How is the rotary encoder encoded

"Encoding" is a mathematical model, which means that the information is arranged and sent out (Encode) during information transmission, and the receiving end receives the encoding and decoding to obtain the information (Decode).

The mathematical modes of modern coding are 0 and 1 (with and without, on and off, long and short, high and low levels, etc.). For example: the semaphore and light language of the two ships in the voyage, the Morse code (long and short tone) of radio transmission; another example: the CD record and record player of the audio encoder, the DVD, Blu-ray and high-definition TV of the video encoder.. .and many more.

The rotary encoder is to "encode" the current angle information or the rotation angle change information.

Relative

Relative rotary "encoding" refers to the need to encode information and output more than twice, and obtain a relatively changed angle value by comparing the encoding information more than twice. It is also called an incremental encoder.

Among them, the photoelectric incremental encoder expresses the interval change between 1 and 0 through the passage and occlusion of the light of the code disc marking. Such an on-off interval change represents the amount of angle change, like a human pulse , Also called pulse output.

This kind of coded angle information must be rotated to have information change and accumulation, just like a CD record must be rotated to have sound. The current angle value can only be obtained by comparing, accumulating and memorizing the previous historical information.

Relative rotary encoder code disc

Absolute

Absolute rotary coding means that there is no need to read the change information more than twice, no need to compare and accumulate, the absolute value is independent of historical information, and the current angular position value can be read directly at one time.

The number of bits of a single-turn absolute value encoder is the number of bits to the Nth power of the encoding method of 0 and 1 to represent its resolution position code, 8 bits have 256 encoding positions, 10 bits are 1024, 12 bits are 4096, 13 Bit 8192, 14 bit 16384, 16 bit 65536... and so on, the highest number of digits in the current industrial market is the 25-bit single-turn absolute encoder from Heidenhain, Germany.

The relative code is "variable angle amount", including internal and external counter count accumulation and memory, and is related to historical count; absolute value code is "direct current angle value" and has nothing to do with historical count.

What is the encoding method of the relative incremental encoder?

The incremental encoder divides a 360° circle by outputting the number of pulses per revolution (PPR), which is also called resolution. The change of each pulse period represents a certain angle change. In this way, the method of pre-dividing the angle to each period and reading the count period change is "incremental encoding". The incremental encoder output is generally A, B, Z three-phase signals , The output of A and B are different by 1/4 cycle, forward rotation is 1/4 phase difference, and reverse rotation is 3/4 phase difference, so you can judge the positive and negative rotation of the encoder; Z phase only provides one position per revolution Pulse to set the zero value.

What is the positive and negative logic of incremental coding?

The square wave of the incremental pulse signal has a logic 1 at high level and a logic 0 at low level. This is called positive logic of encoding, and vice versa.

Generally, the code of NPN open collector output form is negative logic.

What is CW encoding or CCW encoding?

Looking at the shaft end of the encoder, the encoder shaft rotates "clockwise", and the encoded data increases, which is called CW. Facing the shaft end of the encoder, the encoder shaft rotates "counterclockwise", and the increase in data is called CCW.

What are the quadrature A and B phases of incremental signals?

The incremental encoder outputs phase A pulses and phase B pulses. The phase difference between phase A and phase B pulses is 90° (1/4T of a cycle), which is called "quadrature phase A and B". The quadrature phase difference can provide the receiving device to determine the rotation direction of the encoder shaft. The rotation in the positive direction is 1/4T, and the rotation in the reverse direction is 3/4T, so that the subsequent receiving device can easily determine whether the incremental pulse signal is "Increase" or "decrease" the cumulative pulse number.

What is quadruple frequency of incremental quadrature A and B signals?

The quadrature A and B signals have a phase difference of 90°. Within a pulse period, 4 changes can be judged by the changes of the rising and falling edges of the A and B pulse signals, or by the A signal encoding logic 1 and 0 And B signal 1 and 0 combined decoding, A and B form a 2-bit code. There can be four divided intervals in one pulse period. These 4 changes or 4 intervals are interpreted, which is called the incremental quadrature AB signal Quadruple frequency.

Why is the phase difference between A and B of the incremental encoder 90°?

In this way, the forward rotation (1/4 difference) or reverse (3/4 difference) of the rotation direction can be judged more clearly, and there is a good tolerance for errors.

What is an absolute value digital code?

The presence and absence obtained by the sensor's internal signal, and the electrical high and low levels are used as the codes of 1 and 0, and the absolute value code composed of N-bit 0 and 1 codes (also called binary codes). The presence and absence of digital coding is not sensitive to the size changes caused by the interference of various micro-motion factors, so the anti-interference is strong. More and more applications currently use absolute encoders.

What are the ways to encode absolute values?

Decimal code

In our daily life, our numbers are 0-9, which are incremented every decimal, which is called "decimal" digital coding;

Because decimal encoding requires ten steps of distributed physical quantities, physical realization is relatively inconvenient, and this encoding is rarely used in encoders.

Pure binary code

Modern electronic digital codes generally represent 0 and 1 with on-off (light and dark). There are multiple digits to the Nth power of 2 to form the code. The simplest is pure binary code BIN. The changing numbers are 0 and 1, which are entered every two. 1. This is called "pure binary" digital coding, which is often used as a code in computer electronic applications;

Hexadecimal code

Decode 0 to 15 in a set of 4 bits in binary to decimal, but above 10 is not expressed in two digits, but expressed in A to F. For example, 10 is A and 15 is F. This is a hexadecimal code, which is convenient for data Deposit and identification;

BCD code

Each decimal digit is decomposed and decoded into a 4-digit set of ten numbers in binary code. The BCD code is only 0-9 after decoding to decimal, and there is no need to consider the carry problem of the decimal system. It is generally used for 7-segment digital display that the human eye is used to. tube;

Gray code

Gray code is a kind of binary code. Its characteristic is that every time a word changes in increment (or decrement), only one of the digits changes (changes between 0 and 1), and the end of the data cycle to the starting point also follows This principle, so that in the process of changing a word, only one digit jumps, the number "can" changes the least, and the error probability is the lowest. Particularly suitable for multi-code synchronous reading of absolute encoder or multi-digit synchronous output (parallel output), and also suitable for communication transmission;

Gray's remaining code:

Most of the binary codes are power-of-two codes, and the total number of codes is also a power-of-two, such as 8-bit 256, 10-bit 1024, etc. In the angle calculation of the encoder, the angle calculation cannot be divided and rounded. , Sometimes the integer of the angle is used. For this purpose, there is a Gray residual code. A part of the gray code is removed, and the remaining part is used for encoding. Therefore, it is also called a "red code", such as 180, 360, 720 codes. Yu Gray code". The decimal number decoded from Gray's residual code to decimal does not start from 0 and needs to be "complemented".

Other codes

In communication transmission, there are some uncommon codes, which are variations or combinations of the above coding forms, such as Manchester codes, etc., which are not described here.

How to convert Gray code to pure binary code (BIN)?

Decoding process: compare "0" with the highest bit of the Gray code with an "exclusive OR", the result is kept at the highest bit (pure binary code), and then the result is compared with the second highest bit of the Gray code by "exclusive OR", and the result is retained In the (pure binary code) second highest bit; and so on, until the lowest bit.

The decoding from Gray code to pure binary code must be full code decoding, from the highest bit to the lowest bit.

How to convert Gray code to pure binary code?

The decoding method from Gray residual code to pure binary code is exactly the same as the decoding method of Gray code, but after decoding to BIN code, it will be found that the encoding does not start from 0. For example, 180 Gray residual code is taken from 256 Gray code and decoded After that, a constant 38 needs to be subtracted.

When installing an absolute encoder, do I need to start from the coding zero point?

No need. Each encoding position of the absolute encoder is unique, any position can be used as the "zero point", other positions confirm the positional relationship with this "zero point" during installation and initialization and store them permanently in the receiving device. Read during use Take each unique code and decode this relationship.

Why do many absolute encoders use Gray code?

In the coding of Gray code or Gray residual code, in each change of a word's up and down sequence, only the 0, 1 change of one bit is changed, the physical energy change is the smallest, and there is no problem of synchronization reading on each digit. Therefore, it is insensitive to interfering factors and has the smallest error probability. There will be no error problem in the synchronous reading of the internal code of the absolute encoder or the external synchronous output. The absolute encoder uses this code more often.

Compared with other codes, such as pure binary code or BCD code, when the order of a word changes, there may be multiple 0, 1 changes, and the physical performance may change suddenly. If you need to read or output the digits synchronously (For example, parallel output), it is difficult to ensure synchronization consistency in response, and some changes in the reading will be read first, and some changes will be read later, and there will be short-term error code jumps, and the read or output needs to be locked. , But the response speed is greatly reduced in this way. Therefore, codes other than Gray code cannot be used in internal synchronous reading or external synchronous output (such as parallel signal output).

Encoding of multi-turn absolute encoder

Absolute multi-turn encoder means that within its measurement range, not only does it have an "absolute value code" within 360° of a single turn, but also has a unique absolute code that does not rely on counting multi-turn values after 360°. It is generally similar to the clock-style minute and hour hands.

The value reading of the absolute multi-turn encoder must have nothing to do with history, and there is no question of whether there is memory or whether there is a battery. Power-on reading or not reading, power-off memory or not, battery with or without, the absolute value coded data are there. If the encoder is accumulatively counted by an internal or external counter, the output code value is related to the historical count. Once the historical relationship is lost or wrong, the information will also be wrong, and it cannot be called an "absolute value" multi-turn encoder. Does it have batteries?

The multi-turn absolute encoder expresses its resolution and measurement circle stroke in the way of "number of single-turn x number of multi-turns (or number of digits)". Multi-turn is commonly used for 4096 turns, and a small amount can reach 16,384 turns, 65536 Circle (16 bits).

GMX425 series absolute multi-turn encoders can reach 12-bit 4096 revolutions conventionally, and can reach 16384 revolutions and 65536 revolutions for special needs.