DE19518714A1 - Optical sensor with measuring body - Google Patents

Abstract

The measuring system (M) has a micro (S,facet), which by means of a diffractively working optical units (DOE), is imaged on a light sensitive screen as a pattern in the receiver (R). The diffractive optical system (DOE) is integrated in the measuring system (M). The light source is a laser. The diffractive optical system (DOE) is arranged between the measuring system (M) and the light sensitive screen. The light source is a LED. The transmitter is designed as a rotary transmitter, and the measuring system (M) is a pulse disc with a microstructure (S), produced according to the method for compact discs.

Description

The invention is in the field of optical encoders, especially rotary encoder, and relates to an encoder according to the preamble of claim 1.

The basic principle used so far in optical sensors is that on a measuring body, with encoders, for example, a round disc, a high precise line grid is applied. This grid can be both strictly peri be coded odically as well as aperiodically. To the position and speed This grid is recognized by this grid lights up and the light fluctuations that are characterized by different codie tion or, in the case of periodic grids, caused by the movement, with Measured using photosensitive elements. This principle has changed very proven, since digitization is already realized on the measuring body becomes. However, this also results in serious restrictions. Be due to the unavoidable divergence, the grid cannot be used big be made small / tight. This is a major limitation in the realization of very high resolutions or in the production of filming encoders with very small disc diameters. So far, people have tried this Border to move by using the appropriate read heads  positioned very close to the grid or by adding more lenses and lenses which brought into the beam path.

The more cost-effective solution, namely reducing the distance of the reading head, is particularly a hindrance when the donors do not have their own own storage. To use this, in itself, inexpensive construction to be able to measure the desired and also possible accuracy to maintain the bearings of the machine (length encoder) or the Motors (encoders) are carried out much more precisely and therefore more expensive, than the function of the machine or motor would require.

It is now possible to place a shadow image on the light instead sensitive elements to create a real image of the grid to project the photodiodes. This must now, not as expected would be to build an optic from individual lens and aperture elements be, but with the invention presented below are opti functions as fine, diffractive grating structures in the coding grating and corresponding diode patterns on the photosensitive elements housed. According to the invention, this is done with diffractive optical elements so-called DOEs.

Diffractive measures are the essential elements of the invention and the matching read head with the light-sensitive To name an array on a chip (preferably an opto-ASIC). The diffract ve material measure includes, besides the coding grid, for example the periodi Incremental grating, the radial or tangential absolute grating and similar grating types with a resolution of 10 µm-20 000 µm, in addition another diffractive grating with a grating constant in the submicron range,  for example 0.6 µm. The diffractive grating has optical functions such as bundling the light and directing the light in certain directions conditions or orders of diffraction. The reading part consists of a (functio nellen) hybrid part, which is the light source, the light-sensitive receiving unit (e.g. a chip with a diode array) and a beam splitter with integrated diffractive optical elements (DOE).

The advantages of the solution according to the invention are as follows:

• - The distance of the reading head is 2 mm instead of according to the be known technology required 60 µm; the light beam from the transmitter must do not have a particularly narrow divergence, special precautions for precise storage is no longer necessary;
• - The diffractive grating allows optical functions that reflect reflex stung make particularly cheap, which simplifies assembly of the Reading part is possible and the height of the sensor practically halved becomes;
• - the diffractive grating can be placed on quartz, glass, silicon, polycarbonate, Nickel, steel, aluminum or other suitable materials in etching or replication technology and is therefore particularly expensive inexpensive for the production of large quantities, in particular This is the case if, for example, manufacturing methods such as for compact discs can be used;
• - An encoder or length encoder consists essentially of only three Mounting parts: the measuring standard, the light source / reading part and the diffractive intermediate optics. If laser light is used, one can using the fan-out effect, do without an intermediate optic  ten only when using an inexpensive light source, such as e.g. LEDs, is necessary;
• - The compactness of the read head in connection with the diffractive Measuring standards allow a mixed coding to be used, näm Lich an incremental and an absolute grid, whereby the inherent System security is very increased;
• - Due to the large distance between the reading heads, that in the motor or the dimension existing storage may be "used";
• - Even with very high resolution, very small panes are now through knives for encoders possible;
• - The assembly on a miniature motor is now possible small dimensions of the (three-part, compact) coding unit much easier.

A donor with these advantages can on the basis of the in the patent claims Chen specified invention can be realized. With the help of subsequently Figures listed the invention will now be discussed in detail.

Fig. 1 shows a first basic concept of the arrangement of a light source, a pulse disc and a light-sensitive receiver for oblique illumination of the pulse disc and the inclined direct reflection to the receiver.

Fig. 2 shows a second basic concept of the arrangement of a light source, a pulse disk and a light-sensitive receiver for the vertical illumination of the pulse disk with the aid of a beam splitter and perpendicular direct reflection on the receiver.

Fig. 3 shows a third basic concept of the arrangement of a light source, a pulse disc and a light-sensitive receiver for obliquely illuminating the pulse disc and perpendicular direct reflection on the receiver.

Fig. 4 shows various orders of diffraction grating in (four) selected spectral bandwidth of a light source.

FIGS. 5a and b show variants of diffractive multifocal elements, and

Fig. 6a and b show how the "array-compatible" retroreflection from front, moving facets with diffractive multifocus elements on a material measure, in which point-shaped light spots are processed on the light-sensitive element.

Fig. 7 shows a basic structure of a rotary encoder with a general imaging optical system between the pulse disc and the detector, schematically showing an array of light sensitive diodes is shown on the detector, and

Fig. 8 shows a preferred embodiment of an encoder according to the invention with a 3-fold diffractive optics between the pulse disc and the detector.

Fig. 9 shows in connection with the coding, for example, a structure on a material measure, pulse disk or Län genmeßkörper.

Fig. 10 a and b show the transfer functions for an encoder with the (higher) resolution corresponding to a base number, and for an encoder with the (lower) resolution corresponding to multiple basic figures.

The light modulated by the material measure is detected in the inventive encoder not via transmission, as in other known ones Systems, but via reflection. The radiation source and the detector is So arranged on the same side of the material measure, which in principle the overall length of the encoder is reduced. The pulse generating information of the Measuring standards, for example the light / dark distribution, are applied to the detec Transfer gate segments (pixels), for example, an opto-ASIC. This happens either the one with an imaging optics between the measuring standard and the detector or by a microstructure on the material measure, which the ge desired brightness distribution in the detector level, i.e. on the array, generated.

Figs. 1 to 3 show different types of, for example rotary encoders with their specific characteristics.:

• 1. an oblique lighting with the light source L on the pulse disc M and evaluation of the direct reflection on the detector R, has the Disadvantage that a shift of the disc due to a possible pitch Exercise of the figure arises, which is an error with a pixel array generated. To eliminate this disadvantage,  
• 2. vertical lighting of the pulse disc M and evaluation direct reflection can be selected using a beam splitter B, what this additional component B needs. If you have the beam splitter can handle, can
• 3. an oblique illumination and the vertical detection of scattered instead of directly reflected light are selected, which eliminates both disadvantages of the types according to FIGS. 1 and 2. The deflection of light that is radiated obliquely onto the pulse disk M in the normal direction is caused by an indicatrix I generated by the microstructure on the pulse disk, with such microstructures being able to deflect the incident light preferably in the direction of detection wherever one is want to place the detector. The choice of diffractive order also ensures the sufficient irradiance (at least 1 mW / cm²). The indicatrix of, for example, a determined microstructure is shown in the beam path and designated I.

Fig. 4 shows the relative size of grating diffraction orders -6. . . 0. . . 6 on the abscissa with four selected spectral bandwidths 0 to +/- 100 nm of a light source. In principle, it is advantageous that the optical function for generating the light distribution required in the detector plane is integrated directly into the structure of the material measure. The attractive possibility of using a diffractive microstructure to split a laser beam into a large number of sub-beams with different directions of propagation (fan-out elements) is known from the literature. In the simplest case, the diffraction orders of a grating can be used, but the intensity decreases with increasing diffraction order. In the case of a light source with a certain spectral bandwidth, the diffraction orders smear and can also overlap, which is also shown schematically in FIG. 4 just discussed.

With an LED whose spectral bandwidth is less than plus / minus 50 nm, the chromatic error could be killed, provided that no more than 3-4 diffractions regulations are used. With diffractive fan-out elements you can meet these conditions because they are a combination of different Lattice structures exist, of which only the respective first diffraction order comes into play. By a suitable combination of these basic grids these elements can be designed so that in all partial beams the same intensity is present. Such elements of the so-called "Dam man grids ", which only split into partial beams, require one more additional optics for focusing.

In principle, it is possible to integrate the focusing function into the diffractive structure S, so that a point grid in the detector plane D would be generated directly. There are two basic types of elements, as shown in FIGS. 5a / b. In the type of FIG. 5a, the element aperture comprises a plurality of partial beam bundles, whereas in the type in FIG. 5b each partial beam bundle is assigned a partial aperture. The numerical aperture of a partial beam is significantly larger in the type of FIG. 5a, as a result of which the light spot diameter in the focal plane or detector plane D can be smaller than in the type shown in FIG. 5b. The beam splitting in accordance with FIG. 5a is feasible only with diffractive structures, while a division into sub-apertures, as shown in Fig. 5b, with conventional optical elements, such. B. a grid of small concave mirrors is possible. Elements of the type according to FIG. 5a are less suitable for use on the material measure than that according to FIG. 5b, which would vary the illuminance of the spots in the detector plane D very strongly with the disk rotation or the scale shift.

Figs. 6a and 6b show in the running direction (the code associated) Facet th diffractive multifocal elements, the structure S, körperung on the Maßver as they pass through the light beam from the light source here, and the light spot. Fig. 6a shows the light spot as it extends over the expansion of a single facet and Fig. 6b the same light spot as it extends in half over two adjacent facets. The track on the measuring scale is a sequence of facets, each of which is a diffractive multifocus element. The material measure is moved into a position in which the facet N is fully illuminated, with several equally bright spots being generated in the detector plane. After a further movement of the measuring standardization by half the length of the facet N, this and the neighboring facet N + 1 are each half illuminated. The light spots generated by each facet on the detector level D are still equally bright with one another, but the irradiance per spot is only half compared to the position in which an entire facet was fully illuminated. When moving on, the brightness of the spots of facet N continues to decrease and that of the spots of facet N + 1 continues until the full illumination of facet N + 1 is reached and all spots (always equally bright) shine again as strongly as that was also the case with full illumination of facet N. Such a variation in the irradiance of the detector should, of course, be avoided, for example by assigning an individual focusing facet to each spot, which is designed as a diffractive element or as a small concave mirror, or by modulating the measuring standard with the help of an optical element Detector level is mapped.

It turns out, however, that an optical function for generating the in the Detector level D requires light spot distribution only with difficulty in the dimension ver body can be integrated if an LED instead of a laser Light source is used. If no laser is available for cost reasons  the fan-out effect but an LED should be used, which is most of the applications would apply, you would have to do the following Embodiment described with the optical imaging in the detector prefer level.

FIG. 7 schematically shows a system with a general imaging optics O between the measuring standard M and the detector R. The measuring standard M is illuminated obliquely by the light source L, as has already been shown schematically in FIG. 3. The imaging optics O and their arrangement is subject to the dimensional constraint or, more precisely, the areal extension of the detector R. When scanning three tracks on the material measure, the detector, or its light-sensitive surface, can have a shape as indicated (in the form of a T) is shown in the drawing. All calculations show that the full width of 3.5 mm of all traces of the material measure cannot be imaged with a single cell lens in the detection plane. Instead, separate lenses must be assigned to several subfields, which is shown in FIG. 8 as a preferred embodiment.

This preferred embodiment uses a lenticular screen DOE as a diffractive optical element for imaging the microstructure of the individual tracks on the light-sensitive field of the Opto-ASIC R. In order to prevent crosstalk between the optical images of adjacent lenses, corresponding, for example honeycomb-like, must be used Channel separator C can be seen before. This measure requires an additional component, which initially appears to be a disadvantage, but this component can at the same time be designed as a holder for the optical element DOE used between the measuring standard M and the detector R. The light spot on the material measure is also generated here according to FIG. 3 by obliquely incident light from the light source L.

Fig. 9 now shows an example for coding microstructures as a measuring standard. The disk shown here, for example (it may just as well be a scale), contains 7 tracks: 2 tracks for the increment signals A, B with 1024 positions (resolution); 2 tracks for the N signal and 3 tracks for the ABS absolute code with 256 positions. The principle of absolute coding is based on serial coding. The first two tracks contain the code, the third track is used to select the more contrasting zone. The scanning is differential, which is why all tracks are double. The corresponding receiving zone of an opto-ASIC contains: 2 scanning diodes for each A, B track, ie one for signal A and one for signal B; 2 scanning diodes for the N path, also one for each track; 34 scanning diodes for scanning the ABS line, namely for tracks 1 and 2 each 8 pairs of diodes (one pair per bit) and those for track 3 two diodes per bit. Doubling the number of diodes serves to avoid positioning problems, as have already been mentioned above. The maximum area of the diode pairs for scanning the A, B and ABS tracks must not exceed the area of a code bit. Such an ASIC would then have 40 scanning diodes. Furthermore, the approximate dimensions of the tracks are given in FIG. 9, which is a few hundred micrometers.

All raster numbers that add up to a basic number, e.g. 2 ^x = 1024, 512, 256, 128. . . , can be processed purely electrically with the same material measure (which contains the maximum number of positions). The simplest way to use this solution is to use a converter from solutcode-in-exit code, e.g. B. a ROM, for each resolution (Absolutcodie tion) and by a frequency divider (increment coding). In the case of multiple basic numbers, the problem of multiple resolution appears especially when more than 2 diodes scan the same slope in series, as is the case here with absolute coding, for example. A less complex solution to this problem is that an absolute code-in-output code converter is additionally provided for each basic number in addition to a given diode network in the ASIC. Example: the resolution is 720 positions or positions, but the measuring standard has 1024 positions or positions; the over-flowing bits, around 300, are individually connected to the others. In this way, approx. 300 bit pairs are created, whereby the accuracy is no longer continuous 1/2 LSB but between half and a whole LSB, which with the transfer functions, abscissa = position and ordinate = output code, in FIGS. 10a and 10b is shown.

The limiting cost factor still lies in the sensitive area of the Opto-ASIC. The larger the scanning field, the more expensive the ASIC. The fewer tracks you scan, the smaller the ASIC can be provided. An incremental encoder without absolute track scanning is therefore cheaper Absolute encoder that shows additional traces. But you can see the "full" in Provide formation (increment and absolute) on the measuring standard and scan and evaluate only a part of it with a "smaller ASIC". The financial effort is then not the material measure but for the module with the larger light-sensitive scanning field.

Claims (12)

1. Optical encoder with material measure, light transmitter and light sensitive receiver, characterized in that light transmitter (L) and receiver (R) are arranged on the same side of the material measure (M) (reflex method), that the receiver (R) is a light-sensitive screened Receiving field (D, pixel array) for the detection of light points (spots) and that the measuring scale (M) has a microstructure (S, facet), which uses diffractive optical means (DOE) on the light-sensitive grid in the receiver (R) are shown. 2. Encoder according to claim 1, characterized in that the diffractive optical means (DOE) is integrated into the material measure (M) (Light source is a laser). 3. Encoder according to claim 1, characterized in that the diffractive optical means (DOE) between the material measure (M) and the light-sensitive grid is arranged (light source is an LED). 4. Encoder according to claim 1, characterized in that it as a rotary Is formed and the measuring standard (M) is a pulse disk with a microstructure (S, facet), produced by the process for compact discs.   5. Encoder according to claim 1, characterized in that it as lengths is designed and the measuring standard (M) with a Microstructure (S, facet) is provided, which after the embossing process ren as for compact disks. 6. Encoder according to one of claims 1 to 5, characterized in that the rasterized receiving field (pixel array) in the detector (R), what light from the structured material measure (M) emp catches is a CCD array. 7. Encoder according to one of claims 1 to 5, characterized in that that the rasterized receiving field (pixel array) in the detector (R), what light from the structured material measure (M) emp catches is an opto-ASIC. 8. Encoder according to claim 6 or 7, characterized in that for different code tracks on the material measure (M) assigned Pixel fields are provided in the rasterized reception field. 9. Encoder according to claim 8, characterized in that for the assignment of code tracks and pixel fields so many diffractive optical Means or lenses (DOE) are provided, such as reali assignments should be settled and the beam paths formed by corre sponding accordingly arranged channel separator (C) prevented from crosstalk become.   10. Encoder according to one of claims 1 to 10, characterized in that the light source (L) and the light-sensitive receiver (R) are arranged the same substrate and form a structural unit. 11. Encoder according to claim 10, characterized in that the common same substrate for the light source (L) and for the light sensitive Receiver (R) designed as a hybrid circuit in semiconductor technology is. 12. Encoder according to one of claims 9 to 11, characterized in that that the duct separator (C) with the between measuring standard (M) and light-sensitive receiver (R) arranged diffractive opti means (DOE) is connected to the common substrate and in this way represents a transmitter / receiver assembly which assigned to the encoder of the material measure (M) becomes.