JPH0462004B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a rotary encoder, and more particularly, a radiation grating in which a plurality of grid patterns of transparent parts and reflective parts are periodically carved on the circumference is attached to a rotating object, and the radiation grating is attached to the radiation grating. For example, the present invention relates to a rotary encoder that irradiates a beam of light from a laser and uses diffracted light from the radiation grating to photoelectrically detect the rotation angle of a radiation grating or a rotating object.

Conventionally, computer equipment such as driving floppy desks, office equipment such as printers, or
Rotary encoders have been used as a means to detect the rotation angle of rotating mechanisms such as NC machine tools and VTR capsten motors and rotating drums.

The method using a photoelectric rotary encoder is to create a so-called main scale in which transparent parts and light-shielding parts are provided at equal intervals around a disc connected to the rotation axis, and corresponding transparent parts are provided at equal intervals to the main scale. A so-called fixed index scale having a light-shielding section and a light-shielding section are arranged in a so-called index scale system in which both scales are placed facing each other with a light emitting means and a light receiving means sandwiched between the two scales. In this method, as the main scale rotates, a signal is obtained that is synchronized with the interval between the light-transmitting part and the light-blocking part of both scales, and the rotation angle is detected by integrating this signal after waveform shaping.

In a rotary encoder, the finer the scale interval between the light-transmitting part and the light-blocking part of both scales, the higher the detection accuracy can be. However, when the scale interval is made finer, the S/N ratio of the output signal from the light receiving means decreases due to the influence of the diffracted light, resulting in a decrease in detection accuracy. For this reason, it is conceivable to fix the total number of gratings in the light-transmitting part and the light-blocking part of the main scale, and widening the interval between the light-transmitting part and the light-blocking part to the extent that it is not affected by the diffracted light.
However, this has the disadvantage that the diameter and thickness of the main scale disc increases, making the entire device larger, and as a result, the load on the rotating object to be tested increases.

SUMMARY OF THE INVENTION An object of the present invention is to provide a small rotary encoder that reduces the load on a rotating object to be tested, reduces the influence of eccentricity of attachment to the rotating object, and is capable of detecting a rotation angle with high precision.

The main features of the rotary encoder for achieving the object of the present invention are a radiation grating in which a plurality of grid patterns are arranged at equal angles around the circumference of a disk, a rotating object connected to the radiation grating, and a luminous flux incident on the radiation grating. of the diffracted light from the luminous flux incident on the radiation grating, and
at least three reflecting mirrors for re-injecting the two diffracted lights at positions approximately symmetrical to the center of rotation of the rotating object with respect to the incident position on the radiation grating of the light beam by the first illumination means; After superimposing two diffracted lights of a specific order diffracted again by the second illumination means having two sets of relay optical systems consisting of an imaging optical system and the radiation grating, the superimposed light beam is received. The rotation angle of the rotating object is determined using an output signal from the light receiving means.

Next, one embodiment of the present invention will be described with reference to each drawing.

FIG. 1 is a schematic diagram of an embodiment of the present invention. In the figure, 1 is a light source emitting a single wavelength such as a laser, 2 is a collimator lens, 3 and 16 are cylindrical lenses, and 4, 5, 6, 9, 1
0 and 11 are reflecting mirrors, 7 and 12 are imaging optical systems, and reflecting mirrors 4, 5, 6 and the imaging optical system 7 form the first relay optical system, and reflecting mirrors 9, 10, and 11 form the image forming system. The optical system 12 forms a second relay optical system.
Reference numerals 8 and 13 denote quarter-wave plates whose axes are arranged at 45 degrees and -45 degrees with respect to the linearly polarized light from the laser 1, respectively. The two relay optical systems intersect between the optical path between the reflecting mirror 4 and the imaging optical system 7 and the optical path between the reflecting mirror 9 and the imaging optical system 12. Reference numeral 14 denotes a radiation grating in which a grid pattern of transparent parts and reflective parts is provided at equal angles on a disk, for example. Reference numeral 15 denotes a rotation axis of the radiation grating 14, which is connected to the rotation axis of the rotating object to be tested. A beam splitter 17 has semi-transparent surfaces 18, 19, and 20.
Polarizing plates 21, 22, 23, and 24 are arranged with their polarization directions shifted by 45 degrees. 25, 26, 27, and 28 are light receiving elements, respectively.

The light beam emitted from the laser 1 is turned into a substantially parallel light beam by the collimator lens 2, and is moved to a position _M1 on the radiation grating ₁₄ by the cylindrical lens 31.
is irradiated in a linear manner. By irradiating in a linear manner in this manner, it is possible to reduce the pitch error in the grid pattern of the light transmitting portion and the reflecting portion corresponding to the irradiated portion of the light beam on the radiation grating 14.

Incidentally, instead of the cylindrical lens, a slit or a lens and a slit may be used for linear irradiation.

The light beam from the laser 1 is reflected and diffracted by the grating pattern of the radiation grating 14. The current irradiation position of the luminous flux
If the pitch of the checkered pattern in M ₁ is p, ±m
The diffraction angle θ _n of the next reflected diffraction lights L ₁ and L ₂ is expressed as sin θ _n =±mλ/p (1). Here, λ is the wavelength of the luminous flux.

Assume now that the radiation grating 14 is rotating at an angular velocity ω. If the distance from the rotation center of the radiation grating 14 to the irradiation position M ₁ is r, the circumferential velocity at the irradiation point M ₁ is v=rω. At this time, the frequency of the ±m-order reflected diffraction light undergoes a so-called Doppler shift by an amount expressed by the following equation.

Δ=±vsinθ _n /λ=±rωsinθ _n /λ ...(2) These two diffracted lights are each transmitted to the radiation grating 14 by the second illumination means consisting of the first relay optical system and the second relay optical system. An incident position approximately symmetrical to the center of rotation of the rotating shaft 15 with respect to the above incident position _M1
re-enters _M2 . Here, the 1/4 wavelength plates 8 and 13 are arranged so that their respective axes are at 45 degrees and -45 degrees with respect to the linear polarization direction of the incident light beam, respectively.

Further, the angle of incidence on the irradiation position M ₂ is set so that each diffracted light is equal to the reflected diffraction angle θ _n at the irradiation position M ₁ and the angle with the circumferential velocity direction of the radiation grating 14 is also equal. First and second relay optical systems are arranged.

Then, at the irradiation position M ₂ , the reflected diffraction beams of order ±m overlap, and the cylindrical lens 1
6, it becomes a parallel beam of light again, and is split into four beams by the beam splitter 17.
, and enters the light receiving elements 25 to 28.

The overlapping ±m-order diffracted lights reflected at the irradiation position M ₂ are again reflected as the radiation grating 14 rotates.
Since it is subject to the Doppler frequency shift Δ in equation (2),
In addition to the frequency shift when reflected at the irradiation position M ₁ , the frequency shift amount of the ±m-order diffracted light reflected at the irradiation position M ₂ becomes ±2Δ. In this way, since the light that has undergone ±m-order diffraction twice overlaps, the frequency of the output signal from the light receiving elements 25 to 28 becomes 2Δ−(−2Δ)=4Δ. In other words, the frequency of the output signals of the light receiving elements 25 to 28 is =
4Δ=4rωsinθ _n /λ, and from the diffraction condition equation (1), it becomes =4mrω/p.

If the total number of grid patterns of the radiation grating 14 is N and the equiangular pitch is Δ, then from p=rΔ and Δ=2π/N, =2mNω/π (3). Now, if the wave number of the output signal of the light receiving element during time Δt is n, and the rotation angle of the radiation grating 14 during Δt is θ, then from n=Δt and θ=ωΔt, n=2mNθ/π... (4) By counting the wave number of the output signal waveform of the light receiving element, the rotation angle θ of the radiation grating 14 can be determined.
can be determined using equation (4).

By the way, it is more preferable if the direction of rotation can be detected when detecting the rotation angle. Therefore, in this embodiment, as is well known in conventional photoelectric rotary encoders, a plurality of light receiving elements are prepared and arranged so that the phases of their signals are shifted by 90 degrees.
A method is used to extract a signal indicating the direction of rotation from a 90° phase difference signal that accompanies rotation.

In addition, due to errors in line width between the transparent part and the reflective part of the radiation grating 14, or fluctuations in laser output, etc.
The center level of the output signal of the light receiving element may fluctuate. Therefore, in this embodiment, in order to suppress this fluctuation, keep the center level constant, and stabilize the subsequent signal processing, we take the differential of two output signals with a 180° phase difference and remove the DC component. It uses the so-called push-pull method.

In this way, in this embodiment, in order to detect the rotation direction and at the same time keep the center level of the signal constant,
Four phase difference signals of 0°, 90°, 180°, and 270° are used as output signals of the light receiving element. In the configuration of this embodiment, these four phase difference signals are generated by a combination of linearly polarized laser light, two quarter-wave plates 8 and 13, and four polarizing plates 21 to 24. Lasers are generally linearly polarized, but
With respect to this polarization direction, quarter-wave plates 9 and 13 are arranged in each optical path of the ±m-order diffracted light so that their axes are at 45 degrees and -45 degrees, as described above. Then,
The light beams transmitted through the quarter-wave plates 8 and 13 become circularly polarized light in opposite directions, and when they overlap again as ±m-order reflected and diffracted light at the irradiation position _M2 , they become linearly polarized light again, but the polarization The orientation changes as the radiation grating 14 rotates. This luminous flux is divided into four luminous fluxes by the beam splitter 17 as described above,
The light enters light receiving elements 25 to 28 via polarizing plates 21 to 24 whose polarization directions are shifted by 45 degrees. As the radiation grating 14 rotates, signals whose phases are shifted by 90 degrees are obtained from the light receiving elements 25 to 28. For example, if the phase of the output signal of the light receiving element 25 is 0°, the phases of the output signals of the light receiving elements 26, 27, and 28 are 90°, 180°, and 270°, respectively. Let these output signals be P ₀ , P ₉₀ , P ₁₈₀ , and P ₂₇₀ , respectively. As shown in Figure 1, the output signals P ₀ and P ₁₈₀ ,
When P ₉₀ and P ₂₇₀ are respectively input to the differential amplifiers 30 and 31, there is a difference between the output signals of the differential amplifiers 30 and 31.
There is a phase difference of 90 degrees, and each output signal has a constant center level with the DC component removed. Then, after waveform-shaping these signals and detecting the rotation direction, the rotation angle can be determined by inputting the signals into a counter and integrating them.

By the way, in the conventionally used index scale type photoelectric rotary encoder, the wave number n of the output signal from the light receiving element, the total number N of the main scale, and the rotation angle θ correspond to equation (4). The relationship is n=Nθ/2π...(5), so the rotation angle Δθ per wave number is Δθ=2π/N (radians)...(6). On the other hand, in the present invention, from equation (4), Δθ=π/2mN (radian) (7). Therefore, according to this embodiment, when a scale with the same number of divisions is used, rotation angle detection accuracy that is 4 m times higher than that of the conventional example can be obtained.

Further, in a conventional photoelectric rotary encoder, the distance between the light-transmitting part and the light-blocking part is limited to about 10 μm, considering the influence of light diffraction.

Now, in order to obtain a rotation angle detection accuracy of, for example, 30 seconds, in the conventional example, the number of divisions of the main scale is calculated from equation (6), N = 360 × 60 × 60 / 30 =
Only 43200 is needed. Therefore, if the interval between the light-transmitting part and the light-blocking part at the outermost circumference of the main scale is 10 μm, the diameter of the main scale is 0.01 mm x 43200/
π=137.5mm is required. However, according to this embodiment, in order to obtain the same rotational angle detection accuracy as in the conventional example, the number of divisions of the radiation grating may be 1/4 m. ±1
In the case of m=1 using the following diffracted light, the number of grating divisions of the radiation grating 14 to obtain rotation angle detection accuracy of 30 seconds may be 43200/4=10800. In this example, if laser diffraction light is used, the distance between the transparent part and the reflective part can be narrow, so for example, if this is 4 μm, the diameter of the radiation grating is 0.004 mm.
×10800/π=13.75mm is sufficient. In other words, according to this embodiment, the shape can be 1/10 or less in size to obtain rotational angle detection accuracy equivalent to that of a conventional index scale type photoelectric rotary encoder. Therefore, the load on the rotating object to be tested is much smaller than in the conventional example, and accurate measurements can be performed.

FIG. 2 is an explanatory diagram of the irradiation positions M ₁ and M ₂ of the light beam on the radiation grating 5 of a part of FIG. 1 and the eccentricity between the center of the radiation grating 5 and the rotation center of the rotating object to be detected.

In this embodiment, two points M ₁ and M ₂ on the radiation grating 14 that are symmetrical with respect to the rotation center are used as irradiation points, that is, measurement points, and the eccentricity between the center of the radiation grating 14 and the rotation center of the rotating body to be tested is This reduces the impact of That is, it is difficult to make the center of the radiation grating 14 and the center of rotation completely coincide with each other, and eccentricity between the two is unavoidable. For example, as shown in Figure 2,
Between the center O of the radiation grating 14 and the rotation center O',
When the amount of eccentricity is a, the distance r from the center of rotation
The Doppler frequency shift at the measurement point M ₁ located at the position is r/(r+
a) to r/(ra-a). on the other hand,
At this time, the frequency shift at position M ₁ and measurement point M ₂ , which is symmetrical with respect to the center of rotation, is
Contrary to the change in M ₁ , from r/(ra-a) to r/
Since it changes to (r+a), the positions M ₁ and M ₂ ,
By using two measurement points at the same time, the influence of eccentricity can be reduced.

Next, the incident point M ₁ in the embodiment shown in FIG.
The imaging relationship between _M2 is shown.

The Doppler shift of the luminous flux at the incident point _M1 is expressed by the second equation, where the wave number vector of the incident luminous flux is k _i , the wave number vector of the reflected diffracted light is k _s , and the velocity vector of the radiation grating is V, as shown in the following formula k _s , When k _i is expressed as a vector, Δ=1/2π(k _s −k _i )・V ・ becomes the inner product of vectors. Here, |k _s |=|k _i |=2π/λ.

Figure _3A shows the vector (k _s −
FIG. 3 is an explanatory diagram of the vector representation of the Doppler shift when k _i )·V>0.

Here, at the incident point M ₁ , the reflected diffracted light is at the incident point
In order to undergo Doppler shift again at M ₂ and the amount of Doppler shift becomes 2Δ, each wave number vector k _i ′ of the incident light flux and the reflected diffracted light at the incident point M ₂ is
Let k _s ′ be (k _s ′−k _i ′)·V>0. For this purpose, the vectors k _i ', k _s ', and V at the incident point M ₂ must be as shown in FIG. 3B. In other words, the light beam reflected and diffracted at the angle of reflection θ in the direction of rotation of the radiation grating at the incident point _M1 is
It must be incident at an angle of incidence θ from the direction of rotation at M ₂ .

On the other hand, in order to establish a conjugate relationship between the images between the incident points M ₁ and M ₂ , it is necessary to satisfy the so-called shear proof law.

A relay optical system that satisfies these conditions is, for example, as shown in Fig. 4, in which the band included in the plane formed by _the vector _ks and the vector V at the incident point M1 is twisted once and then enters the incident point _M2 . becomes.

In this embodiment, the relay optical system shown in FIG. 4 is achieved using at least three reflecting mirrors and at least one imaging optical system.

In each of the embodiments described above, two diffracted lights of order ±m were used, but two diffracted lights of different orders may be used instead of diffracted lights of order ±m. Alternatively, the lattice pattern on the radiation grating may be composed of only transmitting portions or only reflecting portions, and only transmitted diffracted light or reflected diffracted light may be used.

Further, in the present invention, only the rotation angle may be detected, in which case a quarter wavelength plate, a polarizing plate, and a beam splitter are not required, and only one light receiving element is required.

Further, the light source in the present invention is not limited to a laser, but any light source that emits a single wavelength can be used.

As described above, according to the present invention, it is possible to achieve a small, high-precision rotary encoder that has a small load on the rotating object to be tested and reduces the eccentricity error between the center of the radiation grating and the rotation center of the rotating object.

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram showing an embodiment of the present invention, Fig. 2 is an explanatory diagram showing the eccentricity between the center of the radiation grating and the center of rotation, and Fig. 3 A and B are vectors of the incident light flux and reflected diffracted light, respectively. An explanatory diagram of the display, Fig. 4 shows the incident point M ₁ ,
FIG. ₂ is an explanatory diagram showing the imaging relationship between M2. 1 in the diagram
is a light source, 2 is a collimator lens, 3, 16 is a cylindrical lens, 4, 5, 6, 9, 10, 1
1 is a reflecting mirror, 7 and 12 are imaging optical systems, 14 is a radiation grating, 15 is a rotating shaft, 17 is a beam splitter, 21 to 24 are polarizing plates, and 25 to 28 are light receiving elements.

Claims (1)

[Scope of Claims] 1. A radiation grating in which a plurality of grid patterns are arranged at equal angles on the periphery of a disk, a rotating object connected to the radiation grating, and a first illumination means for making a luminous flux incident on the radiation grating. and two diffracted lights of specific orders among the diffracted lights from the light flux that have entered the radiation grating, and the first illumination means generate a point approximately equal to the center of rotation of the rotating object with respect to the incident position of the light flux on the radiation grating. at least three for re-injecting each at a symmetrical position.
After superimposing the two diffracted lights of a specific order diffracted again by the second illumination means having two sets of relay optical systems each consisting of a pair of reflecting mirrors and at least one imaging optical system, and the radiation grating, A rotary encoder comprising a light receiving means for receiving the superimposed light beams, and determining a rotation angle of the rotating object using an output signal from the light receiving means. 2. The rotary encoder according to claim 1, wherein the first and second illumination means irradiate the luminous flux linearly in a direction perpendicular to the radiation direction of the radiation grating.