonance contained in the source signal, the clearer the periodicity of the vibration rectification error VRE(k). Furthermore, since the signal component at half the frequency of the cantilever resonance frequency is reduced by performing band-limiting filter processing on the signal to be processed, the increase and decrease range of the maximum value of the vibration rectification error VRE(k) is reduced in J1 and J2 compared to B1 and B2 in Fig. 26.

[0168] FIG. 35 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of formula (7). In FIG. 35, K1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. K2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. K3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. K4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the signal to be processed, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the signal to be processed. In any of K1 to K4, the periodicity of the vibration rectification error VRE(k) can be confirmed. In addition, it can be seen from K1 to K4 that the periodicity of the vibration rectification error VRE(k) becomes clearer as the signal component due to cantilever resonance contained in the source signal becomes larger. Furthermore, since the signal to be processed is subjected to band-limiting filter processing, the signal components at half the frequency of the cantilever resonance frequency are reduced, and therefore, the increase / decrease range of the maximum value of the vibration rectification error VRE(k) is reduced in K1 and K2 compared to C1 and C2 in FIG. 27.

[0169] Fig. 36 and Fig. 37 are plots of vibration rectification error VRE(k) obtained by product-sum calculation for four cases of N=2048, 512, 128, and 32, where k=1 to 2048, using a source signal acquired under the second measurement condition. In Fig. 36 and Fig. 37, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum value and the minimum value is a constant value. In Fig. 36 and Fig. 37, the formula of product-sum calculation used to calculate the vibration rectification error VRE(k) is different.

[0170] FIG. 36 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the formula (6). In FIG. 36, L1 is the vibration rectification error VRE(k) obtained when N=2048. L2 is the vibration rectification error VRE(k) obtained when N=512. L3 is the vibration rectification error VRE(k) obtained when N=128. L4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in L1 and L2. In L3 and L4, the number of additions N in the product-sum operation is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear. Furthermore, since the signal to be processed is subjected to band-limiting filter processing, the signal components at half the frequency of the cantilever resonant frequency are reduced, and therefore, in L1, the increase / decrease range of the maximum value of the vibration rectification error VRE(k) is reduced compared to E1 in Figure 29.

[0171] FIG. 37 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the formula (7). In FIG. 37, M1 is the vibration rectification error VRE(k) obtained when N=2048. M2 is the vibration rectification error VRE(k) obtained when N=512. M3 is the vibration rectification error VRE(k) obtained when N=128. M4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the signal to be processed, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the signal to be processed. In M1 to M3, the periodicity of the vibration rectification error VRE(k) can be confirmed. In addition, compared with L1 and L2 in FIG. 36, the periodicity of the vibration rectification error VRE(k) can be clearly confirmed in M1 and M2. In M4, the number of additions N in the product-sum operation is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear. Furthermore, the signal component at half the frequency of the cantilever resonance frequency is reduced by performing band-limiting filter processing on the signal to be processed, so the increase and decrease range of the maximum value of the vibration rectification error VRE(k) is reduced in M1 compared to F1 in Fig. 30.

[0172] Fig. 38 and Fig. 39 are plots of vibration rectification error VRE(k) obtained by product-sum calculation for four cases of N=2048, 512, 128, and 32, where k=1 to 2048, using a source signal acquired under the first measurement condition. In Fig. 38 and Fig. 39, the horizontal axis is k and the vertical axis is VRE(k), and VRE(k) is normalized so that the difference between the maximum value and the minimum value is a constant value. In Fig. 38 and Fig. 39, the formula of product-sum calculation used to calculate the vibration rectification error VRE(k) is different.

[0173] FIG. 38 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the formula (6). In FIG. 38, N1 is the vibration rectification error VRE(k) obtained when N=2048. N2 is the vibration rectification error VRE(k) obtained when N=512. N3 is the vibration rectification error VRE(k) obtained when N=128. N4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the processing target signal, and the periodicity of the vibration rectification error VRE(k) can be confirmed in N1 to N3. In N4, the number of additions N in the product-sum operation is insufficient, and the periodicity of the vibration rectification error VRE(k) is unclear. Also, since the signal component due to cantilever resonance contained in the source signal is larger compared to the case of Fig. 36, it can be seen that the periodicity of the vibration rectification error VRE(k) is clearer in N1 to N3 compared to L1 to L3 in Fig. 36. Furthermore, since the signal component at half the frequency of the cantilever resonance frequency is reduced by performing band-limiting filter processing on the signal to be processed, the increase / decrease range of the maximum value of the vibration rectification error VRE(k) is smaller in N1 to N3 compared to H1 to H3 in Fig. 32.

[0174] FIG. 39 is a plot of the vibration rectification error VRE(k) obtained by the product-sum operation of the formula (7). In FIG. 39, O1 is the vibration rectification error VRE(k) obtained when N=2048. O2 is the vibration rectification error VRE(k) obtained when N=512. O3 is the vibration rectification error VRE(k) obtained when N=128. O4 is the vibration rectification error VRE(k) obtained when N=32. The noise components contained in the first signal are reduced by performing band-limiting filter processing on the signal to be processed, and the noise components contained in the second signal are reduced by performing smoothing filter processing on the signal obtained by shifting the phase of the signal to be processed. In any of O1 to O4, the periodicity of the vibration rectification error VRE(k) can be confirmed. In addition, compared with N1 to N4 in FIG. 38, the periodicity of the vibration rectification error VRE(k) can be clearly confirmed in O1 to O4. Also, since the signal component due to cantilever resonance contained in the source signal is larger compared to the case of Fig. 37, it can be seen that the periodicity of the vibration rectification error VRE(k) is clearer in O1 to O4 compared to M1 to M4 in Fig. 37. Furthermore, since the signal component at half the frequency of the cantilever resonance frequency is reduced by performing band-limiting filter processing on the signal to be processed, the increase / decrease range of the maximum value of the vibration rectification error VRE(k) is smaller in M1 to M4 compared to I1 to I4 in Fig. 33.

[0175] The second embodiment described above provides the same effects as the first embodiment. Furthermore, according to the second embodiment, if the first signal is a signal obtained by band-limiting filtering the signal to be processed, many signal components other than the signal component of the resonant frequency contained in the first signal are reduced, and if the second signal is a signal obtained by band-limiting filtering the signal obtained by shifting the phase of the signal to be processed, many signal components other than the signal component of the resonant frequency contained in the second signal are reduced, thereby improving the detection accuracy of the signal component of the resonant frequency.

[0176] 3. Third embodiment In the following description of the third embodiment, the same components as those in the first or second embodiment are denoted by the same reference numerals, and descriptions that overlap with those in the first or second embodiment are omitted or simplified. The differences from the first and second embodiments will be described.

[0177] In the first and second embodiments, when a part of a time series signal included in a source signal is cut out to generate a processing target signal, the first sample value and the last sample value become discontinuous in the first signal based on the processing target signal or the second signal based on a signal obtained by shifting the phase of the processing target signal, and the accuracy of the vibration rectification error obtained by the product-sum operation may decrease due to noise caused by this discontinuity. Therefore, in the third embodiment, in order to reduce the discontinuity of the sample value, the first signal is a signal obtained by multiplying the processing target signal by a window function, or the second signal is a signal obtained by multiplying the processing target signal by a window function. The type of the window function is not particularly limited, and examples of the window function include a Hanning window function, a rectangular window function, a Gaussian window function, a Hamming window function, a Blackman window function, and a Kaiser window function.

[0178] The first signal may be a signal obtained by filtering the signal to be processed and applying a window function. For example, the filtering may be smoothing filtering or band-limiting filtering. The first signal may be a signal obtained by removing or reducing a direct current component from the signal to be processed and applying a window function. The first signal may be a signal obtained by removing or reducing a direct current component from the signal to be processed, filtering, and applying a window function.

[0179] The second signal may be a signal obtained by filtering a signal obtained by shifting the phase of the signal to be processed and applying a window function. For example, the filtering may be a smoothing filter process or a band-limiting filter process. The second signal may be a signal obtained by removing or reducing a direct current component from a signal obtained by shifting the phase of the signal to be processed and applying a window function. The second signal may be a signal obtained by removing or reducing a direct current component from a signal obtained by shifting the phase of the signal to be processed, filtering, and applying a window function.

[0180] For example, when the first signal is a signal obtained by removing or reducing the DC component of the signal to be processed, filtering, and applying a window function to the signal to be processed, and the second signal is a signal obtained by shifting the phase of the signal to be processed and filtering the same, the k-th vibration rectification error VRE(k) is calculated by the formula (8). window is the window function. Also, F window (i)·f LPF (f HPF (S(i)) is the i-th sample value of the first signal, and f LPF (S(i+k)) is the i-th sample value of the second signal.

[0181]

number

[0182] In addition, in equation (8), the division by N may be omitted. Furthermore, in equation (8), a first signal obtained by multiplying the target signal by a window function is used, but a second signal obtained by multiplying a phase-shifted signal obtained by multiplying the target signal by a window function may be used.

[0183] The procedure of the signal processing method of the third embodiment is the same as that of Fig. 19, and therefore illustration and description thereof will be omitted. Also, the configuration of the signal processing device 400 of the third embodiment is the same as that of Fig. 18, and therefore illustration and description thereof will be omitted.

[0184] FIG. 40 is a plot of the vibration rectification error VRE(k) obtained by a product-sum operation with k=1 to 2048 and N=256 using the source signal acquired under the second measurement condition. In FIG. 40, the horizontal axis is k and the vertical axis is VRE(k). VRE(k) is a value between the maximum value and the minimum value. The difference between the minimum value and the maximum value is normalized to a constant value. In FIG. 40, the product-sum calculation formula used to calculate the vibration rectification error VRE(k) is different.

[0185] In FIG. 40, P1 is the window function F window is the vibration rectified error VRE(k) obtained by equation (8) using as a Hanning window function. P2 is the vibration rectified error VRE(k) obtained by the above equation (5). By applying a window function to the signal to be processed, the discontinuity of the sample values ​​is mitigated, so the periodicity of the vibration rectified error VRE(k) is clearer in P1 than in P2.

[0186] The third embodiment described above provides the same effects as those of the first or second embodiment. Furthermore, according to the third embodiment, if the first signal is a signal obtained by multiplying the target signal by a window function, discontinuity between the first sample value and the last sample value of the first signal is alleviated, and if the second signal is a signal obtained by multiplying the target signal by a window function with a phase shift, discontinuity between the first sample value and the last sample value of the second signal is alleviated, thereby improving the detection accuracy of a signal component having periodicity.

[0187] 4. Fourth embodiment Hereinafter, for the fourth embodiment, components similar to those of any of the first to third embodiments will be given the same symbols, explanations that overlap with those of any of the first to third embodiments will be omitted or simplified, and the following will mainly describe the contents that are different from any of the first to third embodiments.

[0188] Fig. 41 is a diagram showing an example of the configuration of a signal processing device 400 according to the fourth embodiment. As shown in Fig. 41, the signal processing device 400 includes a processing circuit 410, a storage circuit 420, an operation unit 430, a display unit 440, a sound output unit 450, and a communication unit 460. Note that the signal processing device 400 may be configured such that some of the components in Fig. 41 are omitted or changed, or other components are added.

[0189] The configurations and functions of the memory circuitry 420, the operation section 430, the display section 440, the sound output section 450, and the communication section 460 are similar to those of any of the first to third embodiments, and therefore will not be described.

[0190] The processing circuit 410 executes a signal processing program 421 to function as a source signal acquisition circuit 411, a processed signal generation circuit 412, a vibration rectification error calculation circuit 413, a resonance frequency calculation circuit 414, and a determination circuit 415. That is, the signal processing device 400 includes the source signal acquisition circuit 411, the processed signal generation circuit 412, the vibration rectification error calculation circuit 413, the resonance frequency calculation circuit 414, and the determination circuit 415.

[0191] The functions of the source signal acquisition circuit 411, the processing target signal generation circuit 412, and the vibration rectification error calculation circuit 413 are similar to those of any of the first to third embodiments, and therefore a description thereof will be omitted.

[0192] The resonance frequency calculation circuit 414 calculates the resonance frequency of the sensor module 1, which is the target, based on the multiple vibration rectification errors calculated by the vibration rectification error calculation circuit 413. The resonance frequency of the sensor module 1 calculated by the resonance frequency calculation circuit 414 is stored in the memory circuit 420 as a resonance frequency 424. For example, the resonance frequency of the sensor module 1 is the cantilever resonance frequency. Specifically, the resonance frequency calculation circuit 414 reads out the vibration rectification error information 423 stored in the memory circuit 420, and acquires two values ​​of k, k1 and k2, at which the vibration rectification error VRE(k) becomes two successive maximum values ​​or two successive minimum values. Then, when the frequency of the measured signal SIN is fx, the resonance frequency calculation circuit 414 calculates , the cantilever resonance frequency f CL can be calculated.

[0193]

number

[0194] The resonance frequency calculation circuit 414 obtains three or more values ​​of k that result in three or more consecutive maximum values ​​or three or more consecutive minimum values, calculates a plurality of cantilever resonance frequencies using the formula (9), and calculates the average value of these frequencies as the cantilever resonance frequency f CL It may be calculated as:

[0195] The determination circuit 415 determines whether the resonance frequency calculated by the resonance frequency calculation circuit 414 is correct or not based on the difference between the maximum value and the minimum value of the multiple vibration rectification errors calculated by the vibration rectification error calculation circuit 413. The determination result by the determination circuit 415 is stored in the storage circuit 420 as a determination result 425. Specifically, the resonance frequency calculation circuit 414 reads out the vibration rectification error information 423 stored in the storage circuit 420, acquires the maximum value and the minimum value of the vibration rectification error VRE(k), and calculates the difference between them. Then, the determination circuit 415 compares the difference between the maximum value and the minimum value with a predetermined threshold value, and determines that the calculated resonance frequency is correct if the difference is greater than the threshold value, and determines that the calculated resonance frequency is incorrect if the difference is less than the threshold value. That is, the determination circuit 415 determines that the resonance frequency calculated in an environment in which the cantilever resonance is sufficiently excited is reliable and therefore the resonance frequency is correct.

[0196] In addition, since the sensitivity of the physical quantity sensor 200 is strongly correlated with the cantilever resonance frequency, the determination circuit 415 can also check the sensitivity anomaly of the physical quantity sensor 200 based on the resonance frequency calculated by the resonance frequency calculation circuit 414. For example, if the weight fixed to the cantilever is missing for some reason, the mass of the cantilever decreases, and the cantilever resonance frequency shifts to a higher frequency. At the same time, the sensitivity of the physical quantity sensor 200 decreases, which appears as a sensitivity anomaly of the physical quantity sensor 200. In addition, if the cantilever is damaged by a strong impact or the like, this also appears as a sensitivity anomaly of the physical quantity sensor 200, and the cantilever resonance frequency also shifts. Therefore, identifying the cantilever resonance frequency is one method for determining whether the sensitivity of the physical quantity sensor 200 is within the specifications. Therefore, the determination circuit 415 can determine whether the sensitivity of the physical quantity sensor 200 is within the specifications based on whether the resonance frequency calculated by the resonance frequency calculation circuit 414 is within a predetermined frequency range.

[0197] The RAM of the storage circuit 420 stores signals generated and information calculated by the processing circuit 410, such as a processing target signal 422, vibration rectification error information 423, a resonance frequency 424, and a determination result 425.

[0198] The display unit 440 may display an image in which the vibration rectification error information 423 is plotted, and information including the resonance frequency 424 and the determination result 425, based on a display signal output from the processing circuit 410.

[0199] At least some of the source signal acquisition circuit 411, the processing target signal generation circuit 412, the vibration rectification error calculation circuit 413, the resonance frequency calculation circuit 414, and the judgment circuit 415 may be realized by dedicated hardware. The signal processing device 400 may be a single device or may be configured by multiple devices. For example, the processing circuit 410 and the storage circuit 420 may be realized by a device such as a cloud server, which calculates the vibration rectification error information 423, the resonance frequency 424, and the judgment result 425, and stores the calculated vibration rectification error information 423, the resonance frequency 424, and the judgment result 425. The number 424 and the determination result 425 may be transmitted to a terminal including an operation unit 430, a display unit 440, a sound output unit 450, and a communication unit 460 via a communication line.

[0200] FIG. 42 is a flow chart showing the procedure of the signal processing method according to the fourth embodiment.

[0201] 42, the signal processing method of the fourth embodiment includes a source signal acquisition step S1, a processing target signal generation step S2, a vibration rectification error calculation step S3, a resonance frequency calculation step S4, and a determination step S5. The signal processing method of the present embodiment is performed by, for example, a signal processing device 400.

[0202] The signal processing device 400 first performs a source signal obtaining step S1, a processing target signal generating step S2, and a vibration rectification error calculating step S3, in the same manner as in any of the first to third embodiments.

[0203] Next, in a resonance frequency calculation step S4, the signal processing device 400 calculates the resonance frequency of the sensor module 1, which is the target, based on the multiple vibration rectification errors calculated in step S3.

[0204] Finally, in a determination step S5, the signal processing device 400 determines whether the resonance frequency calculated in step S4 is correct based on the difference between the maximum and minimum values ​​of the multiple vibration rectification errors calculated in step S3. Furthermore, in the determination step S5, the signal processing device 400 may determine whether the sensitivity of the physical quantity sensor 200 is within the specifications based on the resonance frequency calculated in step S4.

[0205] In the judgment step S5, the signal processing device 400 may calculate the difference between the maximum value and the minimum value of the multiple vibration rectification errors calculated in step S3, and output the difference value as a judgment index for the correctness of the resonance frequency instead of judging the correctness of the resonance frequency. In this case, the signal processing device 400 may judge the correctness of the resonance frequency based on the judgment index.

[0206] Figures 43 and 44 are plots of the vibration rectification error VRE(k) obtained by a product-sum operation with k = 1 to 2048 and N = 2048 using source signals acquired under each of the four measurement conditions described above. In Figures 43 and 44, the horizontal axis is k and the vertical axis is VRE(k). The product-sum operation formula used to calculate the vibration rectification error VRE(k) is different between Figures 43 and 44.

[0207] FIG. 43 is a plot of the vibration rectification error VRE(k) obtained by the product-sum calculation of the above-mentioned formula (4). In FIG. 43, Q1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. Q2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. Q3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. Q4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition.

[0208] FIG. 44 is a plot of the vibration rectification error VRE(k) obtained by the product-sum calculation of the above-mentioned formula (5). In FIG. 44, R1 is the vibration rectification error VRE(k) obtained using the source signal acquired under the first measurement condition. R2 is the vibration rectification error VRE(k) obtained using the source signal acquired under the second measurement condition. R3 is the vibration rectification error VRE(k) obtained using the source signal acquired under the third measurement condition. R4 is the vibration rectification error VRE(k) obtained using the source signal acquired under the fourth measurement condition.

[0209] For example, in the signal processing device 400, since the difference between the maximum and minimum values ​​of VRE(k) in Q1, Q2, R1, and R2 is greater than the threshold, the signal processing device 400 uses VRE(k) of Q1, Q2, R1, and R2. The calculated resonant frequency is determined to be correct, and whether or not the sensitivity of the physical quantity sensor 200 is within the specifications is determined based on the calculated resonant frequency. On the other hand, the difference between the maximum and minimum values ​​of VRE(k) for Q3, Q4, R3, and R4 is smaller than the threshold value, and the signal processing device 400 determines that the resonant frequency calculated using VRE(k) for Q3, Q4, R3, and R4 is incorrect.

[0210] The fourth embodiment described above provides the same effects as the first to third embodiments. Furthermore, according to the fourth embodiment, the signal processing device 400 calculates the cantilever resonance frequency, so that the user or the signal processing device 400 can perform various analyses based on the cantilever resonance frequency.

[0211] Furthermore, according to the fourth embodiment, the greater the cantilever resonance is excited, the greater the difference between the maximum and minimum values ​​of the calculated vibration rectification errors, and the signal processing device 400 can determine whether the calculated cantilever resonance frequency is correct or not based on the difference. For example, only when the calculated cantilever resonance frequency is determined to be correct, the user or the signal processing device 400 can perform various analyses based on the cantilever resonance frequency, for example, determine whether the sensitivity of the physical quantity sensor 200 is within the specifications.

[0212] 5. Fifth embodiment Hereinafter, for the fifth embodiment, components similar to those of any of the first to fourth embodiments will be given the same symbols, explanations that overlap with those of any of the first to fourth embodiments will be omitted or simplified, and the following will mainly describe the differences from any of the first to fourth embodiments.

[0213] As described with reference to FIG. 16, in the frequency ratio measurement circuit 202 provided in the physical quantity measurement device 2 of the sensor module 1, the vibration rectification error changes in a constant period with respect to the change in the group delay amount of the first low-pass filter 310. This period is determined by the cantilever resonance frequency and the frequency of the physical quantity detection element 40, and when the cantilever resonance frequency or the frequency of the physical quantity detection element 40 changes, the fluctuation period of the vibration rectification error also changes. Therefore, by measuring the period of change in the vibration rectification error with respect to the change in the group delay amount of the first low-pass filter 310, it is possible to obtain a judgment index for whether the sensitivity of the physical quantity sensor 200 is within the specifications. Therefore, in the fifth embodiment, a plurality of vibration rectification errors are generated by changing the group delay amount of the first low-pass filter 310.

[0214] In the fifth embodiment, the structure and functional configuration of the sensor module 1 are the same as those in FIGS. 1 to 8, and therefore are not illustrated.

[0215] In the fifth embodiment, the physical quantity measuring device 2 of the sensor module 1 has a normal operation mode for measuring the frequency ratio between the above-mentioned measured signal SIN and the reference signal CLK, and an inspection mode for checking the sensitivity of the physical quantity sensor 200. The micro control unit 210 receives a predetermined command from the signal processing device 400 via the interface circuit 230, whereby the physical quantity measuring device 2 is set to the normal operation mode or the inspection mode. For example, in the manufacturing process of the sensor module 1, the signal processing device 400 may set the physical quantity measuring device 2 to the inspection mode, and the physical quantity measuring device 2 may check the sensitivity of the physical quantity sensor 200. The signal processing device 400 may select non-defective products of the sensor module 1 based on the result of the sensitivity check. Alternatively, after the sensor module 1 is installed and before it is operated, the signal processing device 400 may set the physical quantity measuring device 2 to the inspection mode, and the physical quantity measuring device 2 may check the sensitivity of the physical quantity sensor 200. If there is no abnormality in the sensitivity of the physical quantity sensor 200 based on the result of the sensitivity check, the signal processing device 400 sets the physical quantity measuring device 2 to a normal operation mode and operates the sensor module 1. In the normal operation mode, the vibration adjustment A measurement value in which the flow error is corrected can be obtained. Furthermore, the signal processing device 400 may periodically set the physical quantity measuring device 2 to an inspection mode, and the physical quantity measuring device 2 may check the sensitivity of the physical quantity sensor 200. Note that the normal operation mode is an example of a "first operation mode", and the inspection mode is an example of a "second operation mode".

[0216] In the inspection mode, the physical quantity sensor 200 is operated in a stable vibration environment, and the micro control unit 210 of the physical quantity measuring device 2 functions as a control circuit to acquire the group delay amount dependency of the vibration rectification error based on the output signal of the physical quantity sensor 200 while changing the group delay amount of the first low-pass filter 310. To achieve this, the micro control unit 210 first sets the cutoff frequency of the second low-pass filter 330 lower than that in the normal operation mode. Specifically, the micro control unit 210 sets the cutoff frequency of the second low-pass filter 330 to, for example, several Hz so that the vibration rectification error included in the output value of the second low-pass filter 330 is emphasized. For example, the micro control unit 210 may lower the cutoff frequency by increasing the number of taps of the second low-pass filter 330 compared to that in the normal operation mode.

[0217] Furthermore, the micro control unit 210 acquires the vibration rectification error of the measurement value while sequentially changing the tap number na for the first low-pass filter 310 having the configuration shown in FIG. 14, and stores the tap number and the vibration rectification error in correspondence with each other in the memory section 220.

[0218] The signal processing device 400 reads out the correspondence information between the number of taps and the vibration rectification error from the storage unit 220 via the interface circuit 230, and calculates the period in which the vibration rectification error changes from a graph plotting the relationship between the number of taps and the vibration rectification error as shown in Fig. 17. Since this period is determined by the cantilever resonance frequency and the frequency of the physical quantity detection element 40, the signal processing device 400 can back-calculate the cantilever resonance frequency. Based on the calculated cantilever resonance frequency, the signal processing device 400 can determine whether the sensitivity of the physical quantity sensor 200 is within the specifications.

[0219] Alternatively, the micro control unit 210 may read out correspondence information between the number of taps and the vibration rectification error from the memory unit 220, calculate the cantilever resonance frequency based on a graph plotting the relationship between the number of taps and the vibration rectification error, and determine whether the sensitivity of the physical quantity sensor 200 is within the specifications.

[0220] The first low-pass filter 310 is an example of a "first filter." The second low-pass filter 330 is an example of a "second filter."

[0221] FIG. 45 is a flowchart showing an example of the procedure of the signal processing method according to the fifth embodiment.

[0222] As shown in FIG. 45, first, when the normal operation mode is set in step S110, the physical quantity measuring device 2 measures the frequency ratio between the signal under measurement SIN and the reference signal CLK in step S120.

[0223] In step S130, the physical quantity measuring device 2 repeats step S120 until the measurement is completed.

[0224] If the normal operation mode is not set in step S110 and the inspection mode is set in step S140, the physical quantity measuring device 2 sets the cutoff frequency of the second low-pass filter 330 to a lower frequency than in the normal operation mode in step S150.

[0225] Next, in step S160, the physical quantity measuring device 2 performs The amount of group delay is set to a predetermined value. Specifically, the physical quantity measuring device 2 sets the number of taps na to a predetermined value.

[0226] Next, in step S170, the physical quantity measuring device 2 acquires the output value of the second low-pass filter 330.

[0227] Next, in step S180, the physical quantity measuring device 2 determines whether or not all output values ​​of the second low-pass filter 330 required for sensitivity determination have been acquired.

[0228] If acquisition of the necessary output values ​​has not been completed, in step S190, the physical quantity measurement device 2 changes the group delay amount of the first low-pass filter 310. Specifically, the physical quantity measurement device 2 changes the number of taps na.

[0229] Then, when acquisition of the necessary output values ​​has been completed, in step S200, the signal processing device 400 or the physical quantity measuring device 2 calculates the period of change in the vibration rectification error using the multiple output values ​​of the second low-pass filter 330 acquired in step S170.

[0230] Next, in step S210, the signal processing device 400 or the physical quantity measuring device 2 calculates the cantilever resonance frequency from the period of change in the vibration rectification error.

[0231] Next, in step S220, the signal processing device 400 or the physical quantity measuring device 2 determines whether the sensitivity of the physical quantity sensor 200 is within specifications based on the cantilever resonance frequency.

[0232] Then, in step S230, the inspection mode of the physical quantity measuring device 2 is ended, and step S110 and subsequent steps are repeated.

[0233] In the fifth embodiment described above, in the normal operation mode of the physical quantity measuring device 2, as described in the first embodiment, the first low-pass filter 310 operates in synchronization with the signal SIN to be measured, and the second low-pass filter 330 operates in synchronization with the reference signal CLK different from the signal SIN to be measured, so that nonlinearity occurs in the relationship between the frequency delta-sigma modulated signal input to the first low-pass filter 310 and the output signal of the second low-pass filter 330. The vibration rectification error caused by this nonlinearity changes depending on the group delay amount of the first low-pass filter 310. Therefore, according to the fifth embodiment, in the normal operation mode of the physical quantity measuring device 2, by setting the group delay amount of the first low-pass filter 310 to an appropriate value, the vibration rectification error caused by this nonlinearity and the vibration rectification error caused by the asymmetry of the signal SIN to be measured cancel each other out, and the vibration rectification error included in the output signal of the second low-pass filter 330 is reduced.

[0234] On the other hand, in the inspection mode of the physical quantity measuring device 2, the cutoff frequency of the second low-pass filter 330 is set lower than that in the normal operation mode, so that the vibration rectification error included in the signal output from the second low-pass filter 330 is emphasized. Therefore, by acquiring the output signal of the second low-pass filter 330 while changing the group delay amount of the first low-pass filter 310, information indicating the relationship between the group delay amount and the vibration rectification error is obtained. The period of the change in the vibration rectification error included in this information correlates with the period of the signal component due to the cantilever resonance included in the measured signal SIN. Therefore, according to the fifth embodiment, in the inspection mode of the physical quantity measuring device 2, it is possible to detect a signal component having periodicity included in the measured signal SIN without performing envelope processing. Also, for example, the physical quantity measuring device 2 or the signal processing device 400 calculates the cantilever resonance frequency of the physical quantity sensor 200 by using the information indicating the relationship between the group delay amount and the vibration rectification error, and calculates the cantilever resonance frequency based on the cantilever resonance frequency. Based on this, it can be determined whether the sensitivity of the physical quantity sensor 200 is within specifications.

[0235] 6. Variations The present invention is not limited to the present embodiment, and various modifications are possible within the scope of the present invention.

[0236] For example, in each of the above embodiments, the sensor module 1 has three physical quantity sensors 200, but the number of physical quantity sensors 200 included in the sensor module 1 may be one, two, or four or more.

[0237] In addition, in each of the above embodiments, the sensor module 1 including an acceleration sensor as the physical quantity sensor 200 has been exemplified, but the sensor module 1 may include sensors such as an angular velocity sensor, a pressure sensor, an optical sensor, etc. as the physical quantity sensor 200. The sensor module 1 may include two or more types of physical quantity sensors among various physical quantity sensors such as an acceleration sensor, an angular velocity sensor, a pressure sensor, an optical sensor, etc.

[0238] In addition, in each of the above embodiments, an element configured using quartz crystal has been given as an example of the physical quantity detection element 40 of the physical quantity sensor 200, but the physical quantity detection element 40 may be configured using a piezoelectric element other than quartz crystal, or may be a capacitance type MEMS element. MEMS is an abbreviation for Micro Electro Mechanical Systems.

[0239] The present invention is not limited to the present embodiment, and various modifications are possible within the scope of the present invention.

[0240] The above-described embodiment and modifications are merely examples, and the present invention is not limited to these. For example, the embodiments and modifications can be appropriately combined.

[0241] The present invention includes configurations that are substantially the same as the configurations described in the embodiments, for example, configurations with the same functions, methods, and results, or configurations with the same purpose and effects. The present invention also includes configurations in which non-essential parts of the configurations described in the embodiments are replaced. The present invention also includes configurations that achieve the same effects as the configurations described in the embodiments, or configurations that can achieve the same purpose. The present invention also includes configurations in which publicly known technology is added to the configurations described in the embodiments.

[0242] The following can be derived from the above-described embodiment and modifications.

[0243] One aspect of the signal processing method includes: a process to generate a process target signal which is a time-series signal based on a source signal which is a time-series signal output from an object; The method includes a vibration rectification error calculation process for calculating a plurality of vibration rectification errors by performing a product-sum operation a plurality of times with changing the shift amount of a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed.

[0244] In this signal processing method, the first signal based on the processing target signal generated based on the source signal and the second signal based on the processing target signal with a phase shifted signal both contain a periodic signal component generated by the object. Therefore, by performing a product-sum operation process on the first signal and the second signal, the signal components with ergodic properties such as noise are attenuated, while the signal components with periodicity strengthen or weaken each other depending on the phase difference between the first signal and the second signal. As a result, The multiple vibration rectification errors obtained by performing the product-sum operation multiple times with different shift amounts have different magnitudes depending on the phase difference between the first signal and the second signal and the period of the periodic signal component. Therefore, according to this signal processing method, it is possible to detect the periodic signal component contained in the signal output from the target without performing envelope processing.

[0245] In addition, according to this signal processing method, it is only necessary to acquire the source signal for a specified time once in order to calculate multiple vibration rectification errors, enabling high-speed calculation processing, and since environmental changes such as temperature changes in the short period of time in which the source signal is acquired are extremely small, calculation errors caused by environmental changes are reduced.

[0246] In one aspect of the signal processing method, The first signal may be a signal obtained by filtering the target signal.

[0247] According to this signal processing method, the noise components contained in the first signal are reduced by the filtering process, and therefore the detection accuracy of the signal components having periodicity is improved.

[0248] In one aspect of the signal processing method, The second signal may be a signal obtained by filtering a phase-shifted signal of the target signal.

[0249] According to this signal processing method, the noise components contained in the second signal are reduced by the filtering process, and therefore the detection accuracy of the signal components having periodicity is improved.

[0250] In one aspect of the signal processing method, The filtering may be a smoothing filtering.

[0251] According to this signal processing method, the smoothing filter process reduces high-frequency noise components contained in the first signal or the second signal, improving the detection accuracy of signal components having periodicity.

[0252] In one aspect of the signal processing method, The filtering may be band-limiting filtering.

[0253] According to this signal processing method, many signal components other than the periodic signal components contained in the first signal and the second signal are reduced by the band-limiting filter processing, thereby improving the detection accuracy of the periodic signal components.

[0254] In one aspect of the signal processing method, The first signal may be a signal obtained by removing or reducing a direct current component from the signal to be processed.

[0255] According to this signal processing method, each sample value of the first signal becomes small, so that the load of the product-sum operation of the first and second signals is reduced and the operation accuracy is improved.

[0256] In one aspect of the signal processing method, The second signal may be a signal obtained by shifting the phase of the processing target signal and removing or reducing a direct current component from the signal.

[0257] According to this signal processing method, each sample value of the second signal becomes small, so that the load of the product-sum operation of the first signal and the second signal is reduced and the operation accuracy is improved.

[0258] In one aspect of the signal processing method, The first signal may be a signal obtained by multiplying the processing target signal by a window function.

[0259] According to this signal processing method, the discontinuity between the first sample value and the last sample value of the first signal is mitigated, thereby improving the detection accuracy of the periodic signal component.

[0260] In one aspect of the signal processing method, The second signal may be a signal obtained by shifting the phase of the target signal and applying a window function to the shifted signal.

[0261] According to this signal processing method, the discontinuity between the first sample value and the last sample value of the second signal is mitigated, thereby improving the detection accuracy of the signal component having periodicity.

[0262] In one aspect of the signal processing method, The number of additions in the product-sum calculation process may be greater than a value obtained by dividing a sampling frequency of the source signal by a resonance frequency of the object.

[0263] According to this signal processing method, in the product-sum operation, the signal components of the resonant frequency of the object contained in the first signal and the second signal are integrated for one period or more, so that the signal components of the resonant frequency are effectively detected.

[0264] One aspect of the signal processing method includes: The method may further include a resonance frequency calculation step of calculating a resonance frequency of the object based on the plurality of vibration rectification errors.

[0265] According to this signal processing method, the resonant frequency of the object is calculated, enabling the user to perform various analyses based on the resonant frequency.

[0266] One aspect of the signal processing method includes: The method may further include a determination step of determining whether the calculated resonance frequency is correct or not based on a difference between a maximum value and a minimum value of the plurality of vibration rectification errors.

[0267] In this signal processing method, the greater the resonance excited in the object, the greater the difference between the maximum and minimum values ​​of the calculated vibration rectification errors, and the correctness of the calculated resonance frequency can be determined based on the difference. Then, according to this signal processing method, for example, the user can perform various analyses based on the resonance frequency only when the calculated resonance frequency is determined to be correct.

[0268] One aspect of the signal processing device includes: a processing target signal generating circuit that generates a processing target signal, which is a time-series signal, based on a source signal, which is a time-series signal output from an object; The vibration rectification error calculation circuit includes a vibration rectification error calculation circuit that performs a product-sum operation on a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed multiple times while changing the shift amount, thereby generating multiple vibration rectification errors.

[0269] In this signal processing device, the first signal based on the processing target signal generated based on the source signal and the second signal based on the phase-shifted signal of the processing target signal both contain a periodic signal component output from the target object. Therefore, by performing a product-sum operation process on the first signal and the second signal, the signal components having ergodic properties such as noise are attenuated, while the periodic signal components are attenuated. The signal components strengthen or weaken each other depending on the phase difference between the first signal and the second signal. As a result, the multiple vibration rectification errors obtained by performing the product-sum operation multiple times with different shift amounts have different magnitudes depending on the phase difference between the first signal and the second signal and the period of the periodic signal component. Therefore, this signal processing device can detect the periodic signal component contained in the signal output from the target without performing envelope processing.

[0270] In addition, with this signal processing device, it is only necessary to acquire the source signal for a specified time once in order to calculate multiple vibration rectification errors, enabling high-speed calculation processing, and since environmental changes such as temperature changes in the short period of time in which the source signal is acquired are extremely small, calculation errors caused by environmental changes are reduced.

[0271] One aspect of the physical quantity measuring device is a reference signal generating circuit that outputs a reference signal; a frequency delta-sigma modulation circuit that uses a signal under test to frequency delta-sigma modulate the reference signal to generate a frequency delta-sigma modulated signal; a first filter that operates in synchronization with the signal under test and has a variable group delay; a second filter that operates in synchronization with the reference signal; the first filter is provided on a signal path from an output of the frequency delta-sigma modulation circuit to an input of the second filter; The analyzer has a first operation mode for measuring a frequency ratio between the signal under test and the reference signal, and a second operation mode in which the cutoff frequency of the second filter is lower than that in the first operation mode.

[0272] In this physical quantity measuring device, in the first operation mode, the first filter operates in synchronization with the signal under measurement, and the second filter operates in synchronization with a reference signal different from the signal under measurement, so that nonlinearity occurs in the relationship between the frequency delta-sigma modulated signal and the output signal of the second filter. The vibration rectification error caused by this nonlinearity changes depending on the group delay of the first filter. Therefore, according to this physical quantity measuring device, by setting the group delay of the first filter to an appropriate value in the first operation mode, the vibration rectification error caused by this nonlinearity and the vibration rectification error caused by the asymmetry of the signal under measurement cancel each other out, and the vibration rectification error included in the output signal of the second filter is reduced.

[0273] In addition, in the second operation mode of the physical quantity measuring device, the cutoff frequency of the second filter is set lower than that of the first operation mode, thereby emphasizing the vibration rectification error contained in the signal output from the second filter. Therefore, by acquiring the output signal of the second filter while changing the group delay amount of the first filter, information indicating the relationship between the group delay amount and the vibration rectification error is obtained. The period of change of the vibration rectification error contained in this information is correlated with the period of the signal component having periodicity contained in the signal to be measured. Therefore, according to this physical quantity measuring device, in the second operation mode, it is possible to detect the signal component having periodicity contained in the signal to be measured without performing envelope processing.

[0274] One aspect of the sensor module is One aspect of the physical quantity measuring device; A physical quantity sensor, The measured signal is a signal based on an output signal of the physical quantity sensor.

[0275] According to this sensor module, by being equipped with a physical quantity measuring device, in the first operating mode, highly accurate measurement values ​​of physical quantities with reduced vibration rectification errors can be obtained, and in the second operating mode, signal components of the resonant frequency generated by structural resonance of the physical quantity sensor can be detected. [Explanation of symbols]

[0276] 1...sensor module, 2...physical quantity measuring device, 5...substrate portion, 10...base portion, 12...joint portion, 13...movable portion, 30a, 30b...support portion, 34...package joint portion, 36a, 36b...joint portion, 38a, 38b...extension portion, 40...physical quantity detection element, 50, 52, 54, 56...weight, 62...joint member, 101...container, 102...lid, 103...screw hole, 104...fixing protrusion, 111...side wall, 112...bottom wall, 115...circuit board, 115f...first surface, 115r...second surface, 116...connector 121: opening, 122: inner surface, 123: opening surface, 125: second base, 127: first base, 129: protrusion, 130: fixing member, 133, 134: constricted portion, 141: sealing member, 172: screw, 174: female screw, 176: through hole, 200, 200X, 200Y, 200Z: physical quantity sensor, 201X, 201Y, 201Z: oscillator circuit, 202, 202X, 202Y, 202Z: frequency ratio measuring circuit, 203: reference signal generating circuit, 210: microcontroller 1. A digital signal processing unit, 220... a memory unit, 230... an interface circuit, 300... a frequency delta-sigma modulation circuit, 301... a counter, 302... a latch circuit, 303... a latch circuit, 304... a subtractor, 310... a first low-pass filter, 311... a delay element, 312... an integrator, 313... an integrator, 314... a decimator, 315... a delay element, 316... a differentiator, 317... a delay element, 318... a differentiator, 320... a latch circuit, 330... a second low-pass filter, 331... an integrator, 332 ...delay element, 333...differentiator, 334...decimator, 400...signal processing device, 410...processing circuit, 411...source signal acquisition circuit, 412...processing target signal generation circuit, 413...vibration rectification error calculation circuit, 414...resonance frequency calculation circuit, 415...determination circuit, 420...storage circuit, 421...signal processing program, 422...processing target signal, 423...vibration rectification error information, 424...resonance frequency, 425...determination result, 430...operation unit, 440...display unit, 450...sound output unit, 460...communication unit

Claims

1. a process to generate a process target signal which is a time-series signal based on a source signal which is a time-series signal output from an object; a vibration rectification error calculation process for calculating a plurality of vibration rectification errors by performing a product-sum operation a plurality of times with changing shift amounts on a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed.

2. In claim 1, A signal processing method, wherein the first signal is a signal obtained by filtering the signal to be processed.

3. In claim 1 or 2, A signal processing method, wherein the second signal is a signal obtained by filtering a phase-shifted signal of the processing target signal.

4. In claim 2 or 3, A signal processing method, wherein the filtering is a smoothing filtering.

5. In claim 2 or 3, A signal processing method, wherein the filtering is band-limiting filtering.

6. In any one of claims 1 to 5, A signal processing method, wherein the first signal is a signal obtained by removing or reducing a DC component from the processing target signal.

7. In any one of claims 1 to 5, A signal processing method, wherein the second signal is a signal obtained by removing or reducing a DC component from a signal obtained by shifting the phase of the processing target signal.

8. In any one of claims 1 to 7, A signal processing method, wherein the first signal is a signal obtained by multiplying the processing target signal by a window function.

9. In any one of claims 1 to 7, A signal processing method, wherein the second signal is a signal obtained by shifting the phase of the signal to be processed and multiplying the signal by a window function.

10. In any one of claims 1 to 9, A signal processing method, wherein the number of additions in the product-sum operation is greater than a value obtained by dividing a sampling frequency of the source signal by a resonant frequency of the object.

11. In any one of claims 1 to 10, A signal processing method comprising: a resonance frequency calculation step of calculating a resonance frequency of the object based on the plurality of vibration rectification errors.

12. In claim 11, The signal processing method includes a determination step of determining whether the calculated resonance frequency is correct or not based on a difference between a maximum value and a minimum value of the plurality of vibration rectification errors.

13. a processing target signal generating circuit that generates a processing target signal, which is a time-series signal, based on a source signal, which is a time-series signal output from an object; a vibration rectification error calculation circuit that performs a product-sum operation on a first signal based on the signal to be processed and a second signal based on a signal obtained by shifting the phase of the signal to be processed multiple times while changing the shift amount, thereby generating multiple vibration rectification errors.

14. a reference signal generating circuit that outputs a reference signal; a frequency delta-sigma modulation circuit that uses a signal under test to frequency delta-sigma modulate the reference signal to generate a frequency delta-sigma modulated signal; a first filter that operates in synchronization with the signal under test and has a variable group delay; a second filter that operates in synchronization with the reference signal; the first filter is provided on a signal path from an output of the frequency delta-sigma modulation circuit to an input of the second filter; a first operation mode for measuring a frequency ratio between the signal under measurement and the reference signal, and a second operation mode in which a cutoff frequency of the second filter is lower than that of the first operation mode.

15. The physical quantity measuring device according to claim 14 ; A physical quantity sensor, The measured signal is a signal based on an output signal of the physical quantity sensor.

Images