JPH09264761A - Position detecting apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to accurately read by simplifying the structure of a head in an apparatus for reading an absolute code recorded on a code plate by using a head having an MR (magnetoresistance effect) element. SOLUTION: This position detecting apparatus comprises an absolute track having an absolute code formed in combination of logic values '0' and '1' so that either '0' or '1' is a magnetized part and the other is a nonmagnetic part, a code plate having an incremental track together with the absolute track, and a position detector having a head for reading the absolute track and a head for reading the incremental track and provided movably in the longitudinal direction of the absolute and incremental tracks. In this case, detecting heads Hn ((n) is output bit number) for reading the absolute track are each formed of three magnetoresistance elements sn1 to sn3 equivalently isolated at each element interval length.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position detecting device used for detecting an absolute position of a linear or rotary moving object, and to a position detecting device which outputs a stable absolute value signal.

[0002]

2. Description of the Related Art Encoders as shown in FIGS. 12 to 14 have been developed as position detecting devices for linear movement and rotational movement. Each of these encoders has a scale and a detection head for reading the scale.

As a code recorded on the scale, a magnetic absolute encoder having a one-track type absolute pattern to which a non-repetitive code is applied has already been proposed. Regarding these encoders, for example, Japanese Patent Laid-Open Nos. 1-79619 and 1-3112.
No. 21, which is incorporated herein by reference, refer to them.

Here, only the necessary portions related to the present invention will be briefly described. An absolute encoder is composed of a scale and a sensor that reads the scale. The scale is a track in which a code is magnetically recorded, and the scale and the sensor move relatively along the track. Then the code is read.

FIG. 6 shows a scale of a one-track ABS encoder and sensors S1, S2, S3 and S4. In this case, four sensors are provided and these are arranged in series in the longitudinal direction of the track. It moves as a unit on the scale and outputs a 4-bit output. In the figure, the blank portion on the scale is a non-magnetized portion, and S
The small section represented by N is a magnetized portion.

In FIG. 6, a non-magnetized portion is associated with a logic "0" and a magnetized portion is associated with a logic "1", and the scale corresponds to the arrangement of the non-magnetized portion and the magnetized portion on the scale. The corresponding logical value is shown above. The output of each sensor as it is moved along this scale is shown below the scale.

The value obtained by combining the outputs of the four sensors S1, S2, S3, S4, that is, the value obtained by reading the sensor output of FIG. 6 in the vertical direction is the absolute position of the sensor on the scale,
As indicated by the hexadecimal code corresponding to the bottom of the figure, the same number sequence is not repeated.

The basic structure of the code recorded on the scale is a code using a non-repeating pattern such as M series as shown in FIG. It should be noted that FIG. 6 exemplifies cases of 5 bits, 6 bits, 8 bits, and 10 bits.

When forming an absolute track using one of these codes, the patterns "0" and "1"
One of the two is magnetized and the other is not magnetized, and the absolute pattern is recorded by making 1 bit of the recording portion correspond to the alternating magnetism of wavelength λ. Further, as will be described later, an incremental track recorded by alternating magnetism with a wavelength of 2λ may be provided in parallel with the absolute track so as to synchronize with the absolute pattern and achieve high resolution.

Since the encoder is often stationary in the environment of use, a pattern detector that can detect a static magnetic field is used. In particular, in the case of a rotary encoder, since it rotates at a high speed, an MR element (magnetoresistive effect element) or the like that can detect in a non-contact manner is generally used.

Therefore, an absolute pattern reading using this MR element sensor will be briefly described with reference to FIGS. 8 and 9.

The absolute pattern (1-
On a), there are provided four sensors S (S1, S2, S3, S4) for each bit interval λ, each sensor S consisting of a pair of MR elements (MR1 and MR2) whose element intervals are λ / 4 apart. Of (1
On the incremental pattern shown in -b), two sensors S5 and S6 each consisting of a pair of MR elements each having an element spacing of λ / 2 are provided at an interval (7/4) λ.

First, when the reproduction output when the absolute track of FIG. 8 is reproduced by one MR element sensor (MR1) is examined, the output waveform shown in the uppermost stage of FIG. 9 is obtained. That is,
The reproduced waveform is a waveform divided at the magnetized portion just at λ / 2.

In order to eliminate this division, the sensor MR1
And another sensor MR2 separated by a distance λ / 4,
Play the same absolute track. At this time, the reproduction output of the sensor MR2 is MR1 as shown in the second row of FIG.
The output waveform has a phase difference of λ / 4 from the reproduction output of.

Therefore, the outputs of these sensors MR1 and MR
When the two outputs are combined by a bridge circuit as shown in FIG. 15, it is possible to obtain an output waveform excellent in a rectangular shape without division as shown in the third stage of FIG. By cutting this output waveform further at a predetermined threshold level, a clean binary signal as shown at the bottom of the figure is obtained.

Therefore, the output waveform obtained from each of the sensors S1 to S4 composed of a pair of MR elements is shaped by a waveform shaper or the like, and four outputs are combined to obtain a 4-bit absolute code. You can

Each output when the incremental track is reproduced by the sensors S5 and S6 is exactly 90.
The signals become sine and cosine signals having a period λ with a phase difference of °, and an absolute section absolute value signal can be obtained within the section of wavelength λ by these two signals.

An example thereof is shown in FIG. Here, four absolute value signals (11, 10, 00, 01) are obtained by dividing a sine wave and a cosine wave in one section by applying a threshold when their output levels are 0. ,
Similarly, if the output level is divided into m (m is a positive integer), 2
It is possible to obtain m absolute values.

By combining the absolute code obtained from the above absolute track and the absolute value signal in the section obtained from the incremental track,
It is possible to obtain a highly accurate absolute signal over the entire area.

[0020]

The absolute encoder as described above provides an absolute output during normal use, but the MR element sensor group on the absolute track receives the track pattern when the power is turned on at the time of power failure or start-up. When located at the boundary between the magnetized portion and the non-magnetized portion, the output of the sensor becomes unstable.

Since the non-repeating codes are randomly arranged due to their nature, even if one bit output is different, the absolute code obtained is completely different from the code at the required position, which is a mistake. There is a risk of providing location information.

In order to avoid this reading error, two sets of heads (sensors) having outputs with different phases are generally used, and when one head is at the boundary, the output is obtained from the other head. As described above, two sets of absolute output signals are alternately read to correct the boundary portion.

However, in the case of the magnetic absolute encoder, as shown in FIG. 11, two sensors each consisting of a pair of MR elements must be used for the output of each bit, and in addition to that, two sets of heads are used. If the boundary correction of
Since a total of 4 MR sensors are required for 1 bit, increasing the number of output bits causes the detector to become larger and the wiring to become complicated, and solving these problems is an important issue in the production of absolute encoders. It was.

[0024]

In order to solve the above-mentioned problems, according to the present invention, an absolute code "0" or "1" which is a combination of logical values "0" and "1" is used.
A track having an absolute pattern, one of which is a magnetized portion and the other a non-magnetized portion, and an alternating magnetism of a predetermined wavelength, which is provided along with the track having the absolute pattern, is recorded in the longitudinal direction. A position where a code plate having a track having a mental pattern, a track having the absolute pattern and a track having the incremental pattern are relatively movable in the longitudinal direction, and a head for detecting the absolute pattern and the incremental pattern is held together. A position detector is constituted by the detector.

Then, a track having the above absolute pattern (hereinafter referred to as an absolute track)
Head H for each bit of the position detector for detecting
n (n is an output bit number) is composed of three MR elements Sn1, Sn2 and Sn3 which are evenly spaced by the element interval length L, and two absolute signals with a phase difference L are taken out from the detection head Hn and are incremented. A track with a mental pattern (hereinafter referred to as an incremental track)
The section absolute value signal obtained by the above is almost the central position of the portion sandwiched between the displacement position of the bit of one absolute signal and the bit change position of the other adjacent absolute signal that changes in bit with a phase difference L. That is, the MR sensor or the pattern is arranged so that the two absolute signals are switched and selected at the center L / 2 of the phase difference between the two absolute signals, and synchronization is established.

In the position detecting device according to the present invention, one head is composed of two sets of sensors consisting of a pair of MR elements, and one of the two MR elements constituting each sensor is composed of two sets of sensors. It is formed so as to be shared so that the same boundary correction as the conventional type can be obtained with three MR elements.

More specifically, when one sensor, for example, the bridge output between Sn1 and Sn2 is an unstable signal output at the boundary, the other sensor Sn2- having a phase difference of the element interval L. Since the bridge output of Sn3 outputs a stable signal, an absolute signal selector that switches the sensor according to the absolute position signal obtained from the incremental part is provided, and it is only L / 2 from the absolute signal boundary. If there is a phase difference, the sensor output Sn2 of the absolute head is shared and Sn1 and Sn3 are synchronized so as to switch to a sensor with stable output, and if the bridge output is selected, it is synchronized with the section absolute value signal. , Absolute output signal is obtained without reading the boundary.

As yet another specific example of the present invention, FIG. 1 shows the overall structure of a position detecting device for outputting a 4-bit code.
FIG. 2 shows a head pattern showing an example of the detection head used in the detection apparatus.

Here, the apparatus of FIG. 1 will be described together with its operation. Although the heads H1 to H4 are shown in FIG.
Since these heads operate in the same manner, the head H1 will be described here as an example.

Among the signals obtained from the three MR elements of the head H1, the signals from the elements S11 and S13 are sent to the absolute signal selection circuit. This signal selection circuit is composed of an analog switch or the like, and sends one of the signals from the elements S11 and S13 selected by the selection signal as shown in FIG. 10 to the bridge circuit, where it is supplied from the element S12. Then, it takes the bridge output with the signal whose phase is inverted by the anti-phase shifter.

The output of this bridge circuit is supplied to a differential amplifier, where it is differentially amplified and becomes an absolute signal whose waveform is shaped through a comparator. The other heads are similarly processed to take out a 4-bit absolute signal.

Since the non-repetitive code thus obtained is a random array code, this signal is sent to a signal conversion circuit composed of a ROM or the like, and a binary (binary) or decimal (decimal) code or the like is requested there. Convert to signal. By combining the output of this signal conversion circuit and the section absolute value signal, it is possible to obtain an absolute value signal over the entire area as in the conventional type.

As described above with reference to FIG. 10, the sine wave signal and the cosine wave signal are obtained from the heads H5 and H6, and these are supplied to the section absolute value signal generating circuit to generate the section absolute value signal. . Further, a selection signal is supplied from the section absolute value signal generation circuit to the ABS signal selection circuit.

[0034]

BEST MODE FOR CARRYING OUT THE INVENTION Next, an embodiment of the present invention will be described with reference to FIGS. FIG. 2 conceptually shows an MR pattern of a position detector in which heads H1 to H4 for reading an absolute pattern and heads H5 and H6 for reading an incremental pattern are formed.

Each absolute pattern reading head is formed of three MR elements, one end of which is connected to a common bias power source. The incremental pattern reading head comprises a sine wave output head H5 and a cosine wave output head H6.
Has a sensor unit for sine wave output and inverted sine wave output composed of two MR elements, the common connection point thereof is grounded, and both ends thereof are connected to the bias power supply. Similarly,
The cosine wave output head H6 has a sensor part for cosine wave output and an inverted cosine wave output composed of two MR elements, the common connection point thereof is grounded, and both ends thereof are connected to the bias power supply.

FIG. 3 shows an example of a position detecting device using the position detector shown in FIG. As shown in the figure, in the code plate of the present invention, each bit interval is λ, “0” is a non-magnetized portion,
An absolute track composed of an absolute pattern (ABS) composed of "1" as a magnetized portion and an incremental track composed of an incremental pattern (INC) recorded by alternating magnetism with a wavelength of 2λ are provided side by side. , Four heads H1 to H4 are arranged on the absolute pattern with a distance λ between them.

On a track having an incremental pattern (INC), heads H5 and H6 composed of a pair of MR elements having an element interval of λ / 2 are arranged at an interval of (7/4) λ.

The above-mentioned heads H1 to H4 output a 4-bit absolute code when moving along the absolute track, but only the head H1 will be described here for the sake of simplicity.

This head H1 has three MR elements S1.
1, S12, S13, S11 and S12 constitute one sensor (S11-S12), and S12 and S13 constitute another sensor (S12-S13).
Two elements S11, S12 of the sensor (S11-S12)
The distance between them is λ / 4, and the sensor (S12-S13)
The distance between the two elements S12 and S13 is also λ / 4. Furthermore, the gap between these two sensors is also λ / 4.
It is.

The sensors (S11-S12) are the sensors (S1
2-S13), the output of the sensor (S11-S12) is the sensor (S12-S13).
The phase is delayed by λ / 4 from the output of. This state is shown in the third and fourth stages of the figure with ON / OFF waveform diagrams.

On the other hand, the incremental track is read by the two heads H5 and H6, and from the outputs of them, 5 in FIG.
A selection signal as shown in the stage is generated, and the two sensors (S11-S12) and (S12-
The detection output at the stable point is extracted by sampling one of the outputs of S13).

By synthesizing the output signals thus extracted, an accurate absolute signal synchronized with the incremental signal as shown at the bottom of the figure can be obtained. In addition,
The symbols a and b entered in the waveforms shown in the fifth row of FIG. 3 indicate which sensor is selected.

Next, another embodiment of the present invention will be described with reference to FIG. In this embodiment, the magnetized portion of the absolute track in the above embodiment is improved. Specifically,
Magnetization part A part where 1 bit exists independently (for example, 0,
(1 portion of 1, 0) is recorded by alternating magnetism of a wavelength obtained by subtracting (3/4) λ, that is, the element interval length λ / 4.
Further, for example, for a portion where the number of magnetized bits is continuous, for example, (3/4) λ for one bit of the continuous portion.
Recording with alternating magnetic field (remaining part is recorded with wavelength λ) or range of continuous wavelength less element spacing length = recording wavelength = {1- (1 / 4n)} λ (n is a continuous bit Record with alternating magnetism of (number). For example, in the case where the magnetized portion for 4 bits is continuous, one bit length is calculated as n = 4 in the above equation, and recording wavelength = (15/1
6) It becomes λ.

When the pattern thus recorded is reproduced by the detector, it is possible to obtain an absolute signal having a uniform pitch width and no division. According to this method, the absolute code obtained from each sensor is a more accurate code as compared with the case of the first embodiment, and therefore the reliability of correction when the present invention is applied is improved.

FIG. 5 shows still another embodiment of the present invention. In this embodiment, it is composed of an absolute track composed of four tracks and one incremental track.

Next, with reference to FIG. 5, as still another embodiment of the present invention, a case where the above-described technical idea of switching the head of the present invention is applied to a multi-track type absolute pattern will be described. .

In FIG. 5, track T1 is the least significant bit of the binary code, track T2 is the second digit of the binary, track T
Reference numeral 3 indicates a binary third digit, and track T4 indicates a binary fourth digit, which are weighted by 0, 1, 2, 3, and 2, respectively.

The interval between the MR elements is λ / 4, and since the three MR elements forming two sensors are lined up continuously, the total interval is (1/4) λ × 2 = λ / 2. Therefore, 1 bit of the track T1 is represented by λ. The length of the track T2 is 2λ, the length of the track T3 is 4λ, and the length of the track T4 is 8λ.

Looking at the code sequence of the natural binary code shown here, there is a part where two or more bits are changed in any of the preceding and following codes, so that the phase of the read output of the magnetizing part is changed. There is a possibility that a misalignment will occur and a read error will occur at the code boundary, as in the one-track type.

However, by applying the head switching of the present invention to the extraction of the output signal at the stable point, the reliability is improved and an excellent absolute encoder without a read error can be realized.

[0051]

According to the present invention, since the number of MR elements of the read head in the magnetic absolute encoder can be reduced, the structure of the encoder can be improved as compared with the conventional one.
Easy to make. Further, by switching and using the outputs from the two sensors arranged with a predetermined gap, the reliability of the extracted signal is improved, and an excellent absolute encoder without reading errors can be realized.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a position detection device of the present invention.

FIG. 2 is a schematic diagram of an MR pattern applied to a detection head used in the position detection device of the present invention.

FIG. 3 is a configuration diagram of an absolute encoder.

FIG. 4 is a configuration diagram of an absolute encoder.

FIG. 5 is a configuration diagram of an absolute encoder.

FIG. 6 is a waveform diagram of outputs and absolute codes obtained from four sets of MR elements.

FIG. 7 is an explanatory diagram showing an example of a 5- to 10-bit absolute code.

FIG. 8 is a configuration diagram of a conventional 4-track code output 1-track type absolute pattern.

FIG. 9 is an output waveform diagram of a sensor using two conventional MR elements.

FIG. 10 is an output waveform diagram of an incremental track.

FIG. 11 is a layout diagram of tracks and heads of a one-track type absolute encoder.

FIG. 12 is an external view of an encoder.

FIG. 13 is an external view of an encoder.

FIG. 14 is an external view of an encoder.

FIG. 15 is a bridge circuit for absolute signals.

[Explanation of symbols]

H1 to H4 head, ABS absolute signal,
INC Incremental signal, ROM Read only memory

Claims (2)

[Claims] 1. An absolute code in which one of "0" and "1" of an absolute code composed of a combination of logical values "0" and "1" is formed as a magnetized portion and the other is formed as a non-magnetized portion. A track, a code plate provided with the absolute track and provided with an incremental track, and a position detector including a head for reading the absolute track and a detection head for reading the incremental track. The position detector is a position detecting device provided so as to be relatively movable in the longitudinal direction of the absolute track and the incremental track of the code plate, and a detection head Hn for reading the absolute track.
Each of (n is an output bit number) has three magnetoresistive effect elements Sn1, Sn2, which are evenly spaced by the element interval length L.
Two absolute signals having a phase difference L are extracted from the detection head Hn, and at the center L / 2 of the phase difference between the two absolute signals, a selection signal obtained from the incremental track 2 A position detection device having a detection head and a selection function for switching and selecting two absolute signals. 2. The position detecting device according to claim 1, wherein an interval L between elements of the three magnetoresistive elements is:
A position detecting device characterized in that when the minimum resolution length of the absolute signal is λ, it is given by L = λ / 4m (m = 1, 2, 3, ...).