Using Rotary Encoders are a Cheap, Versatile and Robust Way for Users to Interact with Your Devices
Rotary Encoders are input devices. While they look similar to potentiometers, they are completely different: a Rotary Encoder knob can be endlessly turned right or left, and when you press the encoder, it often additionally works like a switch (i.e. to confirm a setting).
To register when the knob turns, the Rotary Encoder internally has a disk with holes and two switches. When you turn the encoder knob, the disk turns. Whenever a hole in the disk passes the two switches, they get contact to GND, one after the other.
When you turn the knob in one direction, then switch A gets contact before switch B, and when you turn the knob in the other direction, it is the other way around. The frequency of impulses sent by both switches indicates the speed in which the knob is turned.
Some Rotary Encoder have a third (and independent) switch that is activated when the knob is pressed.
The picture below shows a raw Rotary Encoder at the left side. Often, they are mounted to a simple breakout board like the one on the right side:
Internally, a Rotary Encoder is a very simple construct consisting of a disk and two or three switches. When you connect a Rotary Encoder directly to your microcontroller, then you have to do all the heavy-lifting yourself:
- Debouncing: mechanical switches bounce (when pressed, they can vibrate and send more than just one impulse). You would have to electrically or in software debounce the encoder input.
- Signal Analysis/Post-Processing: as the encoder is just sending two phase-shifted High/Low signals from its two internal switches, your software needs to figure out the direction of phase shift to determine the rotational direction, and analyze the signal frequency to determine the rotational speed.
Fortunately, there are a lot of libraries available that take care of these things. Two things remain, though, that the libraries can't leverage when you work directly with Rotary Encoders:
- Many GPIO Pins: You need one wire per switch plus GND, so depending on whether the encoder has an additional press switch, between 3 and 4 wires and between 2 and 3 GPIO pins.
- CPU Load: Whenever the Rotary Encoder moves, your CPU has work to do, on top of the monitoring load.
That's why there are also smart Rotary Encoders: all of the hassle described above is shiftet over to a dedicated microprocessor. Here is a picture of such a smart Rotary Encoders:
You can clearly see the dedicated microprocessor that makes this Rotary Encoder smart.
Such boards typically communicate via IC2: just two wires are needed (not four), and just two GPIO pins. These GPIO pins are shared among all other I2C devices, too. If you increase the number of Rotary Encoders in your project or use other I2C devices like OLED screens, the number of required GPIO stays the same.
- Mechanical Rotary Encoder: I hook up a "pure* mechanical Rotary Encoder to a microprocessor. This bare-bone example works beautifully for simple use cases and does not need any special libraries. It illustrates the essentials you need to know when working with Rotary Encoders.
Tags: HID, Control, Encoder, User Interface, Switch
Visit Page on Website - created 2024-02-27 - last edited 2024-02-27