Rotary Encoders<\/h2>\n\n\n\n
A Rotary Encoder is a device that looks (physically) like a potentiometer but it does not vary the resistance between two\/three connectors. Instead it gives pulses when the axle is rotated. Another difference with the potentiometer is that you can rotate the axle over 360\u00b0 (in fact, there is no limit in how many degrees you can rotate the axle).<\/p>\n\n\n\n
![\"\"](\"https:\/\/willem.aandewiel.nl\/wp-content\/uploads\/2019\/10\/RGB_Rotary_Encoder.png\")
<\/figure><\/div>\n\n\n\nMost Rotary Encoders have a switch that is activated when you press the axle. And then there are Rotary Encoders that also have a three colour LED inside (RGB Rotary Encoder<\/em>) which give you the option to feedback to the user a pallet of colours. <\/p>\n\n\n\n The Rotary Encoder generate pulses when the axle is rotated. Most Rotary Encoders generate 24 pulses at every 360\u00b0 rotation. In your program you can count these pulses and let the program act on it (for instance: de- or in-crease the voltage of a bench power supply or step through a menu-list).<\/p>\n\n\n\n You may ask: \u201cwhy use a Rotary Encoder and not just use a potentiometer?<\/em>\u201d. Well of course there is no \u201cthis is better than that<\/em>\u201d answer. It depends on the project at hand. If you build an analogue HiFi amplifier the potentiometer is probably the better choice. If you have a project with a microprocessor the (digital) Rotary Encoder gives you a lot of options to create a nice user interface (with maybe a fine\/coarse change depending on the axle pushed short or long and a coloured feedback to show which mode is active).<\/p>\n\n\n\n This is an example Sketch to \u201cread\u201d a Rotary Encoder. You need hardware bounce control circuitry on pin A and B of the rotary Encoder to make it work reliably:<\/p>\n\n\n\n There is a drawback on using a Rotary Encoder in your project: You need a lot of pins<\/em>! This is called a (data) \u201cbus<\/em>\u201d because you can connect a lot of devices at it. The \u201cbus\u201d exists of two lines. A clock line (SCL<\/em>) and a data line (SDA<\/em>). There is always (at least) one \u201cmaster<\/em>\u201d device and all the other devices are \u201cslave<\/em>\u201d devices. Every device has a unique address to differentiate between each other. I will not go into the depths of the I\u00b2C protocol, but normally the master will claim control over the bus and sends a request to a slave with a specific address. The slave in turn will take action on the request either by performing a specific action in the slave itself, by sending data back to the master or simply by sending an acknowledge back to the master to let him know he received the request. Here<\/a> you can read more about the I\u00b2C protocol.<\/p>\n\n\n\n Wouldn\u2019t it be nice to connect a Rotary Encoder using only the two wires of the I\u00b2C bus!?<\/p>\n\n\n\n And that is what this post really is about: A Rotary Encoder that you connect via the I\u00b2C bus and control with the I\u00b2C protocol.<\/p>\n\n\n\n I developed the firmware and a small PCB with an ATtiny841 micro processor. I chose the ATtiny841 because it has Wire<\/em> hardware (the layer underneath I\u00b2C) for a Slave builtin. <\/p>\n\n\n\n The firmware makes the ATtiny841 act like a \u2018normal<\/em>\u2019 I\u00b2C Slave at one end and interface with a RGB Rotary Encoder<\/em> at the other end. For ease of use I wrote an Arduino\/ESP8266 Library<\/a> to interface with the I\u00b2C RotaryEncoder.<\/p>\n\n\n\n With this setup it is possible to connect as many (literally!<\/em>) I\u00b2C RotaryEncoders to your microprocessor as you like and still only use two GPIO pins. You can even connect other I\u00b2C devices (display\u2019s, sensors etc.) to the same two GPIO pins leaving a lot of GPIO pins free for other purposes. <\/p>\n\n\n\n If you want, you can use the Interrupt pin of the I\u00b2C RotaryEncoder. <\/p>\n\n\n\n This pin generates an interrupt for every change in the position of the rotary axle or push button. All I\u00b2C RotaryEncoders share the same Interrupt line.<\/p>\n\n\n\n#define outputA 6\n#define outputB 7\n\nint rotVal = 0; \nint rotState;\nint rotLastState; \n\nvoid setup() { \n pinMode (outputA,INPUT);\n pinMode (outputB,INPUT);\n Serial.begin (115200);\n \/\/ Reads the initial state of the outputA\n rotLastState = digitalRead(outputA); \n} \/\/ setup()\n\nvoid loop() \n{ \n rotState = digitalRead(outputA);\n \/\/ If the previous and the current state of \n \/\/ the outputA are different, that means a\n \/\/ Pulse has occured\n if (rotState != rotLastState){ \n \/\/ If the outputB state is different to the outputA state, \n \/\/ the encoder is rotating clockwise\n if (digitalRead(outputB) != rotState) { \n rotVal ++;\n } else {\n rotVal --;\n }\n Serial.print(F(\"Position: \"));\n Serial.println(rotVal);\n } \n rotLastState = rotState; \/\/ Update the previous state to \n \/\/ the current state\n} \/\/ loop()<\/pre>\n\n\n\n
One RGB Rotary Encoder needs two GPIO pins for the Encoder itself (plus GND), then you need one GPIO pin for the push switch and three GPIO pins for the RGB LED. Thats a total of six GPIO pins! On, for instance an ESP8266 that leaves only three free GPIO pins to control the rest of your project! On an Arduino UNO you have more free GPIO pins so one Rotary Encoder will not be a problem. But what if you want to connect two or even three Rotary Encoders. On an ESP8266 that\u2019s impossible, but on an Arduino UNO you will also run out of free GPIO pins.<\/p>\n\n\n\nLuckily, there is a solution!<\/h3>\n\n\n\n
The Inter-Integrated Circuit bus (I\u00b2C)<\/h2>\n\n\n\n
The I\u00b2C Rotary Encoder<\/h2>\n\n\n\n
![\"\"](\"https:\/\/willem.aandewiel.nl\/wp-content\/uploads\/2019\/10\/I2C_RotaryEncoder_v22-PCB_top-1024x848.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/willem.aandewiel.nl\/wp-content\/uploads\/2019\/10\/I2C.svg_-1024x362.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/willem.aandewiel.nl\/wp-content\/uploads\/2019\/10\/I2C_RE_bus.png\")
<\/figure>\n\n\n\n