RU2419067C2 - Absolute angle transducer (versions) - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: absolute angle transducer, according to the first embodiment, has a control element mounted on the controlled object and a receiving unit. The control element has a light source and a signal mask with a transparent slit lying in the focal plane of the lens. The receiving unit has a lens in whose focal plane there is a receiving matrix. The output of the receiving matrix is connected to a computer which determines the turning angle of the control element from the spatial position of the image of luminous slit relative the receiving matrix. According to the second version, the control element, which is mounted on the controlled object, has a signal mask with a mirror strip. The receiving unit has an illuminator with a light source for illuminating the signal mask and a receiving lens which constructs the image of the mirror strip in the plane of the receiving matrix.

EFFECT: improved accuracy and operational characteristics and reduced size and weight of the transducer.

2 cl, 3 dwg

Description

The proposed device relates to the field of measurement technology and can be used in automation, machine tool, instrument making, robotics, tracking systems and other fields of technology.

Numerous precision angle converters are known, described, for example, in the book: Babushkin S.G. etc. "Optical-mechanical devices", 1963. The most advanced of them are electromechanical devices such as a rotating transformer and optical goniometers.

In optical goniometers, the collimation principle of measuring the angle of rotation of the shaft is used, on which multifaceted mirror prisms are rigidly fixed. Domestic goniometers such as GS-10, GS-5 and GS-2 have an accuracy of 10 ", 5" and 2 ", respectively, in an unlimited range of angles. However, the dimensions of these devices are very large (more than 0.5 × 0.5 × 0.5 m ³ ).

The main disadvantages of these devices, taken as analogues, are:

- large dimensions and mass of devices;

- limited measurement accuracy due to the influence of backlash, gaps, rotational axes and interface nodes with the axis of rotation of the object;

- the high cost of devices, which for modern converters of accuracy class 1 ÷ 3 arc seconds reaches $ 3000 and more.

The shaft-code photoelectric encoders are also known (see, for example, book: N. Melkin, “Photoelectric Converters of an Angular Value to a Digital Code” // Sudpromgiz, 1966) and the most precise of them are devices with interpolation in the latter category (Vulbet J. Sensors in digital systems // M .: Energoizdat, 1981, p. 201). Modern domestic devices manufactured by SKBIS (St. Petersburg) have a third class of measurement error of up to 3.5 angles. sec, for example, the LIR-DA 190 device (http://www.skbs.ru) with dimensions: diameter 90 mm × 60 mm and weight 0.7 kg.

Known devices according to US patent No. 5018853 (05/28/1991), ed. USSR No. 1805287 (03.30.1993). However, in these devices a CCD array is used, and not a two-dimensional CCD array, which is a fundamental difference between the claimed device. Indeed, the operation of a CCD line with one signal slit is fundamentally impossible for constructing a full-fledged angle converter with a measurement range of 0 ÷ 360 °. Therefore, the device according to US patent No. 5018853 is operable only in the range 0 ÷ 90 ° (description of US patent No. 5018853, Summary of the invention, column 2, line 20). To measure the angle in the required range of 0–360 °, it is necessary to have two mutually perpendicular slots with a complex system of fiber-optic communication between the object and the meter to switch the backlight of these slots (see, for example, device according to ed. St. USSR No. 1805287). In addition, due to the fact that the number of pixels in the CCD matrix is approximately 10 ³ times the number of pixels in the CCD array, it is possible to construct axisymmetric measuring circuits for measuring the angle, which leads to an increase in the measurement accuracy by a factor of 100 or more, compared with the schemes analogues. Thus, the devices described in these patents solve narrow problems with low accuracy.

Also known are the devices described in the applications DE 102007023993 (11.27.2008), JP 2069641 (08.03.1990). These devices, on the contrary, contain a receiving CCD matrix and do not contain the signal slit necessary for precision angle measurements. They make a rough measurement of the angle together with the measurement of other linear quantities (Δх, Δу, etc.). Therefore, the size of the images of the signal structures in the plane of the receiving CCD matrices is compelled to be much smaller than the sizes of the matrices themselves, which leads to a sharp decrease in the accuracy of determination of the measured parameters (including the angle of rotation of the object).

The closest analogue (prototype) selects the device described in the book: Vulbet J. Sensors in digital systems // Moscow, Energoizdat, 1981, p. 201.

The circuit of the analogue device contains a mechanical coupling of the controlled object with the axis of the rotor of the converter. A glass disk with circular frame paths (encoded in the Gray code) is rigidly fixed on the rotor axis, and the device stator contains a system of lighting diodes and photodetectors, the number of which is equal to the number of code tracks.

In the chain of receiving photodetectors two-level signals (0 and 1) of the binary Gray code are generated, which are converted by the receiving circuit of the device into a binary code of the measured angle of rotation of the rotor shaft.

A physical limitation on the accuracy of this converter is the linear size of the period of the least significant bit of the code, which cannot be less than the wavelength of light λ (≈0.5 μm) due to light diffraction.

From here follow the main disadvantages of the prototype:

- large dimensions and masses of the device;

- complex mechanical components that provide rotations of the code disk and its conjugation with the axis of rotation of the object;

- high cost of the device.

The invention is aimed at solving the following problems:

- improving the accuracy and performance;

- reducing the size and mass of the converters;

- reduction in the cost of devices.

The tasks are solved by the fact that, firstly, a new photodetector for this type of device is used in the converters - multi-element devices, CCD arrays, on which linear displacements of 0.01 μm and above can be measured, which, accordingly, leads to a decrease in dimensions 10 times compared with the prototype. In addition, the cost of serial samples of CCDs is $ 100 ÷ 250, which determines the low cost of the device itself.

Secondly, optical circuits (both in clearance and reflection) and control elements installed on the measured object are very simple and cheap, which also leads to a reduction in the cost of the converter.

Thirdly, the optical connection between the device and the measurement object (instead of the mechanical connection in the prototype device) leads to the fact that the design of the converter does not contain any moving elements at all, and the distance between the device and the object can take different values. This, in turn, leads to an increase in the real accuracy of measurements and the expansion of the functionality of the converter.

Fourthly, the controller (PC) is used in the instrument circuitry, which implements the converter operation software and stores the results of metrological calibration, which also leads to an increase in measurement accuracy.

The invention is illustrated in figures 1, 2, 3.

Figure 1 shows a schematic diagram of a Converter with an active energy control element.

Figure 2 shows the signal optical image and its electrical equivalent.

Figure 3 shows a diagram of a device with a passive energy control element.

Figure 1 shows the control element CE, rigidly mounted on a rotary shaft, and the measuring unit IB.

The control element of the FE contains an LED 1, a pinhole 2, a condenser 3, a signal mask 4 located in the focal plane of the lens 5. The receiving unit contains a lens 6 and a receiving CCD - (CMOS) matrix 7, the signal from which is received via the USB-2.0 port to a personal computer PC or a specialized controller.

The device operates as follows.

The illuminator of the control element (LED 1, pinhole 2 and condenser 3) generates a uniform parallel light stream that illuminates the signal mask 4. The signal mask is a transparent narrow slit on an opaque substrate (Fig.1B). The mask 4 can be performed on a glass disk part or in the form of metal knives. The lens 5, in the focal plane of which the signal mask 4 is located, forms a collimation image of the light gap of the signal mask 4, which is rotated by the measured angle φ relative to the measuring unit.

The lens 6 of the receiving node builds in its focal plane an image of the luminous mask 4 CE. This plane is aligned with the receiving plane of the receiving matrix 7, which detects this image of the luminous mask 4.

Figure 2 presents the image of the light stroke of the mask 4 in the plane of the receiving matrix 7, and figure 2b shows a graph of the electrical signal on a separate row of the matrix.

Using the receiving matrix 7, the optical image is converted into an electronic digital equivalent and is sent to a personal computer PC.

Using specialized information processing algorithms in the PC, each row j of the receiving matrix is calculated with the coordinates x _{oj of the} energy center of the image points of the signal stroke (Fig. 2a).

The angle φ _j is determined from the obvious relation

where Δy _p is the pixel size of the matrix 7 (Δy _p = 3 ÷ 5 μm),

j is the row number of the matrix 7.

For values of φ> 45 °, rows and columns are transposed with the replacement j → i, and calculations are performed for each column i of matrix 7.

Thus, the measured angle φ is determined in the range 0 ÷ 360 °.

The calculated algorithm of the angle φ is

where M is the number of rows (columns) of the receiving matrix (M = 10 ³ ).

Therefore, the random error δφ of measurements

The experimentally measured value of the coordinate x _{oj of the} energy center of the stroke image is 0.03 Δx _p (Δy _p ), where Δx _p (Δy _p ) - x and y are the pixel sizes of the receiving matrix 7.

Therefore, by formulas (2) and (3) we obtain

All systematic errors of the device can be detected and eliminated at the calibration stage (with the accuracy of the reference device used). Therefore, the actual measurement accuracy of the device is determined by the value of δφ.

At the moment, made and tested the layout of the device according to the scheme of figure 1. The layout used a CCD matrix with a resolution of 640 × 480 pixels. The random error on the layout was δφ = 0.3 ang. sec for all angles in the range of 360 °.

The light diameters of the device are determined by the size of the receiving matrix and are equal to 6 × 6 mm. However, the dimensions of the matrix strapping board are ≈30 × 30 mm ² , which determines the actual dimensions of the device, which are much smaller than the dimensions of the photoelectric code sensors of comparable measurement accuracy.

The cost of the device according to the scheme in figure 1 is determined mainly by the cost of the matrix 7 together with the lens 6, which is equal to 200 ÷ 250 $.

The total cost of the device does not exceed $ 400 ÷ 500, which is much less than the price of modern devices of accuracy class 0.2-0.3 arc seconds.

The scheme of figure 1 also has a very important advantage, which consists in the fact that the distance between the FE and the receiving module can be variable over a wide enough range. This is because the circuit of the device is collimating.

In addition, since there is optical communication between the FE and the receiving module, the device does not have moving parts and any transition elements (couplings) between the FE and the receiving unit.

Figure 3 shows the autoreflex modification of the converter circuit. The receiving unit of this circuit contains a lighting unit (LED 1, aperture 2, condenser 3, beam splitter plate 8) and a receiving unit (lens 6 and matrix 7). The control element is represented by a mask 4 with a narrow specular stroke instead of a transparent slit in the circuit in figure 1.

The image of this stroke, illuminated by the illuminator, is built by the lens 6 in the receiving plane of the matrix 7.

This image, its detection and information processing algorithms are similar to those that were given in the description of the operation of the circuit in figure 1.

The only significant difference between the circuit in Fig. 2 and the circuit in Fig. 1 is that the distance between the FE (item 4) and the receiving module is constant, because this circuit of the device is not (auto) collimating.

Claims (2)

1. An absolute angle transducer, characterized in that it contains a control element mounted on the controlled object and consisting of a light source, a signal mask with a transparent slit located in the focal plane of the lens, and a receiving unit consisting of a lens in the focal plane of which is installed receiving matrix (CCD or CMOS matrix), the output of which is connected to a personal computer (controller), which determines the angle of rotation of the control element by the spatial position of the image eniya luminous gap relative to the receiving array. 2. Absolute angle transducer, characterized in that it contains a control element mounted on the monitored object and consisting of a signal mask with a mirror stroke, and a receiving unit, including a illuminator with a light source to illuminate the signal mask, a receiving lens that builds an image of the mirror stroke in the plane of the CCD or CMOS matrix, the output of which is connected to a personal computer (controller), which determines the angle of rotation of the control element by the spatial position a mirror stroke relative to the receiving matrix.

Images