Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/robmarkcole/circuitpython-on-home-assistant

Programming and managing circuitpython boards from Home-Assistant
https://github.com/robmarkcole/circuitpython-on-home-assistant

circuitpython home-assistant

Last synced: 4 days ago
JSON representation

Programming and managing circuitpython boards from Home-Assistant

Awesome Lists containing this project

README 

        
## Project goal

My inspiration is the [esphomelib](https://esphomelib.com/) project, which makes it easy to integrate DIY hardware with [Home-Assistant](https://www.home-assistant.io/). Where my project differs, is that I will use CircuitPython instead of C++ on the boards (and a fully python 3 backend), meaning that users only need to know python to make a contribution to the project. Users will also enjoy the rapid development cycle possible with micropython and a REPL. Home-Assistant will provide an IDE and tools to make it easy to flash firmware to the board, and provide a framework for integrating the board with other services already exposed to Home-Assistant. This should get a lot of the tedious data and networking 'plumbing' aspects of DIY projects out of the way, so users can focus on making cool hardware for their home automation projects. Note - I've selected Circuitpython over Micropython as the Circuitpython ecosystem (code and docs) are well maintained and curated by Adafrtuit, whilst Micropython is more a wild-west of code and documents.  



## Home-Assistant setup

I am running [Hassbian](https://www.home-assistant.io/docs/installation/hassbian/) distribution of [Home-Assistant](https://www.home-assistant.io/) on a pi 3B+ and am using the Cloud9 IDE addon, with setup process described [here](https://www.hackster.io/robin-cole/tensorflow-object-detection-with-home-assistant-7cc04b).










## CircuitPython for ESP8266

Most of the currently available CircuitPython boards lack wifi, which is a limitation for some home automation projects. Fortunately we can use CircuitPython on the ESP8266 boards which have both wifi and bluetooth (we will be using wifi and [MQTT](https://github.com/micropython/micropython-lib/tree/master/umqtt.simple) for posting data to Home-Assistant). We therefor get the benefits of the nicely curated CircuitPython ecosystem, as well as low cost hardware with wifi. CircuitPython can run on both the official feather-huzzah-esp8266 board, but also a raw ESP8266 module (but with incorrectly mapped pins). I will develop this project using the feather-huzzah-esp8266 board, and then later add specific instructions for using the the raw ESP8266 board. However since most of the instructions will be identical, I will just refer to 'the ESP8266 board' in the following instructions. Note that CircuitPython boards without wifi will also be supported, and these can be integrated to Home-Assistant using the serial port as I have done [for other projects](https://github.com/robmarkcole/HASS-circuitpython-air-quality-sensor-node).



## Adafruit feather huzzah esp8266

The official esp8266 board supported by CircuitPython is the [feather-huzzah-esp8266](https://learn.adafruit.com/adafruit-feather-huzzah-esp8266). This board has connection for a battery, and the pinouts are correct, so it is the baseline board for this project.










## Flashing firmware

The official instructions for flashing the board firmware are [here](https://learn.adafruit.com/welcome-to-circuitpython/circuitpython-for-esp8266). We use the [esptool](https://github.com/espressif/esptool) for this process, and I created a new directory `circuitpython` using the IDE:



1. `sudo pip3 install esptool` -> install the esptool and check it was successful by entering `esptool.py -h` which prints the help info.

2. `sudo curl -OL https://github.com/adafruit/circuitpython/releases/download/3.1.1/adafruit-circuitpython-feather_huzzah-3.1.1.bin` -> download the firmware `.bin` file, here for Circuitpython release 3.1.1. Check [here](https://github.com/adafruit/circuitpython/releases/latest) for the URL to the latest Circuitpython release.

3. `esptool.py --port ESP8266_PORTNAME erase_flash` -> erase the current firmware, where in my case `ESP8266_PORTNAME` is `/dev/ttyUSB0`

4. `esptool.py --port ESP8266_PORTNAME --baud 115200 write_flash --flash_size=detect 0 firmware.bin` -> write the new firmware, again replacing `ESP8266_PORTNAME` with your port and replacing `firmware.bin` with `adafruit-circuitpython-feather_huzzah-3.1.1.bin`. Note that for the raw esp8266 I had to use slightly different settings, changing the baud and adding `-fm dio` -> `esptool.py --port ESP8266_PORTNAME --baud 460800 write_flash --flash_size=detect -fm dio 0 firmware.bin`



You have now updated the board firmware, which in this example results in the board running CircuitPython 3.1.1. Note that the above process worked without problems for both the feather-huzzah-esp8266 and the raw esp8266 board. Note also that exactly the same process is used to flash [vanilla micropython](https://docs.micropython.org/en/latest/index.html) firmware to the boards, the firmware files are [here](http://micropython.org/download#esp8266).



## Circuitpython REPL from Home-Assistant

General advice on connecting to a Circuitpython REPL over serial (USB) is [here](https://learn.adafruit.com/welcome-to-circuitpython/advanced-serial-console-on-mac-and-linux). The following advice is specific to connecting to the ESP8266 board via the Cloud9 IDE, and assumes you are logged in as the default `pi` user.



1. `sudo apt-get install screen` -> install [screen](https://www.gnu.org/software/screen/), if you have never done so before.

2. `ls /dev/tty*` -> list available devices/terminals, and I identified the ESP8266 by plugging and unplugging the board and checking what was added to this list. My boards shows up as `/dev/ttyUSB0`

3. `screen /dev/ttyUSB0 115200` -> connect to the board via screen. If you get the error `Cannot open your terminal '/dev/pts/2' - please check.`, run `script /dev/null` and try again.

4. Do whatever you want to do via the REPL. To exit screen session `control + A + D`

5. `screen -list` -> To check running screen processes, useful if you become detached, this will return something like `6248.pts-3.hassbian     (20/11/18 20:37:25)     (Detached)`. To kill that process `screen -S 6248 -X quit`










## Circuitpython files

Adafruit publish a python package [ampy](https://github.com/adafruit/ampy) which we can use to explore files on the board, as well as post `.py` files to the board.



1. `sudo pip3 install adafruit-ampy` -> to install, if you haven't before. Note sudo is required

2. `ampy --port /dev/ttyUSB0 ls` -> list files, note that screen cannot be running (unplug and replug the board to close screen and reset the connection).

3. Create a `main.py` file in Cloud9, then post and view it using ampy:










## Data plotting

Home-Assistant integration will allow logging of board data to a SQL database for long term storage. Home-Assistant can also be used for plotting this data, but its plotting capabilities are pretty basic. Therefore the scope of this project will also include the development of tools for plotting board data, for example using Bokeh.