Photoelectric encoder principle structure diagram
Incremental photoelectric rotary encoder
An encoder is a device that converts a physical quantity into a digital format. In motion control systems, the encoder's function is to convert parameters such as position and angle into digital quantities. Various types of encoders can be formed using mechanisms such as electrical contact, magnetic effect, capacitance effect, and photoelectric conversion. The most common type of encoder in motion control systems is the photoelectric encoder.
Photoelectric encoders are classified into rotary photoelectric encoders and linear photoelectric encoders according to their applications, used to measure rotational angles and linear dimensions, respectively. The key component of a photoelectric encoder is the photoelectric encoding device, which is a circular code wheel (or code disk) in rotary photoelectric encoders, and a ruler-shaped code strip in linear photoelectric encoders. The code wheel and code strip can be made of materials such as metal, glass, and polymers, depending on the application and cost requirements. Their principle is the same: generating digitized optical signals representing the position of motion during operation.
Figure 12.1 illustrates the principle of a transmission-type rotary photoelectric encoder. A combination of light-blocking and light-transmitting parts, arranged according to a specific encoding rule, is etched onto a code disk concentric with the shaft being measured. On one side of the code ring is a light-emitting diode or incandescent lamp source, and on the other side is a photoelectric device that receives the light. As the code disk rotates with the shaft being measured, the light beam passing through it becomes intermittent. Through the reception of the photoelectric device and the processing of the electronic circuitry, a specific electrical signal is output. This signal is then digitally processed to calculate the position and speed information.
The above describes the principle of a transmission-type photoelectric encoder. Clearly, photoelectric encoders can also be made using the principle of light reflection.
The code disk of an incremental encoder is shown in Figure 12.2. On modern high-resolution code disks, both the light-transmitting and light-blocking portions are very fine slits and lines, hence the term circular grating. The angle between adjacent slits is called the grating pitch angle, with the light-transmitting slit and the light-blocking portion each occupying approximately half of the grating pitch angle. The resolution of the code disk is expressed in counts per revolution (CPR), which is the number of pulses generated in the photoelectric detection section for one revolution of the code disk. For example, a code disk with a CPR of 2048 can resolve an angle of 10,311.8". Often, a special slit (or a set of slits) is also arranged on the code disk to generate an index or zero signal. The measuring device or motion control system can use this signal to generate a homing or reset operation.
From a theoretical perspective, the electrical signal output by an optoelectronic device should be a triangular wave. However, due to light diffraction caused by the gap between the moving and stationary parts and the characteristics of the optoelectronic device, the resulting waveform approximates a sine wave, and its amplitude is independent of the resolution of the code disk.
The design in Figure 12.1 arranges six sets of such baffle and photoelectric device combinations, two of which are used to generate the index pulse signal I (Z in some literature). The other four sets, due to their positional arrangement, generate four quasi-sine wave signals with a phase difference of 90°, referred to as A, B, A', and B'. Sending A and A', with a phase difference of 180°, to the two inputs of a comparator yields a square wave signal A with a 50% duty cycle at the comparator's output. Similarly, a square wave signal B can be obtained from B and B'. Thus, through the special arrangement of the photoelectric detection devices, dual-channel photoelectric pulse output signals A and B are obtained (see Figure 12.3). These two signals have the following characteristics:
(1) The duty cycle of both is 0%; Figure 12.3 Formation of dual-channel signal
(2) If signal A leads signal B by 90° in phase when rotating in one direction, then when the rotation direction is reversed, signal B leads signal A by 90° in phase.
This dual-channel signal characteristic provides the conditions for improving measurement resolution and acquiring direction signals.
Square wave signals A and B with a duty cycle of 0% have four special moments, namely the leading edge and trailing edge of their waveforms.
The preceding and following signals are evenly distributed at 90° within one cycle of the waveform. By extracting and utilizing these preceding signals, a pulse signal with a frequency multiplied by 4 can be obtained, thus increasing the resolution of the photoelectric encoder by 4 times.
Figure 12.4 shows a processing circuit composed of digital circuits, which employs a Schmitt trigger inverter, XOR gate, OR gate, and D flip-flops. The waveforms at various points in the circuit are shown in the figure, separated by dashed lines to represent the waveforms in forward and reverse rotation. It can be seen that the circuit generates a 4x frequency counting signal and a direction signal. Using these signals, along with positioning pulses, the electronic circuitry can determine the position of the motion system by counting the pulses. A counter can be used to increment when the shaft rotates in one direction and decrement when it rotates in the opposite direction, thus maintaining the memory of the absolute position without power loss.
Telescope axial position indicator
Figure 3.17 shows the working principle diagram of (a) the code disk and (b) the encoder of an eight-bit encoder.
Modern industry has provided a range of axial position indication devices for telescope axial systems. These devices include photoelectric encoders, circular induction synchronizers, and grating scales.
(1) Photoelectric encoder
An optical encoder is a binary photoelectric position indicator. Its basic principle involves using alternating light and dark stripes of varying degrees of intensity to obtain a binary digital signal representing the angular position via a photoelectric element. This signal is then decoded to obtain the absolute or relative value of the angular position. The code pattern of an absolute encoder is always unique, indicating the length or angular position. An optical encoder consists of a light source, a code disk, and a photoelectric receiver. The code disk is the most important component of the encoder. Figure 3.17 shows the code disk of an eight-bit encoder and its working principle diagram. The code disk shown here is a natural code disk. Absolute encoders have various code patterns. One type of code disk, called a Grei code, is particularly suitable for optical encoders (see Figure 3.18(a)). This type of code disk changes only one digit per increment, making it less prone to errors.
Figure 3.19 shows the working principle of incremental encoder code disk pulse information subdivision, where z represents the zero position.
Another type of photoelectric encoder is the incremental encoder. The code disk of an incremental encoder is shown in Figure 3.18(b). Its code disk is composed of alternating bright and dark stripes. Generally speaking, incremental encoders with the same resolution are much cheaper than absolute encoders. There are also some methods to improve the resolution of incremental encoders. Typically, an incremental grating code disk has four tracks, two of which are alternating bright and dark stripe codes, and the other two are power brightness indicator codes. These two stripe codes are staggered, so this type of encoder can provide not only the angle and magnitude of the code disk's movement, but also the direction of the movement. Simultaneously, when the square wave pulse information of the grating code disk is input into clockwise and counterclockwise increment/decrement counters, the square wave information of these two stripe codes can be decomposed into one, two, or four times finer signals to improve the encoder's resolution. If the grating code disk is of good quality, this fine four-fold signal can be accurate to half the value of each signal pulse.
To achieve finer resolution, a method using a grating reader can be employed. (See Figure 3.20) In this case, a small sub-grating is added behind the rotating grating. When coherent light illuminates the grating disk, the light intensity on the sub-grating surface is (Leki, 1999):
Figure 3.20 Schematic diagram of the working principle of subdivision of the sub-grating code disk in an incremental encoder (Leki, 1999)
In the formula, t1 is the projective index of the grating. If the period of the first grating is p, the period of the second grating is also p. Using w as the spatial frequency on the focal plane, the light energy on the focal plane is:
Figure 3.21 shows the relationship between the light intensity signal and displacement of the sub-grating code disk subdivision in the incremental encoder, where AU represents arbitrary...
Unit(leki,1999)ReprintedwithpermissionfromTaylor&Francis,Inc.
When M=0, the light energy of this signal can be expressed as a series. If only the first two terms are taken, the light energy at the focal point is a cosine function. Thus, through electrical subdivision, we can obtain even finer resolution. In practical applications, four sets of sub-gratings can be used simultaneously on the upper and lower sets of fringes to improve the accuracy of electrical subdivision. However, as shown in Figure 3.21, the focal energy of a periodic grating is not a true cosine curve. Therefore, if a modulated sub-grating as shown in Figure 3.22 is used, its focal energy is the true cosine curve, and the resolution after subdivision will be more accurate. Alternatively, by using a modulated parallel light source, using two surface light sources of different areas, the focal energy can also become a correct cosine function. By applying combinations of incremental gratings with different resolutions, sine and cosine values of different frequencies can be obtained, thus enabling the creation of a very high-precision absolute encoder. Generally, such high-precision encoders have multiple code tracks, consisting of a DC reference code and three to fifteen bits of sine and cosine codes.
Figure 3.22 Specific dimensions of the two modulator sub-gratings in the incremental encoder (Leki, 1999)
The inherent precision of modern grating technology can also greatly improve the accuracy of photoelectric encoders. A 16-bit incremental encoder, by adding a 16-bit absolute code pattern to its code disk and imaging two adjacent stripes of the incremental code simultaneously, can provide the precise position of the code disk, thus achieving the accuracy of an absolute encoder of more than 24 bits. This is a very important technological advancement.
(2) Circular induction synchro
Another similar shaft angle encoding device is the circular inductive synchro. Unlike photoelectric encoders, the circular inductive synchro is an analog device. The changes in values are continuous, not abrupt. The basic principle of the circular inductive synchro is shown in Figure 3.23. It consists of a stator and a mover. Its mover has only one coil, while its stator has several coils forming poles. The angle between each coil is degrees. When an AC voltage is input to the mover, and the mover axis deviates from the zero point of the stator by a certain angle, different amounts of current will be generated in the coils on the stator. As shown in Figure 3.24, we have:
Figure 3.23 Basic principle of circular inductive synchro
Figure 3.24 Output voltages in each coil of the stator of the circular induction synchro
In the formula, is a proportionality constant. If the coils on the stator are connected together as shown in Figure 3.23, the following current will be generated on the stator:
By utilizing this characteristic of circular inductive synchronizers, it is possible to measure minute changes in angle. When using a circular inductive synchronizer, a coarse code disk is also required to determine the absolute position of the angle. Compared to photoelectric encoders, circular inductive synchronizers have the following advantages: (a) the coil moving and fixed disks are relatively inexpensive; (b) they have lower environmental requirements and can be used in environments with temperature variations and vibrations.
(3) Application of encoders and other angle measurement methods
Photoelectric encoders require analog-to-digital converters in control loops, while circular induction synchronizers can be directly used for synchronous drive control. Both can achieve absolute or incremental indication of axial position. They offer high positional accuracy and good error repeatability, but high-digit indicators are more expensive. The axial position indication method using a grating scale and moiré fringes is a recent development, particularly suitable for large-aperture telescopes. This grating scale has an accuracy of less than 1 micrometer and is typically uniformly attached to the edge of a large drive wheel, providing high resolution through moiré fringes. A drawback of the grating scale is that it cannot guarantee the uniformity of all fringes, requiring calibration using a tabular method in computer control. In telescopes, grating scales are commonly used for absolute position calibration.
Absolute positioning accuracy in telescopes is necessary for accurate star guidance and positioning, while incremental positioning is for precise star guidance. Therefore, incremental encoders require high resolution. Absolute encoders can be directly connected to the telescope's drive shaft, eliminating other error factors in position indication. However, sometimes, due to a low encoder bit depth or the need for light to pass through the telescope's drive shaft, the encoder can be mounted on the first-stage gear. This amplifies the encoder's resolution, but gear errors also affect the accuracy of the absolute angle display. This error significantly impacts absolute position calibration. However, in recent years, many telescopes have adopted high-resolution incremental amplification indicators, using other highly repeatable devices, such as high-sensitivity levels or special grating lines, to provide the absolute zero point of the axial angular position, thus eliminating the need for expensive absolute encoders. Some newer telescopes also utilize precision electromagnetic switches as encoders for the axial angular absolute position, with repeatability of approximately 1 micrometer. In this design, a precision electromagnetic switch is installed every 10 or 15 degrees. Between each precision electromagnetic switch, an incremental encoder is used; sometimes, a friction surface can even drive a lower-order incremental encoder. This design is more cost-effective compared to other designs. Proper installation is crucial for encoders to achieve their resolution accuracy. When connecting the encoder to the shaft, it is paramount to avoid applying any force or torque to the encoder shaft. Therefore, the encoder coupling should have relatively low strength in the axial and radial directions, but high strength in the circumferential direction.
Some new six-bar platform telescopes also incorporate an angle measuring device called a fiber optic resonant gyroscope. A fiber optic resonant gyroscope comprises three fiber optic loops. A beam of light is emitted from a narrow-bandwidth laser diode to one end of an optical fiber. Simultaneously, the end of this fiber loops back to the starting end and couples with the fiber at the starting end via an optical coupler, forming a loop through which light passes in both directions. In the middle of this loop, another optical coupler couples the first loop with a second fiber loop. Opposite to the second fiber loop, a third optical coupler couples the second fiber loop with the third fiber loop. The third fiber loop is an open-loop circuit, connected to detectors at both ends. In this system, if all loops and couplers are fixed, and the two couplers in the second fiber loop are located at the symmetrical point of the loop, it will resonate with light of a specific wavelength. However, when the second loop has a small angle relative to the first loop, the light path increases in one direction and decreases in the other, thus the new system resonates at two different frequencies. By comparing the original resonant frequencies, one frequency is higher and the other is lower. Measuring the change in resonant frequency allows us to understand the change in angle, thus achieving the purpose of angle measurement.
