JP2007155618A - Rotation angle detection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a rotation angle detection device capable of maintaining properly detection accuracy of a rotator to be detected, even when a deviation is generated in an amplitude value or an offset value of an output signal when detecting a rotation angle of the rotator to be detected based on an output signal by a magnetic detection element. <P>SOLUTION: This rotation angle detection device has a set of three Hall elements 111a-111c for detecting a change of a magnetic vector MV emitted from a polarized rotor 200 rotating following rotation of a crankshaft 300 as sine wave signals A, B having each phase shifted by 90 degrees. The set of the three Hall elements 111a-111c is arranged slantingly respectively. The rotation angle detection device calculates the rotation angle θ of the crankshaft 300 based on an operation expression 'θ=tan<SP>-1</SP>(A/B)' relative to each of the set of the three Hall elements 111a-111c, and then operates the mean value thereof, while keeping a phase relation among three calculated signals. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a rotation angle detection device that detects a rotation angle of various detected rotating bodies such as a crankshaft using a magnetic sensing element.

  Conventionally, as this type of rotation angle detection device, for example, a rotation angle detection device described in Patent Document 1 is known. FIG. 8 shows an outline of a conventional general rotation angle detection device that detects the rotation angle of the crankshaft using a magnetic sensing element, including the rotation angle detection device described in Patent Document 1.

  As shown in FIGS. 8A and 8B, this apparatus adopts a horizontal Hall element that is generally used as a magnetic sensing element, and the two horizontal Hall elements 11a and 11b are IC chips. 1a and 1b are formed by resin molding. These two IC chips 1a and 1b are disposed at an angle of 90 degrees with each other, and are disposed opposite to a disk-shaped magnetized rotor RT in which N and S poles are separately magnetized. That is, the IC chips 1a and 1b detect the angle of the magnetic vector MV applied from the magnetized rotor RT to the Hall elements 11a and 11b and detect the crankshaft CS that is also the central axis of the magnetized rotor RT. The rotation angle is obtained.

  FIG. 9 is a block diagram showing the internal circuits of such IC chips 1a and 1b. Hereinafter, the electrical configuration and the operation based on the configuration will be further described with reference to FIG. explain. Such a rotation angle detection device can be realized by various electrical configurations. Here, a circuit that is currently practically used as an internal circuit of the IC chips 1a and 1b will be described.

  As shown in FIG. 9, in the IC chips 1a and 1b, the Hall elements 11a and 11b are respectively driven by a constant current or a constant voltage from the drive circuits 10a and 10b. Here, as shown in FIG. 8B, these Hall elements 11a and 11b are provided such that their magnetic sensing surfaces are orthogonal to each other. The direction (angle) of the magnetic vector MV applied to the Hall elements 11a and 11b directly reflects the rotation angle of the magnetized rotor RT (crankshaft CS). With such a configuration, from the Hall elements 11a and 11b, the sin waveform voltage (Hall voltage) A, which is shifted in phase by 90 ° as shown in FIG. B will be taken as their output. The output signals A and B are sequentially taken into the amplifier circuits 12a and 12b and further to the A / D converters 13a and 13b in the IC chips 1a and 1b. Then, the amplifier circuits 12a and 12b respectively amplify as desired, and the A / D converters 13a and 13b respectively quantize them with a required resolution and then convert them into discrete values (digital values). The two output signals A and B discretized in this way are taken out from the output terminals of the IC chips 1a and 1b as angle information of the crankshaft CS, respectively, and then, for example, are used for controlling the fuel injection of the in-vehicle engine. It is taken into the angle calculation unit 24 in the control device 20. Normally, the angle calculation unit 24 digitally corrects (angle calculation) a signal that linearly changes with respect to the rotation of the crankshaft CS, so that the rotation of the crankshaft CS is based on this signal. An angle is detected.

Incidentally, in this angle calculation unit 24, first, when the output signal A is sin θ and the other output signal B is cos θ when the rotation angle of the magnetized rotor RT (crankshaft CS) is θ, The rotation angle θ is calculated as “θ = tan ⁻¹ (A / B)”. However, the signal waveform calculated in this way is such that the value falls to the minimum value at each angle position of “θ = 90 °” and “θ = 270 °” as shown in FIG. It is difficult to uniquely derive the rotation angle of the crankshaft CS only from the calculation result. In view of this, the angle calculation unit 24 next selects, for example, “θ = of the signal waveforms shown in FIG. 10B based on the values of the two discrete output signals A and B and their magnitude relationship. A predetermined offset value is added to or subtracted from an angle range of “0 ° to 90 °”, an angle range of “θ = 90 ° to 270 °”, and an angle range of “θ = 270 ° to 360 °”. This is converted into a signal waveform shown in FIG. As a result, a linear signal waveform having a specific value with respect to the rotation angle θ for each rotation (0 ° to 360 °) of the crankshaft CS can be obtained. The rotation angle θ of CS can be detected finely.
JP 2004-340740 A

  By the way, the rotation angle detection apparatus as described above is not necessarily in a desirable environment. For example, when a disturbance magnetic field is generated due to operation of a peripheral device, the magnetic vector MV There is concern that the size will change. That is, when the magnitude of the magnetic vector MV changes in this way, the magnitude of each vector component acting on the Hall elements 11a and 11b of the magnetic vector MV also changes. For this reason, a deviation occurs in the amplitude values of the output signals A and B by the Hall elements 11a and 11b, and as a result, such deviation of the amplitude values of the output signals A and B appears as an angular error of the detected rotation angle θ. It becomes like this.

On the other hand, although the above-described angle calculation (“θ = tan ⁻¹ (A / B)”) has been confirmed by the inventor to have a certain effect of reducing such an angle error, it is more reliable. Development of a rotation angle detection device capable of obtaining highly reliable angle information is desired.

  It should be noted that the rotation angle detection device using another magnetic sensing element such as a magnetoresistive element is not limited to the rotation angle detection device using the Hall element, but the amplitude value or offset of the output signal from the magnetic sensing element is not limited. The tendency for the detection accuracy to decrease when a deviation occurs is generally the same.

  The present invention has been made in view of such circumstances, and its purpose is to detect a deviation in the amplitude value or offset value of the output signal when detecting the rotation angle of the detected rotating body based on the output signal from the magnetic sensing element. It is an object of the present invention to provide a rotation angle detection device that can maintain the detection accuracy of a detected rotating body more favorably even if it occurs.

In order to achieve such an object, in the rotation angle detection device according to claim 1, the change of the magnetic vector emitted from the rotating magnet with the rotation of the rotation shaft to be detected as the rotation angle is phased by 90 degrees. A sensor unit having two magnetic sensing elements arranged to sense as a shifted sine wave signal, the rotation angle of the rotation shaft is “θ”, and one output signal A of the magnetic sensing element is “A = sin θ ”, When the other output signal B of the magnetic sensing element is“ B = cos θ ”, the output signals from the two magnetic sensing elements are based on an arithmetic expression of“ θ = tan ⁻¹ (A / B) ”. A rotation angle detector that detects a rotation angle of the rotation shaft based on a signal extracted from the signal processing unit, and a signal processing unit that converts the signal into a linear signal that changes linearly with respect to the rotation of the rotation shaft. , The two When the magnetic sensing element is a magnetic sensing element pair, the sensor unit has a plurality of magnetic sensing element pairs, and the signal processing unit outputs the linear signal separately from the output signals from the plurality of magnetic sensing element pairs. In addition, one signal calculated by performing an operation capable of absorbing an angular error on the linear signals is output as the rotation angle information of the rotating shaft.

As described above, the calculation formula “θ = tan ⁻¹ (A / B)” reduces the angular error that appears when the amplitude values and offset values of the output signals A and B from the magnetic sensing element are shifted. It has a certain effect. In this regard, in the above-described configuration, such “θ = tan ⁻¹ (A / B)” is calculated separately from the output signals from the plurality of magnetic sensing element pairs, and the angle with respect to the plurality of linear signals obtained thereby is calculated. Since the operation capable of absorbing the error is performed, more reliable angle information can be obtained.

Further, the inventor has found that the effect of reducing the angle error of the arithmetic expression “θ = tan ⁻¹ (A / B)” varies depending on the phase based on “θ” on the left side of the expression. I found it. In this regard, in the rotation angle detection device according to claim 1, in the rotation angle detection device according to claim 2, first, the plurality of magnetic sensing element pairs are caused to change in a magnetic vector as the rotation shaft rotates. Are arranged to be detected as output signals having different phases. For each of the output signals having different phases, an arithmetic expression of “θ = tan ⁻¹ (A / B)” was implemented. On this basis, an operation capable of absorbing the angle error is applied to such linear signals while maintaining their phase relationship. For this reason, it becomes possible to obtain the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” over the wide angle range.

In the rotation angle detection device according to claim 1 or 2, as an operation capable of absorbing the angle error, for example, as with the angle detection device according to claim 3,
An average value calculation for obtaining an average value of the linear signals obtained separately from output signals of the plurality of magnetic sensing element pairs.
Alternatively, as in the rotation angle detection device according to claim 4,
A majority operation for selectively obtaining one of the linear signals having a large number of matches through comparison of the linear signals obtained separately from the output signals of the plurality of magnetic sensing element pairs.
It is conceivable to employ an operation such as, etc. In any of these operations, the detection accuracy for the detected rotating body (rotating shaft) can be more suitably maintained. However, in the rotation angle detection device according to claim 2, when the rotation angle detection device according to claim 3 is employed, the calculation formula of “θ = tan ⁻¹ (A / B)” is particularly used. The effect of reducing the angle error can be obtained more suitably over the wide angle range.

  Further, the magnet that rotates with the rotation of the rotation shaft that is the detection target of the rotation angle and the magnetic sensing element are basically arbitrary. However, as in the rotation angle detection device according to claim 5, the magnet is a disc-shaped magnetized rotor formed integrally with the rotating shaft in such a manner that the N pole and the S pole are separated and magnetized. A vertical Hall element in which the sensor unit and the signal processing unit are integrated as a single semiconductor chip, and the magnetic sensing element senses a magnetic vector parallel to the semiconductor substrate surface based on the Hall effect. In this way, integration (miniaturization) as the rotation angle detection device can be easily achieved. In addition, through the semiconductor process, it is possible to set the positional relationship between the Hall elements and the Hall element pairs more accurately.

  In the rotation angle detection device according to any one of claims 1 to 5, as in the rotation angle detection device according to claim 6, the sensor unit includes the plurality of magnetic sensing element pairs. It is practically desirable to have two magnetic sensing element pairs, or three magnetic sensing element pairs as in the rotation angle detection device according to claim 7.

Hereinafter, an embodiment of a rotation angle detection device according to the present invention will be described in detail with reference to FIGS.
First, the configuration of the rotation angle detection device will be described in detail with reference to FIG. FIG. 1A is a side view showing a schematic configuration of this apparatus, and FIG. 1B is a plan view showing the schematic configuration of this apparatus.

As shown in FIG. 1A, the rotation angle detection device according to this embodiment is also
A sine wave signal whose phase is shifted by 90 degrees from the change in the magnetic vector MV emitted from the magnetized rotor (magnet) 200 that rotates as the crankshaft 300 that is the target of rotation angle detection (FIG. 10 ( a sensor part 111 having two Hall elements to be detected as a).
When the rotation angle of the crankshaft 300 is “θ”, one output signal A of the Hall element is “A = sin θ”, and the other output signal B of the Hall element is “B = cos θ”, the above two A signal processing unit that converts an output signal from the Hall element into one signal (see FIG. 10C) based on an arithmetic expression of “θ = tan ⁻¹ (A / B)”.
And so on. However, in this embodiment, a so-called vertical Hall element that senses a magnetic vector MV parallel to the surface of the semiconductor substrate based on the Hall effect is employed as the Hall element, whereby the sensor unit 111 and the signal processing unit are 1 An integrated circuit is formed as one IC chip 100.

By the way, as described above, such a rotation angle detection device is not necessarily in a desirable environment. For example, when a disturbance magnetic field is generated due to the operation of a peripheral device or the like, An angular error may occur in the detected rotation angle. On the other hand, the arithmetic expression “θ = tan ⁻¹ (A / B)” also has a certain effect of reducing such an angle error as described above.

FIG. 2 and FIG. 3 show the effect of reducing the angle error by the calculation formula “θ = tan ⁻¹ (A / B)” for each “θ” on the left side of the calculation formula. The same effect will be described with reference to FIGS. 2A to 2C show the angle error (%) appearing in the linear signal when a deviation of “1%” occurs in the amplitude values of the output signals A and B by the Hall element. It is a graph. FIGS. 3A to 3C are graphs showing angle errors appearing in the linear signal when a deviation of “1%” occurs in the offset values of the output signals A and B by the Hall element. .

That is, if the rotation angle θ of the crankshaft 300 is obtained as the linear signal based on the arithmetic expression “θ = tan ⁻¹ (A / B)”, it is shown in FIGS. As can be seen, even when the amplitude values of the output signals A and B by the Hall element are shifted by “1%”, the angular error is reduced to “0.1%” or less. Similarly, as shown in FIGS. 2A to 2C, even when a deviation of “1%” occurs in the offset values of the output signals A and B by the Hall element, the angular error is as follows. Reduced to "less than 0.2%" or less.

  However, as shown in FIGS. 2 (a) and 2 (b), the inventor has a case where a deviation occurs only in the amplitude value of one of the output signals A and B by the Hall element, or FIGS. As shown in (c), when there is a deviation in the offset value, the effect of reducing the angle error varies depending on the phase with reference to “θ” on the left side in the above arithmetic expression. I found it. Therefore, in the rotation angle detection device according to this embodiment, when the two Hall elements are Hall element pairs, the sensor unit 111 is configured so that the crankshaft as shown in FIG. Three Hall element pairs 111a to 111c are arranged so as to sense a change in the magnetic vector MV accompanying the rotation of 300 as an output signal having a different phase. In addition, the signal processing unit obtains the linear signal separately from the output signals from the Hall element pairs 111a to 111c, and performs an average value calculation to obtain the average value while maintaining the phase relationship of the linear signals. I am doing so.

In such a configuration, the phase of the linear signal obtained separately from the output signals from the plurality of Hall element pairs 111a to 111c is “120 °” according to the arrangement mode of the plurality of Hall element pairs 111a to 111c. Only they will be different. For this reason, the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” also acts on such a linear signal with a phase shift of “120 °”. Then, since the average value calculation for obtaining the average value is performed while maintaining the phase relationship of such linear signals, the angle according to the calculation formula of “θ = tan ⁻¹ (A / B)” is used. The effect of reducing the error is smoothed, and the same effect can be preferably obtained over the wide angular range.

Here, FIG. 4A is a graph showing how the effect of reducing the angle error acts on the three linear signals with a phase shift of “120 °”. FIG. 4A corresponds to FIG. 2A, and assumes a case where a deviation of “1%” occurs only in the amplitude value of the output signal A by the Hall element. . In FIG. 4A, the solid line corresponds to the signal obtained from the Hall element pair 111a, the two-dot chain line corresponds to the signal obtained from the Hall element pair 111b, and the one-dot chain line represents the Hall element. It corresponds to the signal obtained from the pair 111c. On the other hand, FIG. 4B shows the angular error appearing in the averaged signal as a result of averaging the linear signals for each detected rotation angle θ. As is apparent from FIGS. 4A and 4B, according to the configuration as the rotation angle detection device, the angle error is calculated based on the arithmetic expression “θ = tan ⁻¹ (A / B)”. As a result of suitably obtaining the reduction effect over the wide angular range, it can be seen that the angular error appearing in the linear signal is reduced to about "less than 0.07%".

  FIG. 5 is a block diagram showing an example of an internal circuit for realizing such an IC chip 100. Hereinafter, the electrical configuration and the operation based on the configuration will be further described with reference to FIG.

As shown in FIG. 5, in the IC chip 100, the Hall element pair 111a is composed of two Hall elements that are respectively driven by constant currents or constant voltages from the drive circuits 110a and 110b. As described above, the Hall element pair 111a is provided in such a manner that the sensing surfaces of the two Hall elements are orthogonal to each other as shown in FIG. Further, the direction (angle) of the magnetic vector MV applied to the Hall element pair 111a directly reflects the rotation angle θ of the crankshaft 300. With such a configuration, from the Hall element pair 111a, the sin waveform voltages (Hall voltages) A and B shifted in phase by 90 ° as shown in FIG. Will be extracted as their output. In the IC chip 100, these output signals A and B are sequentially taken into the amplifier circuits 112a and 112b and further the A / D converters 113a and 113b constituting part of the signal processing unit. Then, the amplifiers 112a and 112b amplify the signals as desired, and the A / D converters 113a and 113b respectively quantize them with a required resolution and then convert them into discrete values (digital values). Then, the two output signals A and B discretized in this way are then taken into the angle calculation unit 114ab, respectively, and the angle calculation unit 114ab uses the following equation: “θ = tan ⁻¹ (A / B)”. Then, it is converted into a linear signal X shown in FIG. As will be described later, the linear signal X is taken into the average value calculation unit 116, and the average value calculation unit 116 calculates an average value with linear signals Y and Z described later.

On the other hand, the Hall element pair 111b and the Hall element pair 111c are also composed of two vertical Hall elements respectively driven by constant currents or constant voltages from the drive circuits 110c and 110d or the drive circuits 110e and 110f. The sensing surfaces of the vertical Hall elements are provided so as to be orthogonal. Also, output signals A and B from the Hall element pair 111b and output signals A and B from the Hall element pair 111c are respectively taken into the signal processing unit,
The amplifier circuit 112c or 112d or the amplifier circuit 112e or 112f is amplified as desired.
The A / D converters 113c and 113d or the A / D converters 113e and 113f are each quantized with a required resolution and then converted into discrete values (digital values).
The linear signals Y and Z that change linearly with respect to the rotation of the crankshaft 300 under the arithmetic expression “θ = tan ⁻¹ (A / B)” in the angle calculation unit 114cd or the angle calculation unit 114ef. To be converted to
The process is sequentially performed in a similar manner to the Hall element pair 111a. However, as shown in FIGS. 6D and 6H in comparison with FIG. 6A, these Hall element pairs 111b and 111c are different from the Hall element pair 111a in the IC chip 100. Each is mounted (arranged) by tilting clockwise by “120 °” and “240 °”. Therefore, the output signals from the Hall element pairs 111b and 111c are used to rotate the crankshaft 300 as shown in FIGS. 6 (e) and 6 (i) in comparison with FIG. 6 (b). On the other hand, the phase changes with a delay of “120 °” and “240 °” from the output signal from the Hall element pair 111a. Furthermore, as shown in FIGS. 6 (f) and 6 (j) in comparison with FIG. 6 (c), the linear signals Y and Z extracted from the angle calculators 114cd and 114ef are also converted into the angle calculators. The phase relationship is delayed by “120 °” and “240 °” from the linear signal X extracted from 114ab. Therefore, by obtaining the average value of the three linear signals X to Z while maintaining the phase relationship, the angle according to the arithmetic expression of “θ = tan ⁻¹ (A / B)” is used. The error reduction effect can be smoothed.

  Incidentally, in this embodiment, as shown in FIG. 5, prior to such average value calculation, among the three linear signals X to Z, linear signals Y and Z are output value adjustment unit 115 cd, 115ef is first taken in. The linear signals Y and Z are signals that have substantially the same output value as the linear signal X with respect to the rotation angle of the crankshaft 300 in the output value adjusting units 115cd and 115ef. 6 (g) and (k) are converted into linear signals Y ′ and Z ′, respectively. However, at this time, the phase relationship of the linear signals X to Z (phase shift by “120 °”) is maintained, and this signal conversion (output value adjustment) is performed only by adding or subtracting a predetermined offset value. Then, the three linear signals X, Y ′, and Z ′ obtained in this way are taken into the average value calculation unit 116, and the average value calculation unit 116 performs average value calculation to obtain the average value.

As described above, according to the rotation angle detection device of this embodiment, the following excellent effects can be obtained.
(1) The sensor unit 111 includes three Hall element pairs 111a to 111c arranged to sense changes in the magnetic vector MV accompanying rotation of the crankshaft 300 as output signals having different phases. In addition, the signal processing unit obtains linear signals X to Z separately from the output signals by the plurality of Hall element pairs 111a to 111c, and obtains an average value thereof while maintaining the phase relationship of the linear signals. The operation was performed. For this reason, the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” is smoothed, and the same effect can be suitably obtained over the wide angle range.

  (2) A so-called vertical Hall element that senses a magnetic vector MV parallel to the surface of the semiconductor substrate based on the Hall effect is adopted as the Hall element, and the sensor unit 111 and the signal processing unit are integrated into a single IC chip 100. did. For this reason, size reduction as the rotation angle detection device can be expected. In addition, Hall elements arranged with a 90 degree relationship with each other and Hall element pairs 111a to 111b arranged with a 120 degree relationship can be accurately formed in the semiconductor substrate.

In addition, the said embodiment can also be changed and implemented as follows.
The configuration of the internal circuit for realizing the IC chip 100 of the above embodiment is arbitrary. For example, in the above embodiment, the output values of the linear signals Y and Z are adjusted to the linear signals Y ′ and Z ′, respectively, with the linear signal X as a reference. However, the output values of the linear signals X and Z may be adjusted based on the linear signal Y, or the output values of the linear signals X and Y may be adjusted based on the linear signal Z. . Further, the output values of all the linear signals X to Z may be adjusted. Alternatively, after calculating the average value of the three linear signals X to Y, the output value of the averaged signal may be adjusted. Further, in order to obtain an effect of smoothing the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)”, such output value adjustment is not necessarily performed.

  In the above embodiment, the Hall element pairs 111a to 111c are provided. However, as shown in FIG. 7A, any two or more Hall element pairs may be disposed so as to be inclined. FIG. 7A assumes a case where the six Hall element pairs HC11 to HC16 are arranged so as to be inclined by 60 °. That is, in this case, the phases of the linear signals obtained separately for the Hall element pairs HC11 to HC16 are different from each other by 60 °. Therefore, even in this case, if the average value is obtained while maintaining the phase relationship between the linear signals, the effect (1) can be obtained.

  In the case where three or more Hall element pairs are provided, the angular difference of each Hall element pair may be different. For example, as shown in FIG. 7B, the angle difference between the Hall element pair HC21 and HC22 is set to “45 °”, and the angle difference between the Hall element pair HC22 and HC23 is set to “105 °”. It may be.

  As shown in FIG. 7C, the Hall element pairs HC31 and HC32 may be arranged in parallel as long as the phase of the output signal is digitally corrected to be different from each other. The Hall element pair may be any element as long as it is arranged to detect a change in the magnetic vector MV accompanying the rotation of the crankshaft 300 as an output signal having a different phase.

In the mean value calculation by the mean value calculation unit 116, the linear signal taken into the average value calculation unit 116 may be weighted. At this time, if the linear signal is weighted according to the rotation angle of the crankshaft 300, the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” can be obtained. It becomes possible to obtain more suitably over the wide angle range.

Instead of the average value calculation unit 116, a majority decision calculation unit that selectively outputs one of the linear signals having a large number of matches through comparison of the captured linear signals may be employed. Even in this case, the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” is smoothed, and the angle error of the linear signal can be absorbed.

The arithmetic expression “θ = tan ⁻¹ (A / B)” has a certain effect of reducing the angle error that appears when the amplitude values and offset values of the output signals A and B by the Hall element are shifted. is there. Therefore, an operation capable of calculating such “θ = tan ⁻¹ (A / B)” separately from the output signals from the plurality of Hall element pairs and absorbing the angle error with respect to the plurality of linear signals obtained thereby. As long as it is applied, at least more reliable angle information can be obtained.

  As the Hall element, a lateral Hall element that senses a magnetic vector perpendicular to the semiconductor substrate surface based on the Hall effect may be employed. However, in this case, integration as a rotation angle detection device becomes difficult.

The number of the Hall element pairs may be plural, and the number is arbitrary.
A magnetoresistive element (MRE) may be employed instead of the Hall element. In short, any magnetic sensing element that senses changes in the magnetic vector may be used.

-Although the magnetized rotor 200 was employ | adopted as a magnet, what is necessary is just a magnet rotated with rotation of the rotating shaft made into the detection object of a rotation angle. An electromagnet may also be used.
-In addition to the rotation angle of the crankshaft, the rotation angle of various rotating bodies such as the opening amount of the throttle valve may be detected. Further, an object other than the rotating body can be adopted as the detected rotating body as long as the displacement amount can be converted into the rotation angle.

(A) is a side view which shows typically the side structure including the relationship with a magnet about one Embodiment of the rotation angle detection apparatus concerning this invention. (B) is a top view which shows typically the planar structure including the relationship with a magnet about one Embodiment of the rotation angle detection apparatus concerning this invention. (A) shows the effect of reducing the angle error by the calculation formula of “θ = tan ⁻¹ (A / B)” when a deviation of “1%” occurs in the amplitude value of the output signal A by the Hall element. The graph shown for every rotation angle of the axis. (B) shows the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” when a deviation of “1%” occurs in the amplitude value of the output signal B by the Hall element. The graph shown for every rotation angle of the axis. (C) shows a reduction in angular error by an arithmetic expression of “θ = tan ⁻¹ (A / B)” when a deviation of “1%” occurs in the amplitude values of the output signals A and B by the Hall elements. The graph which showed the effect for every rotation angle of the crankshaft. (A) shows the effect of reducing the angle error by the arithmetic expression of “θ = tan ⁻¹ (A / B)” when the offset value of the output signal A by the Hall element is shifted by “1%”. The graph shown for every rotation angle of the axis. (B) shows the effect of reducing the angle error by the calculation formula of “θ = tan ⁻¹ (A / B)” when a deviation of “1%” occurs in the offset value of the output signal B by the Hall element. The graph shown for every rotation angle of the axis. (C) shows a reduction in angular error by an arithmetic expression of “θ = tan ⁻¹ (A / B)” when a deviation of “1%” occurs in the offset values of the output signals A and B by the Hall elements. The graph which showed the effect for every rotation angle of the crankshaft. (A) shows how the effect of reducing the angle error by the arithmetic expression “θ = tan ⁻¹ (A / B)” acts on three linear signals with a phase shift of “120 °”. Graph. (B) is a graph showing the angle error appearing in the averaged signal as a result of averaging three linear signals for each “θ”. The block diagram which shows the internal circuit of the rotation angle detection apparatus of the embodiment. (A), (d), (h) is a top view which shows typically the arrangement | positioning aspect of a Hall element pair. (B), (e), (i) is a graph which shows the output signal (Hall voltage) by a Hall element pair for every rotation angle of a crankshaft. (C), (f), (j) is a graph which shows the linear signal obtained separately for the output signal by a Hall element pair for every rotation angle of a crankshaft. (G), (k) is a graph which shows the conversion mode of the linear signal by an output value adjustment part. (A)-(c) is a top view which shows other examples, such as the arrangement | positioning aspect of a Hall element pair. (A) is a side view which shows typically the side structure of the conventional rotation angle detection apparatus. (B) is a top view which shows typically the planar structure of the conventional rotation angle detection apparatus. The block diagram which shows the internal circuit of the conventional rotation angle detection apparatus. (A) is a graph which shows the output signal by the conventional Hall element pair for every rotation angle of a crankshaft. (B) is a graph showing a signal obtained when the calculation of “θ = tan ⁻¹ (A / B)” is performed on the output signal from the conventional Hall element pair for each rotation angle of the crankshaft. (C) is a graph which shows the signal finally output from the conventional rotation angle detection apparatus for every rotation angle of a crankshaft.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... IC chip, 110a-110f ... Drive circuit, 111a-111c ... Hall element pair, 112a-112f ... Amplifier circuit, 113a-113f ... A / D converter, 114ab, 114cd, 114ef ... Angle calculation part, 115cd, 115ef ... an output value adjusting unit, 116 ... an average value calculating unit, 200 ... a magnet, 300 ... a crankshaft.

Claims (7)

Two magnetic sensing elements arranged to sense a change in magnetic vector emitted from a magnet that rotates as the rotation axis to be detected as a rotation angle is detected as a sine wave signal that is 90 degrees out of phase. The rotation angle of the rotating shaft is “θ”, one output signal A of the magnetic sensing element is “A = sin θ”, and the other output signal B of the magnetic sensing element is “B = cos θ”. A signal for converting the output signals from the two magnetic sensing elements into a linear signal that changes linearly with respect to the rotation of the rotating shaft based on an arithmetic expression of “θ = tan ⁻¹ (A / B)”. A rotation angle detecting device for detecting a rotation angle of the rotation shaft based on a signal extracted from the signal processing unit,
When the two magnetic sensing elements are magnetic sensing element pairs, the sensor unit includes a plurality of magnetic sensing element pairs, and the signal processing unit is configured to output the linear signal separately from output signals from the plurality of magnetic sensing element pairs. A rotation angle detection device that outputs a signal as rotation angle information of the rotation shaft by obtaining a signal and performing an operation that can absorb an angle error on the linear signal. The plurality of magnetic sensing element pairs are arranged so as to sense a change in magnetic vector accompanying rotation of the rotating shaft as an output signal having a different phase, and the signal processing unit is obtained separately from the output signal. The rotation angle detection device according to claim 1, wherein an operation capable of absorbing the angle error is performed on the linear signals while maintaining their phase relationship. The rotation angle detection device according to claim 1, wherein the operation capable of absorbing the angle error is an average value calculation for obtaining an average value of the linear signals obtained separately from output signals of the plurality of magnetic sensing element pairs. The operation that can absorb the angular error is a majority operation that selectively obtains one of the linear signals that occupies a large number of matches through comparison of the linear signals obtained separately for the output signals of the plurality of magnetic sensing element pairs. The rotation angle detection device according to claim 1 or 2. The magnet is formed of a disk-shaped magnetized rotor integrally formed with the rotating shaft in such a manner that the N pole and the S pole are separated and magnetized, and the sensor unit and the signal processing unit are one semiconductor. 5. The rotation according to claim 1, wherein the magnetic sensing element includes a vertical Hall element that senses a magnetic vector parallel to the semiconductor substrate surface based on a Hall effect. Angle detection device. The sensor unit has two magnetic sensing element pairs as the plurality of magnetic sensing element pairs.

The rotation angle detection apparatus as described in any one of Claims 1-5. The rotation angle detection device according to claim 1, wherein the sensor unit includes three magnetic sensing element pairs as the plurality of magnetic sensing element pairs.

Images