JPS61292014A - Position detector - Google Patents

Abstract

PURPOSE:To simplify to produce the titled detector by constituting the graduations of a scale body using the general unevenness. CONSTITUTION:When the sine-wave type exciting signals with a phase difference by 90 deg. mutually are supplied to the 1st - the 4th detection coils 81-84, the magnetic flux generated by the exciting signal according to an opposing positional relation between the detection coils 81-84 and the uneven graduations of the scale body 70 is decreased and as a result, the impedance of each of the detection coils 81-84 is changed. Then, the output corresponding to the impedance taken out from each of the detection coils 81-84 is added by an adder 86 to obtain positional information from the output (e). In this case, since the graduations of the scale body 70 are constituted by using the general unevenness, the production is made simple.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a linear or rotational position detector;
For example, it is used to extract print head position information in a print head position control system of an electronic printer, or to detect the spindle position of a measuring instrument that presses the spindle tip against a measured object and detects its displacement.

Conventional technology This type of position control system is increasingly required to be more accurate, faster, and simpler, and along with this, position detectors are also required to have higher resolution, responsiveness, and simpler structure. has been done.

Those that meet these conditions include the inductor F-sin type and magnetic scale type disclosed in the document ``Automation of Measurements You Want to Know'' (Published by Japan Machinist Co., October 1, 1976), 137-176. There is a so-called phase modulation type position detector.

As shown in FIG. 4, the former supports a slider body 20 so as to be movable in the direction of the operator while facing a long scale body 10, and the scale body 10 has a rectangular wave-like zigzag graduation coil. 11, and the slider body 20 has a first. forming a second excitation filter 21.22;
An excitation signal sin with a phase difference of 90 degrees is applied to the coils 21 and 22.
ωt, cosωt (ω: angular frequency) are supplied for two hours to excite the coil.
1. Both coils 11 and 21 according to the opposing positional relationship with 22.
．． The degree of electromagnetic coupling between 11 and 22 changes; the degree of coupling is highest when both coils are completely polymerized, and lowest when both coils are not polymerized, and changes by 1 in a sinusoidal waveform between them. Therefore, now, with the above-mentioned complete polymerization state as a reference, the amount of movement χ of the coil 21 and 22 relative to the coil 111 from there is defined as the angular dimension θ (=2π・χ/ P) and using its θ to represent the voltage induced in the coil 11 by electromagnetic induction when the coils 11 and 22 are open, let A be the coefficient, then A cos θ sin ωL,
A sin f9 cos ωt, and in the end, an output of A 5in (ω is 10 θ), which is the sum thereof, is taken out from the coil 11, and as a result, position information θ is obtained every 2π/ω.

The latter magnetic scale type includes a magnetic scale body 30 in which magnetic scales in which the strength of magnetic flux changes in a sine wave pattern are arranged at minute pitches, and only 1/4 pitch of the magnetic scales as shown in FIG. A group of heads 41 in which first and second magnetic flux responsive heads 41.42 are arranged offset from each other is opposed to the head group 41, and the excitation coils of the first and second magnetic flux responsive heads 41.4.2 are arranged at 90°. Each detection coil is excited by an excitation signal having a degree phase difference, and the sum of the outputs taken out from each detection coil is calculated. In this case, the amplitude of the output (the frequency is the same as the excitation signal) taken out from each detection coil according to the positional relationship between the magnetic scale and each head 4, 1, 42 is a sine wave (the amplitude is also (with a phase shift of 90 degrees), and the sum thereof becomes a signal having phase information corresponding to the position, similar to the indatocin type.

When using this type of position detector, for example, to retrieve position information of a print head, one side of the position detector facing the motor, for example, a linear pulse smoke movable element, which is a drive source of the print head, is connected to one side of the position detector, for example, a slider. The body 20 is fixed, and the scale body 10 is placed on the stationary body [
After this is fixed, the predetermined origin position is electrically (or mechanically) calibrated and then used for position detection.

Problems to be Solved by the Invention The above-mentioned position detector is of a phase modulation type, and its resolution and responsiveness almost meet the requirements.

=4- However, there is still a gap between requirements in terms of structural simplicity. That is, when using a position detector, the slider body or head group is integrally fixed to the moving body as described above, and the scale body is placed opposite to it along the path of the moving body. These scale bodies are special with fill or magnetic scales, and cannot be installed directly on the drive source of a moving object, such as the piston shaft of a hydraulic cylinder or the secondary side (fixed side) of a linear pulse motor. This is difficult both structurally and in terms of magnetic noise. For this reason, methods such as placing them in parallel with the secondary side with a gap between them, or in series with the secondary side are taken, which leads to the problem that the overall system configuration becomes complicated and large. was there.

Furthermore, the fill and magnetic graduations of such a scale body require special equipment and techniques to form, and even with that, there is a problem in that it is difficult to form long pieces with high precision.

Therefore, in order to solve the above-mentioned problems, the present invention does not require a special scale, which is easy to manufacture, and also provides a position where the scale can be formed on the moving path itself and used as a scale body. The aim is to provide a detector.

Means for Solving the Problems The present invention examines various types of graduations for the scale body.
As a result of experiments, we found that when a detection coil made by connecting rectangular coils with the same pitch in a zigzag pattern was placed opposite a general concave-convex scale, and an excitation signal was supplied to the coil, the impedance of the fill changed. However, it was completed based on the knowledge that the change becomes sinusoidal depending on the opposing position of the concavo-convex scale and the coil. That is, as shown in FIGS. 3(a) and 3(b), the uneven scale 5o
When the rectangular coil 60 is made to face each other and an alternating current that changes sinusoidally is supplied as an excitation signal, magnetic flux is generated inside and outside the wire forming the fill 50. As a result, an eddy current is generated in the uneven scale due to the magnetic flux, and this eddy current serves to prevent the generation of magnetic flux due to the alternating current.

The height of the eddy current varies depending on the position of the wire and the convex part of the uneven scale, and is greatest in the state (a) facing the convex part,
It is smallest in the state (b) facing the recess. As a result, as the opposing positional relationship between the wire and the convex portion changes from the state (b) to the state (,), the degree to which generation of magnetic flux is obstructed increases, and the magnetic flux decreases accordingly. When the magnetic flux changes in this way, the inductance of the coil changes accordingly, and eventually the impedance of the coil changes. This change in impedance changes sinusoidally around a constant value for each uneven scale pitch.

The present invention has been devised based on the above knowledge, and includes a position detector in which a scale body and a detection head face each other and detect the relative position of the two.
A concavo-convex scale is formed in the detection head, and the first to fourth detection filters, which are rectangular coils having the same pitch P as the previous concave-convex scale, are connected in a zigzag pattern to each other (n+1/4)P[
However, n is Oll, ... integer] and arranged,
A sinusoidal excitation signal with a position difference of 90 degrees is supplied to the first to fourth detection coils from the oscillator, and an electric signal that changes in accordance with the impedance is extracted from the first to fourth detection coils. The sum is calculated by an arithmetic unit.

In the case of more than 100% fertilization, excitation signals sinωt, cosωt, - with a phase difference of 90 degrees are applied to the first to fourth detection filters.
When sinωt and -cosωt are supplied, the magnetic flux generated by the excitation signal is reduced according to the opposing positional relationship between the detection coil and the uneven scale, and as a result,
The impedance of each detection coil changes. Now, the state in which the first detection coil is located in the state shown in FIG. Detection coil 4 moves leftward from that position to 1/4 to 3/3 of the unevenness pitch.
4), and let the amount of movement of the concavo-convex scale to the left from that position be χ, and that χ is the concave-convex scale pitch P.
Expressed as θ (=2π・χ/P) using the angular dimension with 2π, the impedance of each detection coil is
, C, respectively, B (C+ cosθ)sir
+ωt, B (C+sinθ)eO3ωt, B(C-
cos θ) (-sin ωt), B(C-sin θ) (-
cosωt). Therefore, the sum e of the outputs corresponding to the impedances taken out from each detection coil calculated by the arithmetic unit is given by K, which is the conversion coefficient between the impedance and the output corresponding to the impedance. θ) (
1), and the relative movement position χ between the concavo-convex scale and the detection coil of the detection head is calculated for each period of the excitation signal as phase information θ proportional to the relative movement position χ.

Therefore, the uneven scale can be formed using only general machine tools and techniques, and the uneven scale can be formed along the path of the moving object whose position is to be detected.
), and even its secondary side like linear pulse smoke (
If the scale has a concavo-convex scale on the fixed side (this is part of the drive mechanism), the concave-convex scale can be used as it is as a scale body.

Example Hereinafter, the present invention will be explained based on examples.

In FIG. 1, 70 is a long scale body, on which rectangular concave and convex scales are formed at equal pitches P in the longitudinal direction.

Detection head 8 facing the concavo-convex scale with a minute interval
0, there are first to fourth detection coils 81 in the form of three rectangular coils connected in a zigzag pattern with the same pitch P as the concavo-convex scale.
=84 or they are arranged so as to be shifted from each other by 1 (1+1/4)P.

FIG. 2 shows an oscillator 85 that supplies excitation signals to the first to fourth detection coils 81 to 84 and the detection coil 8 caused by the oscillation unit 85.
This is an example of an arithmetic unit that adds coil outputs that change in response to impedance changes of 1 to 84. Excitation signal sinωt of the oscillator 85, −sinωt, C03ω15,
- The output terminals of QO3ω are connected to each corresponding first,
Third, second and fourth detection filters 81.83.82.84
are connected to the input terminals of the respective coils 81.83.8.
2.84 shows impedance change B(C+cosθ)si
nωt, B(C-cosθ)(-sinωI), B(
C + sin θ) cos ωt, B (C sin θ) (-c
osωt). The first and third coils 81
．． 82, the pairs of second and fourth coils 82 and 84 are connected to add each other, and an output corresponding to the sum of impedance changes for each month is taken out, and the output of each phase is further connected to an adder 8.
6, and the output e shown in equation (1) above is calculated.

This type of position detector is described, for example, in the literature [Linear Motor and Its Applications] (IEE of Japan, published March 30, 1980).
When used to extract positional information of a print head of an electronic printer equipped with a linear pulse motor as disclosed in pages 98 to lot of The detection head 80 of the position detector is fixed to a mounting member 92 that protrudes in the movable direction from the movable element 91 of the motor 90, and is opposed to the scale body 70 which also serves as the stator of the linear pulse motor 90.

After electrically (or mechanically) calibrating the predetermined origin position, the position is used for position detection.

The same effect can be obtained even if the scale body 70 is arranged separately from the secondary side (fixed side) of the linear pulse motor and the detection head 80 is opposed to it.

In addition, when used as a size measuring instrument, a concavo-convex scale is formed on the spindle to make the spindle itself the scale body 70, and the detection head is fixed to the casing that supports the spindle so that they face each other. Become.

As an example, the unevenness scale pitch of the scale body 70 is 8 mm5.
Height 4mm+, material low carbon steel, number of pitches of the detection fill of the detection head 80 is 24, wire diameter 0.3+nm, gap between the uneven scale and the fill 0.5mm, voltage of excitation signal 20p
-p (series resistance 1 to Ω) and the frequency is 10 MHz, the amount of change in output due to change in impedance of one coil is 22.5 m V p -p, which is sufficiently large.

In the above embodiment, a rectangular concavo-convex scale is used, but a substantially sinusoidal impedance change can be obtained even when the concavo-convex scale is trapezoidal or triangular.
This shape may also be adopted. Further, in the embodiment 1, the case where the first to fourth coils are formed by zigzag connection of three rectangular fills is illustrated, but an appropriate number may be selected; Instead of being arranged in series, the fourth detection coil may be arranged in a stacked configuration to reduce the length of the detection head.

In addition, although the above embodiment is for linear position detection, the scale body is a disk, and concave and convex graduations with equal pitch angles are formed on the end face or circumferential surface of the scale body, and the detection head is opposed to the concave and convex graduations, and the rotating position is It may also be configured for detection.

The effects of the invention are as described above, and since the present invention uses general unevenness for the graduations of the scale body, it is easy to manufacture, there are no restrictions on installation, and the movement path of the object to be detected can be directly monitored. The structure can be simplified, and the overall size can be reduced.

[Brief explanation of drawings]

Fig. 1 is a perspective view showing an embodiment of the mechanical part of the present invention;
The figure is a block diagram showing the excitation signal supply and signal processing portion of the present invention, Figure 3 is a model diagram for explaining the principle of the present invention, and Figures 4 and 5 are perspective and front views showing the prior art, respectively. It is a diagram. 70 Niskel integrated body 80: Detection head 81-8
4: Detection coil 85: Oscillator 86: Adder

Claims (1)

[Claims] 1. In a position detector in which a scale body and a detection head face each other and detect the relative position of the two, a concave-convex scale is formed on the scale body at an equal pitch P, and a rectangular scale with the same pitch P as the concave-convex scale is formed on the detection head. The first to fourth detection coils, which are formed by connecting suitably shaped coils in a zigzag pattern, are connected to each other (n+1/4
) P [where n is 0, 1, ... integer] are arranged and fixed in a staggered manner, and the first to fourth detection coils are supplied with sinusoidal excitation signals with a phase difference of 90 degrees from the oscillator. 1st~
A position detector in which a calculation unit calculates the sum of outputs corresponding to the impedance of a fourth detection coil.