JP2001050829A - Magnetostrictive torque sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a magnetostrictive torque sensor in which the error due to temperature gradient can be compensated while reducing the size. SOLUTION: The torque sensor comprises magnetic anisotropic parts 11a, 11b provided on a torque transmission shaft 14, excitation coils 12a, 12b positioned in correspondence with the magnetic anisotropic parts 11a, 11b, and detection coils 12c, 12d. The detection coils 12c, 12d are excited with at least two frequencies and the value of torque applied to the to transmission shaft 14 is operated based on the results obtained by separating detection signals of the detection coils 12c, 12d for each frequency.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetostrictive torque sensor having a temperature compensation function.

[0002]

2. Description of the Related Art Conventionally, it has been practiced to connect a driving device, a load device and the like to measure a torque generated on a power shaft of a device body. As a torque sensor for measuring this torque, when a stress is applied to a magnetostrictive material such as permalloy (Fe-Ni) or sendust (Fe-Al-Si) and strain is generated in the magnetostrictive material, the magnetostrictive material is A magnetostrictive torque sensor using an inverse magnetostrictive effect in which the magnetic permeability changes has been proposed. FIG. 8 shows a schematic configuration diagram of a magnetostrictive torque sensor. In FIG. 8, the excitation coils 105a and 105b are excited by the AC signal source 104. In this state, when a torque is applied on the torque transmission shaft 101, the magnetically anisotropic portion 102 made of a magnetostrictive material
The strains a and 102b are distorted by the torque, and the magnetic anisotropy portions 102a and 102b change in magnetic permeability according to the strain. This change in the magnetic permeability causes a change in the mutual inductance between the detection coil and the excitation coil, and as a result, the signals detected by the detection coils 103a and 103b (hereinafter referred to as detection signals) change. Then, only the torque applied to the torque transmission shaft 101 is detected by obtaining the difference between the detection signals by the subtraction circuit. However, when actually measuring the torque value, thermal energy is supplied from the connected driving device, load device, or the like, and a temperature gradient is often generated on the torque transmission shaft 101 in many cases. Thus, when a temperature gradient occurs, the torque transmission shaft 101 and the magnetically anisotropic parts 102a, 10a
Since a gradient also occurs in the magnetic permeability and electric resistance of 2b, an error occurs in the detection signal as a result. At the same time, a temperature gradient generated on the torque transmission shaft 101 during torque measurement changes the coil resistance of the exciting coils 105a and 105b. Thus, the exciting coils 105a, 105b
When the coil resistance changes, the exciting coil 105a,
This causes a voltage drop of the excitation signal applied to 105b,
There is also a problem that an error occurs in the detection signal. Therefore,
To compensate for errors due to temperature gradients that occur in the torque transmission shaft and the magnetically anisotropic portion provided thereon, see Japanese Patent No.
The technology described in 64049 and the like has been proposed.

FIG. 9 is a block diagram showing the basic structure of a magnetostrictive torque sensor described in Japanese Patent No. 2564049. The specific configuration is as follows. Magnetic anisotropic portions 203a, 203b, and 203c are sequentially formed on the torque transmission shaft 202 at equal intervals in the longitudinal direction at an inclination of 45 ° from the central axis. However, only the magnetic anisotropic part 203a is inclined in the opposite direction to the other magnetic anisotropic parts 203a and 203b. Exciting coils 201a, 201b, and 201c and detecting coils 204a, 204b, and 204c are provided around the magnetic anisotropic parts 203a, 203b, and 203c, and the exciting coils 201a, 201b, and 201c are provided.
Is connected to an AC power supply 211 for supplying an exciting current. The detection coils 204a, 204b, 204
c is the rectifying filter 205a, 205b, 20
5c is connected to the input terminal side. Output terminals of the rectifying filters 205a and 205b are connected to an input terminal of a first subtraction circuit 206 for subtraction.
Output terminals of the rectifying filters 205b and 205c are connected to input terminals of a second subtraction circuit 207 for subtraction. Further, the output terminals of these subtraction circuits 206 and 207 are connected to the input terminals of a third subtraction circuit 208 for subtraction. On the other hand, the rectifying filter 205a
And the output terminal side of 205c is an addition circuit 2 for addition.
09 is connected to the input terminal side. This adding circuit 20
9 is connected to a subtraction point 212 for comparison with the reference voltage V, and the result of the comparison is input to the auto gain controller 210. The output side of the auto gain controller 210 is an AC power supply 211.
And the excitation coils 201a, 201b, 201
It is possible to control the power supplied to c. Then, when a temperature gradient occurs on the torque transmission shaft 202,
An error in the measurement result of the torque value due to the temperature gradient generated in the detection coils 204a, 204b, 204c can be compensated for by an operation in an arithmetic circuit.

[0004]

However, in the magnetostrictive torque sensor described in Japanese Patent No. 2564049, the excitation coil, the detection coil, and the magnetic anisotropic portion are required to compensate for an error caused by a temperature gradient. It is necessary to provide at least three or more places. Therefore, there is a problem that the magnetostrictive torque sensor becomes longer in the axial direction of the torque transmission shaft. Further, if the magnetostrictive torque sensor shown in FIG. 8 is used, the size of the magnetostrictive torque sensor can be reduced, but errors due to the temperature gradient cannot be compensated.
In view of the above facts, it is an object of the present invention to reduce the size of a magnetostrictive torque sensor and at the same time to provide a magnetostrictive torque sensor that compensates for an error due to a temperature gradient.

[0005]

According to a first aspect of the present invention, there is provided a magnetostrictive torque sensor according to the present invention.
A magnetically anisotropic part provided on the outer peripheral surface of the torque transmission shaft,
A detection coil that detects a change in magnetic permeability in the magnetic anisotropic part and outputs a detection signal, and excites the detection coil at at least two frequencies to generate a detection signal detected by the detection coil. Detection signal separating means for separating each excitation frequency and outputting it as a separated detection signal; and torque value calculating means for calculating a torque value applied to the torque transmission shaft based on the separated detection signal. It is characterized by. In the magnetostrictive torque sensor according to claim 2, in addition to the features described in claim 1, the magnetic anisotropic part has a first magnetic anisotropic part and a second magnetic anisotropic part, When a torque is applied to the torque transmitting shaft of the magnetic anisotropic portion, the first magnetic anisotropic portion, or one of the second magnetic anisotropic portions receives a tensile stress, 4. The magnetostrictive torque sensor according to claim 3, wherein the other is arranged to receive a compressive stress, in addition to the features according to claim 1 or 2, wherein a material of the torque transmission shaft is provided. Is characterized by using a non-magnetic material. Claim 4
The magnetostrictive torque sensor according to the present invention is characterized in that, in addition to the features according to any one of the first to third aspects, permalloy is used as a magnetostrictive material of the magnetic anisotropic portion. The magnetostrictive torque sensor according to claim 5 has, in addition to the features described in any one of claims 1 to 4, a yoke surrounding the detection coil, wherein the yoke is formed using ferrite. The magnetostrictive torque sensor according to claim 6 is characterized in that, in addition to the features according to any one of claims 1 to 5,
An excitation coil that is excited by an excitation signal to generate a magnetic field, so that the current value of the excitation signal supplied to the excitation coil is kept constant even if the resistance value of the excitation coil changes. It is controlled. A magnetostrictive torque sensor according to a seventh aspect is characterized in that, in addition to the features described in the sixth aspect, the yoke surrounds the detection coil and the excitation coil.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [1] First Embodiment [1.1] Configuration of First Embodiment FIG. 1 is a block diagram showing a basic configuration of a magnetostrictive torque sensor according to a first embodiment of the present invention. . In FIG. 1, the magnetostrictive torque sensor includes a sensor unit 1, an excitation signal generation unit 2, and a detection signal calculation unit 3.

FIG. 2 is a sectional view showing the basic configuration of the sensor section 1. In FIG. 2, reference numeral 16 denotes a cylindrical sensor unit 1.
Of the sensor case. The sensor section 1 has the same central axis as the central axis 4a in the length direction (horizontal direction on the paper) of the sensor case main body 16, and the sensor case main body 16 has
And a torque transmission shaft 14 that penetrates through. As a material used for the torque transmission shaft 14, a material that is nonmagnetic, has wear resistance, and has high tensile strength and fatigue strength is suitable. Examples of the material of the torque transmission shaft 14 include austenitic stainless steel such as SUS301 and 304, a titanium alloy, and duralumin.

On the outer peripheral surface of the torque transmitting shaft 14 surrounded by the sensor case main body 16, magnetic anisotropic portions 11a and 11b made of magnetostrictive material
Is provided. In this case, the magnetic anisotropic part 1
Planes 4c, 4c 'including the longitudinal central axes of 1a, 11b
(Hereinafter referred to as the magnetically anisotropic part planes 4c and 4c ') and the central axis 4a of the torque transmission shaft 14 are provided to have an angle of 45 °. At this time, the magnetic anisotropic portion plane 4c,
4c 'are orthogonal to each other to form a "C-shape" (chevron). These magnetic anisotropic parts 11a, 11b
For example, permalloy (Fe—N
i), sendust (Fe-Al-Si) and the like, and are formed as a thin film on the torque transmission shaft 14 by a method such as a wet plating method, a sputtering method, a plasma spraying method, or an ion plating method.

The torque transmission shaft 14 is provided on the magnetic anisotropic portion 1.
1a, 11b, bearings 15a, 1b
5b so that the sensor case body 1 can rotate.
6 is supported. As the bearings 15a and 15b, for example, ball bearings or the like are used to reduce frictional resistance. The torque transmission shaft 14 is provided with the bearing 15
Since it is rotatably supported by the sensor case main body 16 by a and 15b, when receiving a driving force of a driving device, a load device, or the like (not shown) for which torque is to be measured, a rotational motion can be performed.

In the sensor case main body 16, two exciting coils 1 are provided on the outer peripheral surface of the torque transmitting shaft 14 in which the magnetically anisotropic portions 11a and 11b are formed.
2a and 12b are wound. Two detection coils 12c and 12d are further wound on the outer peripheral surfaces of the two excitation coils 12a and 12b, respectively. Here, the exciting coil 12a and the detecting coil 12
c corresponds to the magnetic anisotropic part 11a, and the exciting coil 12b and the detecting coil 12d correspond to the magnetic anisotropic part 11b.

Further, inside the sensor case main body 16, excitation coils 12a and 12b and detection coils 12c and 1c are provided.
Yoke 1 made of a ferromagnetic material so as to surround 2d
3 is provided for adjusting the magnetic field. This yoke 1
For example, ferrite (MIIO)
Fe ₂ O ₃ ; where MII = Mn, Fe, Co, Ni,
Cu, Zn, Mg, and Cd).

FIG. 3 is a block diagram showing a functional configuration of the magnetostrictive torque sensor. In FIG. 3, the exciting coils 12a and 12b are connected in series, and furthermore, oscillating circuits 21a, 21b, and 2 provided in the exciting signal generator 2.
1c. Oscillation circuits 21a, 21b, 21
c is the frequency f ₁ , f ₂ , f ₃ (where f ₁ <f ₂
<And f _3. ) [Hz] AC signal for exciting coil 1
It is an AC signal generation source supplied to 2a and 12b. Also,
The oscillating circuits 21a, 21b, and 21c are provided by the temperature gradient generated on the torque transmission shaft 14 when measuring the torque value.
Even if the resistance value of the exciting coil changes, control is performed so that the current value of the exciting signal does not change. In this way, by keeping the current value constant, the generated magnetic flux also becomes constant. Here, the frequencies f ₁ , f ₂ , and f ₃ are frequencies that meet the condition that the reversal of the magnetic dipole can sufficiently follow the reversal of the magnetic field vector.

On the other hand, the detection coil 12c has a frequency f ₁
Bandpass filter 2 that passes signals of nearby frequencies
2a, and is connected band-pass filter 22b which passes the frequency of the signal of the frequency f ₂ near the band-pass filter 22c which passes the frequency of the signal of the frequency f ₃ near the input terminal side of the. The detection coil 12d is passing a band-pass filter 22d for passing the frequency of the signal of the frequency f ₁ near the band-pass filter 22e for passing the frequency of the signal of the frequency f ₂ near the frequency of the signal of the frequency f ₃ near It is connected to the input terminal side of the band pass filter 22f. These bandpass filters 22
a, 22b, 22c, 22d, 22e, by 22f, the detection coils 12c, a detection signal outputted from 12d, it is possible to separate the respective components of the frequency _{f 1, f 2, f 3} .

The band pass filters 22a, 22b,
A rectifier circuit 23 having a function of performing full-wave rectification or half-wave rectification on a detection signal separated into components of frequencies f ₁ , f ₂ , and f ₃ is provided on an output terminal side of 22c, 22d, 22e, and 22f.
a, 23b, 23c, 23d, 23e, and 23f, respectively.

The rectifier circuits 23a, 23b, 23
Output terminal sides of c, 23d, 23e, and 23f are low-pass filters 24a, 24b, 24c, 24d, 24e, 2
4f is connected to each input terminal. These low-pass filters 24a, 24b, 24c, 24d, 24
e, 24f are rectifier circuits 23a, 23b, 23c, 23
This is a device that transmits a DC component of a detection signal output from d, 23e, and 23f and further cuts noise of the detection signal.

The low-pass filters 24a, 24b, 2
Output terminal sides of 4c, 24d, 24e, and 24f are amplification factor adjustment circuits 25a, 25b, 25c, 25d, 25e, and 2e.
5f is connected to the input terminal side. The amplification factor adjusting circuits 25a, 25b, 25c, 25d, 25e,
25f, when the temperature of the torque transmission shaft 14 is uniform at a reference temperature T _0, in the context of the strain on the torque transmission shaft 14 does not occur, a low pass filter 24a, 24b,
Assume 24c, 24d, 24e, an initial setting is made so as to amplify the voltage of the detection signal output from 24f to V _0, respectively. This reference temperature is desirably set near room temperature (for example, 25 ° C.).

The amplification factor adjusting circuits 25a, 25b, 2
The output terminal side of 5c is connected to the input terminal side of the addition operation circuit 26a. Further, the amplification factor adjusting circuits 25d and 25d
Output terminals e and 25f are connected to an input terminal side of the addition operation circuit 26a. On the other hand, an offset voltage generation circuit 27a is connected to another input terminal of the addition operation circuit 26a, and an offset voltage generation circuit 27b is connected to another input terminal of the addition operation circuit 26b. The offset voltage generation circuits 27a and 27b
This is an apparatus for applying an offset voltage to signals from a, 25b, 25c, 25d, 25e, and 25f to control an error.

The output terminals of the addition operation circuits 26a and 26b and the output terminals of the low-pass filters 24b and 24e are
It is connected to the input terminal of CPU28. This CPU2
Numeral 8 denotes a circuit capable of performing various operations.
a, 26b and a device that performs calculations in accordance with the output results from the low-pass filters 24b, 24e.

[1.2] Operation of First Embodiment First, in FIG. 2, a drive device, a load device, and the like whose torque is to be measured are connected to the torque transmission shaft 14. When no torque is applied from the connected driving device, load device, or the like, no distortion occurs in the pair of magnetically anisotropic portions 11a and 11b provided on the torque transmission shaft 14. On the other hand, when a torque is applied to the torque transmission shaft 14 from the connected drive device or load device, tensile and compressive stresses are applied to the pair of magnetic anisotropic portions 11a and 11b in the 45 ° direction. . As described above, the magnetic anisotropy planes 4c and 4c ′ are arranged in a “C” shape such that the planes 4c and 4c ′ have an angle of 45 ° with the central axis 4a of the torque transmission shaft 14, and
This is because they are orthogonal to each other (see FIG. 2).

Next, referring to FIG.
1b, 21c, the frequency f _1, f _2, f ₃ of the synthesized two exciting coils 12a excitation signal, supplied to 12b.
When the two excitation coils 12a and 12b are excited, a detection signal proportional to the inductance is induced as an AC signal in the two detection coils 12c and 12d. This detection signal is a composite signal of the frequencies f ₁ , f ₂ and f ₃ . In this state, the torque is applied to the torque transmission shaft 14 so that the magnetically anisotropic portions 11a, 11a
When the magnetic permeability of the magnetostrictive material forming b changes, the detection signals of the detection coils 12c and 12d change.

The detection signal detected by the detection coil 12c is separated into the respective frequencies f ₁ , f ₂ and f ₃ by band-pass filters 22a, 22b and 22c. The detection signals separated for each of the frequencies f ₁ , f ₂ , and f ₃ are subjected to full-wave rectification or half-wave rectification by rectification circuits 23a, 23b, and 23c.
The signals are converted into DC signals (hereinafter referred to as DC detection signals) by 24b and 24c. These frequencies f ₁ , f ₂ , f ₃
The DC detection signal for each of the gain adjustment circuits 25a,
After the amplification factors are adjusted in 25b and 25c, the addition circuit 26
is input to a. Further, an offset voltage is supplied from an offset voltage generation circuit 27a connected to the addition circuit 26a to control an error. The adjustment of the amplification factor and the offset voltage will be described later in detail.

Similarly, the combined AC signal detected by the detection coil 12d is also transmitted to the bandpass filters 22d and 22d.
e, 22f, and detected by the rectifier circuits 23d, 23e, 23
f, and is converted into a DC signal via low-pass filters 24d, 24e, 24f. Thereafter, the amplification factor adjusting circuit 25
d, 25e, and 25f, and are amplified by an addition circuit 26b.
Is input to Further, an offset voltage is supplied from an offset voltage generation circuit 27b connected to the addition circuit 26b in order to control occurrence of an error.

The addition circuits 26a and 26b add the voltage value of the DC detection signal and the offset voltage for each of the frequencies f ₁ , f ₂ and f ₃ , which will be described in detail later.

The CPU 28 performs various calculations on the signals added by the adders 26a and 26b and the DC detection signal of the frequency f ₂ from the low-pass filters 24b and 24e, and outputs the signals as a torque signal 29. This CPU 28
The details of the various processes will be described later.

[1.3] Adder circuits 26a, 26b and C
Calculation Performed by PU28 The frequency f is applied to the exciting coil 12a.₁, F_Two, F_ThreeExcitation signal
Is applied, torque τ is marked on the torque transmission shaft 14.
Detection signal detected by the detection coil 12c
Signal by the band pass filters 22a, 22b, 22c
Frequency f₁, F_Two, F_ThreeThink when separated into components
You. At this time, the frequency f₁About the DC detection signal voltage
To V_A1Then, this V_A1Is represented by the following equation. V_A1= V₀+ A_A1・ (T_A-T₀) + B_A1・ Τ + c_A1・ Τ ・ (T_A-T₀) …… (1) where a_A1・ (T_A-T₀) Indicates that no torque is applied
In the state, the temperature of the magnetic anisotropic portion 11a is T₀To T_ABecame
Voltage change of the DC detection signal at the time, b_A1・ Τ is the reference temperature
T₀DC detection signal voltage when torque τ is applied at
Change, c_A1・ Τ ・ (T_A-T₀) Indicates the magnetic anisotropic portion 11a.
Temperature is T₀To T_AWhen the torque changes to
Is the voltage change of the DC detection signal at the time of occurrence. Paraphrase
If b_A1Is the sensor sensitivity, c_A1Is the temperature of the sensor sensitivity
Dependency. Similarly, the frequency f _Two, F_ThreeDC detection signal
Signal voltage to V_A2, V_A3Then V_A2And V_A3Is
expressed. V_A2= V₀+ A_A2・ (T_A-T₀) + B_A2・ Τ + c_A2・ Τ ・ (T_A-T₀) ……… (2) V_A3= V₀+ A_A3・ (T_A-T₀) + B_A3・ Τ + c_A3・ Τ ・ (T_A-T₀) …… (3) At this time, a_Ai, B_Ai, C_AiThe value of (i = 1 to 3) is prior
Can be determined by measurement, and furthermore, these values
The electrical properties, permeability, inverse magnetostriction, etc. of the constituent materials depend on the frequency.
Values that are unique to each frequency in order to have
It is. Here, equations (1) to (3) are converted through the following calculation procedure.
Equations (4) to (8) are obtained. (Equation (1) · c_A2) − (Equation (2) · c_A1) By τ
・ (T_A-T₀) Is deleted to obtain equation (a1). (Equation (2) · c_A3) − (Equation (3) · c_A2) By τ
・ (T_A-T₀) Is deleted to obtain equation (a2). (Equation (a1) · (b_A2・ C_A3-C_A2・ B_A3))-(Expression
(A2) ・ (b_A1・ C_A2-C_A1・ B_A2) At τ_ASection
Divided by T_A= W_A・ V_A1+ X_A・ V_A2+ Y_A・ V_A3+ Z_A ……… (4) where W_A, X_A, Y_A, Z_AIs represented by the following equation. W_A= C_A2・ (B_A2・ C_A3-C_A2・ B_A3) / ｛(B_A2・ C_A3-C_A2・ B_A3) (A_A1・ C_A2-C_A1・ A_A2)-(B_A1・ C_A2-C_A1・ B_A2) (A_A2・ C_A3-C_A2・ A_A3)｝ ……… (5) X_A= − ｛C_A1・ (B_A2・ C_A3-C_A2・ B_A3) + C_A3・ (B_A1・ C_A2-C_A1・ B_A2)｝ / ｛(B_A2・ C_A3-C_A2・ B_A3) (A_A1・ C_A2-C_A1・ A_A2)-(B_A1・ C_A2-C_A1・ B_A2) (A_A2・ C_A3-C_A2・ A_A3)｝ ............ (6) Y_A= C_A2・ (B_A1・ C_A2-C_A1・ B_A2) / ｛(B_A2・ C_A3-C_A2・ B_A3) (A_A1・ C_A2-C_A1・ A_A2)-(B_A1・ C_A2-C_A1・ B_A2) (A_A2・ C_A3-C_A2・ A_A3)｝ ……… (7) Z_A= ｛(C_A1-C_A2) (B_A2・ C_A3-C_A2・ B_A3)-(C_A2-C_A3) (B_A1・ C_A2-C_A1・ B_A2)｝ ・ V₀ / ｛(B_A2・ C_A3-C_A2・ B_A3) (A_A1・ C_A2-C_A1・ A_A2)-(B_A1・ C_A2-C_A1・ B_A2) (A_A2・ C_A3-C_A2・ A_A3)｝ ……… (8)

On the other hand, in the DC detection signal of the other detection coil 12b, since the magnetically anisotropic parts 11a and 11b are orthogonal to each other, the stress values are reversed. The voltage of the DC detection signal for this for frequency f _i V _Bi
Is expressed by the following equation, where T _B is the temperature at the magnetically anisotropic part 11b. V _Bi = V ₀ + a _Bi · (T _B −T ₀ ) −b _Bi · τ−c _Bi · τ · (T _B −T ₀ ) (9) However, the values of a _Bi , b _Bi , and c _Bi (i = 1 to 3) can be obtained by preliminary measurement. Therefore, it is possible to determine the temperature T _A as well as the temperature T _B from equation (9). _{T B = W B · V B1} + X B · V B2 + Y B · V B3 + Z B ......... (10) _{Further, W B, X B, Y} B, Z B is expressed by the following equation. _{W B = c B2 · (b} B2 · c B3 -c B2 · b B3) / {(b B2 · c B3 -c B2 · b B3) (a B1 · c B2 -c B1 · a B2) - (b _B1 · c _B2 −c _B1 · b _B2 ) (a _B2 · c _B3 −c _B2 · a _B3 )｝ (11) X _B = − ｛c _B1 · (b _B2 · c _B3 −c _B2 · b _{B3) + c B3 · (b} B1 · c B2 -c B1 · b B2)} / {(b B2 · c B3 -c B2 · b B3) (a B1 · c B2 -c B1 · a B2) - (b _B1 · c _B2 −c _B1 · b _B2 ) (a _B2 · c _B3 −c _B2 · a _B3 )｝ (12) Y _B = c _B2 · (b _B1 · c _B2 −c _B1 · b _B2 ) / ｛(B _B2 · c _B3 -c _B2 · b _B3 ) (a _B1 · c _B2 -c _B1 · a _B2 )-(b _B1 · c _B2 -c _B1 · b _B2 ) (a _B2 · c _B3 -c _{B2 · a B3)} ......... (} 13) Z B = {(c B1 -c B2) (b B2 · c B3 -c B2 · b B3) - (c B2 -c B3) (b B1 · c B2 _{-c B1 · b B2)} ·} V 0 _{{(B B2 · c B3 -c} B2 · b B3) (a B1 · c B2 -c B1 · a B2) - (b B1 · c B2 -c B1 · b B2) (a B2 · c B3 -c B2・ A _B3 )｝ ……… (14)

That is, in the adding circuits 26a and 26b,
Is the temperature T according to equations (4) and (10)._A, T_BAsked
Will be done. However, the gain adjustment circuit 25
a, 25b, 25c, 25d, 25e, 25f are reference temperatures
V with no torque applied in degrees₀So that
Adjustment was assumed, but the torque was actually marked at the reference temperature.
V when not added₀・ W_A, V₀・ X_A, V₀・ Y_A,
V₀・ W_B, V₀・ X_B, V ₀・ Y_BAdjusted to output
There is a need to. For this reason, the amplification factor adjusting circuits 25a,
The output at 5b, 25c, 25d, 25e, 25f is given by the formula
(4), W of (10)_A・ V_A1, X_A・ V_A2, Y_A・
V_A3, W_B・ V_B1, X_B・ V_B2, Y_B・ V_B3As a term
This is input to the arithmetic circuits 26a and 26b. Also,
Z in equations (4) and (10)_A, Z_BOffset voltage
It must be generated by the generation circuits 27a and 27b.

The torque τ can be obtained from the difference between the temperatures T _A and T _B obtained so far and an arbitrary frequency. For example, when the frequency f ₂ is used, the following equation is obtained by solving the torque τ by taking the difference between the equations (2) and (9). τ = {V _A2 + V _B2 + a _A2 · T _A + a _B2 · T _B2 + T ₀ (a _A2 −a _B2 )} / {c _A2 · (T _A -T ₀ ) + c _B2 · (T _B -T ₀ ) + (B _A2 + b _B2 )｝ (15) That is, the CPU 28 calculates the torque τ according to the equation (15).
Is required. Here, V _A2 and V _B2 are CPU
Low-pass filter 24 connected to the input terminal side
b, 24e on the output terminal side. C
The expression (15) calculated by the PU 28 is as follows.
The error for the temperature gradient generated above is compensated. Therefore, the torque signal 29 from the CPU 28 is obtained as a value that has been subjected to temperature compensation.

In the present embodiment, the CPU 28 is used as shown in FIG. 3, but a combination of analog circuits capable of calculating the equation (15) may be used. Further, in the present embodiment, three frequencies are used as the excitation frequencies. However, when the temperature dependence of the sensor sensitivity can be ignored, the excitation frequencies are set to the two frequencies and c _A1 ,
It is also possible to obtain temperatures T _A and T _{B by} setting c _A2 , c _B1 , and c _B2 to 0, and to solve for the torque τ. In addition, when the nonlinearity of the material is accurately considered in order to improve the accuracy of the measurement result of the torque value, a higher-order term may be generated. When a higher-order term becomes necessary, the excitation frequency is increased by that amount, and a higher accuracy measurement value is obtained by calculating in the same manner as the calculation method of Expression (4) → Expression (8). It is also possible.

As described above, according to the magnetostrictive torque sensor according to the present embodiment, the size of the magnetostrictive torque sensor can be reduced.
At the same time, it is possible to provide a magnetostrictive torque sensor that compensates for an error due to a temperature gradient.

[2] Second Embodiment [2-1] Configuration of Second Embodiment A block diagram showing a basic configuration of a magnetostrictive torque sensor according to a second embodiment is the same as that of FIG. That is, the magnetostrictive torque sensor shown in the present embodiment has the same sensor unit 1 and exciting current supply device 2 as the magnetostrictive torque sensor shown in the first embodiment.

FIG. 4 is a block diagram showing a functional configuration of the magnetostrictive torque sensor according to the second embodiment. In FIG. 4, the band pass filters 22a, 22b, 22c, 22
d, 22e, 22f, rectifier circuits 23a, 23b, 23
c, 23d, 23e, 23f, low-pass filter 24
a, 24b, 24c, 24d, 24e, 24f, amplification factor adjustment circuits 25a, 25b, 25c, 25d, 25e, 2
5f, addition operation circuits 26a and 26b, offset voltage generation circuits 27a and 27b, and torque signal 29 are the same as those shown in FIG. 3 of the first embodiment.

The configuration of the second embodiment is different from that of the first embodiment in that the input terminals of the CPU 30 are not connected to the output terminals of the low-pass filters 24b and 24e, and the addition performed by the addition circuits 26a and 26b. Results and CPU
Only the result of the operation performed in 30 is shown. Adder 26a,
The addition performed in 26b and the calculation performed in CPU 30 will be described later.

[2.2] Operation of the Second Embodiment The operation of the second embodiment differs from the "operation of the first embodiment" in that the addition performed by the adder circuits 26a and 26b and the CPU
Since only the calculation performed in step 30 is performed, this will be described.

The frequency f is applied to the exciting coil 12a.₁, F_Two, F
_Three, And the torque transmission shaft 14
Is detected by the detection coil 12c when torque τ is applied to
The detected signal to be detected is a bandpass filter 22a, 22b,
Frequency f at 22c₁, F_Two, F _Three, Separated into components
Think about it. At this time, the amplification factor adjusting circuits 25a, 25b,
The voltage of the DC detection signal from_A1, V_A2, V_A3When
I do. At this time, V_A1, V_A2, V_A3Is the equation (1) to (3)
It is expressed by the following equation. However, similar to the first embodiment,
Reference temperature T₀(Eg 25 ° C) and no torque applied
When the output is V₀Assume that Also, a_Ai, B_Ai, C_Ai
The value of (i = 1 to 3) can be obtained by preliminary measurement.
You.

Here, when the equations (1) to (3) are solved by the following procedure, the following equations are obtained. (Equation (1) · c _A2 ) − (Equation (2) · c _A1 )
_{The term of A} · (T _A −T ₀ ) is deleted to obtain equation (a3). (Equation (2) · c _A3 ) − (Equation (3) · c _A2 )
To remove the section of the _A · (T _A -T ₀₎ and formula (a4). (Eq. ( _A3 ) · (a _A2 · c _A3 −c _A2 · a _A3 )) − (Eq. ( _A4 ) · (a _A1 · c _A2 −a _A2 · c _A1 ), delete the term T _A −T _0. Τ = A _A · V _A1 + B _A · V _A2 + C _A · V _A3 + D _A ... (16) where A _A , B _A , and C _A are represented by the following equations. A _A = c _A2 · (a _A2 · c _A3 −c _A2 · a _A3 ) / ｛(a _A2 · c _A3 −c _A2 · a _A3 ) (b _A1 · c _A2 −c _A1 · b _A2 ) − (a _A1 · c _A2 −c _A1 · a _A2 ) (b _A2 · c _A3 −c _A2 · b _A3 )｝ ……… (17) B _A = − ｛c _A1 · (a _A2 · c _A3 −c _A2 · a _A3 ) + c _A33 (a _{A11c A2} -c _{A1ａa A2} ) /｝ (a _A2・ c _A3 -c _A2・ a _A3 ) (b _A1・ c _A2 -c _{A1ｂb A2} )-(a _{A1 · c A2 -c A1 · a} A2) (b A2 · c A3 -c A2 · b A3)} ......... (18) C A = c A2 · (a A1 · c A2 -c A1 · a A2) / ｛(A _A2 · c _A3 -c _A2 · a _A3 ) (b _A1 · c _A2 -c _A1 · b _A2 )-(a _A1 · c _A2 -c _A1 · a _A2 ) (b _A2 · c _A3 -c _A2 · b _A3 )｝ (19) D _A = ｛(c _A1 −c _A2 ) (a _A2 · c _A3 −c _A2 · a _A3 ) − (c _A2 −c _A3 ) (a _A1 · c _{A2 -c A1 · a A2)} ·} V 0 _{{(A A2 · c A3 -c} A2 · a A3) (b A1 · c A2 -c A1 · b A2) - (a A1 · c A2 -c A1 · a A2) (b A2 · c A3 -c A2・ B _A3 )｝ ……… (20)

On the other hand, since the DC output of the other magnetically anisotropic part 11b is represented by the equation (9), the torque τ can be similarly obtained. τ = A _B V _B1 + B _B V _B2 + C _B V _B3 + D _B ............ (21)

In Equations (16) and (21), there is no temperature parameter in the equations, and errors caused by temperature changes can be absorbed. However, the gain adjustment circuits 25a, 25b, 25c, 25d, 25e, 2
5f is V ₀ at the reference temperature with no torque applied.
It is assumed that the adjustment is made such that V ₀ · A _A , V ₀ ·
_{B A, V 0 · C A} , V 0 · A B, V 0 · B B, it is necessary to adjust so that the output of V ₀ · C _B. Therefore, the amplification factor adjusting circuits 25a, 25b, 25c, 25d, 25e, 25
The output at f is represented by A _A · V _A1 and B _{A in} equations (16) and (21).
_{· V A2, C A · V} A3, A B · V B1, B B · V B2, C B · V B3
Are input to the adders 26a and 26b. Further, the formula _{(16) (21) D A} , D B of the voltage value offset voltage generating circuit 27a, and must be generated by 27b.

In this way, the CPU 30 makes the addition circuit 2
The error can be reduced by taking the average value of the detection results of the torque τ subjected to the arithmetic processing in 6a and 26b.

In this embodiment, the CPU 30 is used as shown in FIG.
A combination of analog circuits that can calculate the average value of the detection results of the torque τ subjected to the arithmetic processing in b may be used. In this embodiment, the excitation frequency is 3
Although two frequencies are used, if the temperature dependence of the sensor sensitivity can be ignored, the excitation frequency is set to two frequencies and c
It is also possible to determine the torque τ by setting _A1 , _cA2 , _cB1 , and _cB2 to 0. Furthermore, when the nonlinearity of the material is accurately considered in order to improve the accuracy of the measurement result of the torque value, a higher-order term may be generated. When a higher-order term becomes necessary, the excitation frequency is increased by that amount, and equation (16) →
By performing the calculation in the same manner as the calculation method of Expression (20), it is possible to obtain a more accurate measurement value.

[3] Third Embodiment [3.1] Configuration of Third Embodiment FIG. 5 is a block diagram showing a basic configuration of a magnetostrictive torque sensor according to a third embodiment. The magnetostrictive torque sensor according to the present embodiment includes a sensor unit 10, an exciting current generating unit 2, and a detection signal calculating unit 5, similarly to the magnetostrictive torque sensor according to the second embodiment. 5, the exciting current generator 2 is the same as that shown in FIG. 1 of the first embodiment.

FIG. 6 is a sectional view showing the basic structure of the sensor section 10. As shown in FIG. 6, the sensor case body 16, the torque transmission shaft 14, the torque transmission shaft bearings 15a and 15b, the magnetic anisotropic portion 11a, the exciting coil 12a, the detection coil 12c, and the yoke 13 are the sensor portion of the first embodiment. 1, which is the same as FIG.

FIG. 7 is a block diagram showing a functional configuration of the magnetostrictive torque sensor according to the third embodiment. In FIG. 7, oscillation circuits 21a, 21b, 21c, band-pass filters 22a, 22b, 22c, rectifier circuits 23a,
3b, 23c, low-pass filters 24a, 24b, 24
c, amplification factor adjustment circuits 25a, 25b, 25c, addition operation circuit 26a, offset voltage generation circuit 27a, and torque signal 29 are the same as those shown in FIG. 3 of the first embodiment.

As shown in FIG. 6, the magnetostrictive torque sensor according to the third embodiment is different from the torque sensor according to the second embodiment only in that the magnetic anisotropic portion 11a is provided at only one position. 12c, and the exciting coil 12a is provided at one position. In addition, since the detection coil is provided at one position of the detection coil 12c, the circuit of the detection signal calculation unit is provided only on one side, and the output of the addition calculation circuit 26a is used as the torque signal 29 as it is.

[3.2] Operation of Third Embodiment The only difference between the operation of the present embodiment and the “operation of the second embodiment” is the operation performed by the CPU 30. This will be described.

Here, the operation performed by the addition circuit 26a is performed in the same manner as in the equation (16). Also, in equation (16), there is no temperature parameter in the equation,
Since an error caused by a temperature change can be absorbed, even when the magnetic anisotropic portion of the torque sensor shown in the second embodiment is set to only one portion 11a and the output of the addition operation circuit 26a is used as it is as the torque signal 29, Compensation for the temperature fluctuation of the sensor unit 4 can be performed. However,
Although it has been previously assumed that the amplification factor adjusting circuits 25a, 25b, and 25c are adjusted to V ₀ at a reference temperature and no torque is applied, actually, V is adjusted at a reference temperature and no torque is applied. _{0 · a a, V 0 ·} B a, it is necessary to adjust so that the output of V ₀ · C _a. For this reason, the outputs of the amplification factor adjustment circuits 25a, 25b, and 25c are expressed by the formula (1).
6) are input to the adder 26a as terms of A _A · V _A1 , B _A · V _A2 , and C _A · V _A3 . Equation (1)
The voltage value of the D _A 6) must be generated by the offset voltage generating circuit 27a.

In the present embodiment, the excitation frequency is set to 3
Although two frequencies are used, if the temperature dependence of the sensor sensitivity can be ignored, the excitation frequency is set to two frequencies and c
It is also possible to determine the torque τ by setting _A1 and _cA2 to 0. Furthermore, when the nonlinearity of the material is accurately considered in order to improve the accuracy of the measurement result of the torque value, a higher-order term may be generated. When a higher-order term becomes necessary, the excitation frequency is increased by that amount, and Equation (16) → Equation (2)
By performing the calculation in the same manner as the calculation method of 0), it is possible to obtain a more accurate measurement value.

As described above, the temperature compensation can be performed in the same manner as in the second embodiment by using the torque τ subjected to the arithmetic processing in the addition circuit 26a. Further, it is possible to perform temperature compensation with a small magnetostrictive torque sensor.

[4] Other Embodiments In the present invention, as described above, the formula (15)
The torque value is measured based on (16) and (21). However, the present invention is not limited to this.
Any device can be used as long as torque can be measured by exciting at two or more frequencies.

[0050]

As described above, according to the magnetostrictive torque sensor of the present invention, the use of at least two excitation frequencies makes it possible to compensate for temperature fluctuations occurring in the sensor section. Further, according to the magnetostrictive torque sensor of the present invention, the size of the sensor unit itself can be reduced, so that a small-sized magnetostrictive torque sensor can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of a magnetostrictive torque sensor according to a first embodiment.

FIG. 2 is a cross-sectional view illustrating a basic configuration of a sensor unit according to the first embodiment.

FIG. 3 is a block diagram showing a functional configuration of the magnetostrictive torque sensor according to the first embodiment.

FIG. 4 is a block diagram illustrating a functional configuration of a magnetostrictive torque sensor according to a second embodiment.

FIG. 5 is a block diagram showing a basic configuration of a magnetostrictive torque sensor according to a third embodiment.

FIG. 6 is a cross-sectional view illustrating a basic configuration of a sensor unit according to a third embodiment.

FIG. 7 is a block diagram illustrating a functional configuration of a magnetostrictive torque sensor according to a third embodiment.

FIG. 8 is a block diagram showing a basic configuration of a conventional magnetostrictive torque sensor.

FIG. 9 is a block diagram showing a basic configuration of a magnetostrictive torque sensor described in Japanese Patent No. 2564049.

[Explanation of symbols]

1: sensor part, 11a, b ... magnetic anisotropic part, 12a, b
... Exciting coil, 12c, d ... Detection coil, 13 ... Yoke, 14 ... Torque transmission shaft, 15a, b ... Bearing, 21
a, b, c ... oscillation circuit, 22a, b, c, d, e, f ...
Band-pass filters 23a, b, c, d, e, f ... rectifier circuits 24a, b, c, d, e, f ... low-pass filters 25a, b, c, d, e, f ... amplification factor adjustment circuits 26a, b ... addition operation circuits, 27a, b ... offset voltage generation circuits 28 ... CPU

Claims (7)

[Claims] A magnetic coil having a magnetic anisotropy portion provided on an outer peripheral surface of the torque transmitting shaft; and a detection coil for detecting a change in magnetic permeability in the magnetic anisotropic portion and outputting a detection signal. The detection coil is excited at two frequencies, a detection signal detected by the detection coil is separated for each excitation frequency, and a detection signal separating unit that outputs the separated detection signal is provided. And a torque value calculating means for calculating a torque value applied to the torque transmission shaft. 2. The magnetic anisotropic part has a first magnetic anisotropic part and a second magnetic anisotropic part, and the magnetic anisotropic part is provided when a torque is applied to the torque transmission shaft. Preferably, one of the first magnetic anisotropic part and the second magnetic anisotropic part is arranged to receive a tensile stress and the other to receive a compressive stress. Item 2. A magnetostrictive torque sensor according to item 1. 3. The magnetostrictive torque sensor according to claim 1, wherein a non-magnetic material is used as a material of the torque transmission shaft. 4. The magnetostrictive torque sensor according to claim 1, wherein permalloy is used as a magnetostrictive material of the magnetically anisotropic portion. 5. The magnetostrictive torque sensor according to claim 1, further comprising a yoke surrounding the detection coil, wherein the yoke is formed using ferrite. 6. An excitation coil for generating a magnetic field for exciting the detection coil based on excitation signals corresponding to at least the two frequencies, wherein a resistance value of the excitation coil changes. The magnetostrictive torque sensor according to any one of claims 1 to 4, wherein the current value of the excitation signal supplied to the excitation coil is controlled to be constant. 7. The magnetostrictive torque sensor according to claim 6, further comprising a yoke surrounding the detection coil and the excitation coil, wherein the yoke is formed using ferrite.