Xiaomi Mi Scale

Code to read weight measurements from Mi Body Composition Scale (aka Xiaomi Mi Scale V2)

Note: Framework is present to also read from Xiaomi Scale V1, although I do not own one to test so code has not been maintained

Getting the Mac Address of your Scale:

Retrieve the scale's MAC Address (you can identify your scale by looking for MIBCS entries) using this command:

$ sudo hcitool lescan
LE Scan ...
F8:04:33:AF:AB:A2 [TV] UE48JU6580
C4:D3:8C:12:4C:57 MIBCS
[...]

Note down your MIBCS mac address - we will need to use this as part of your configuration...

Setup & Configuration:

Running script with Docker:

Supported platforms:
1. linux/amd64
2. linux/arm32v6
3. linux/arm32v7
4. linux/arm64v8
Open docker-compose.yml (see below) and edit the environment to suit your configuration...
Stand up the container - docker-compose up -d

docker-compose:

version: '3'
services:

  mi-scale:
    image: lolouk44/xiaomi-mi-scale:latest
    container_name: mi-scale
    restart: always

    network_mode: host
    privileged: true

    environment:
      MISCALE_MAC: 00:00:00:00:00:00 # Mac address of your scale
      MQTT_HOST: 127.0.0.1  # MQTT Server (defaults to 127.0.0.1)
      MQTT_PREFIX: miScale
      # MQTT_USERNAME:        # Username for MQTT server (comment out if not required)
      # MQTT_PASSWORD:        # Password for MQTT (comment out if not required)
      # MQTT_PORT:            # Defaults to 1883
      # MQTT_TIMEOUT: 30      # Defaults to 60

      # Auto-gender selection/config -- This is used to create the calculations such as BMI, Water/Bone Mass etc...
      # Up to 3 users possible as long as weights do not overlap!

      USER1_GT: 70            # If the weight is greater than this number, we'll assume that we're weighing User #1
      USER1_SEX: male
      USER1_NAME: Jo          # Name of the user
      USER1_HEIGHT: 175       # Height (in cm) of the user
      USER1_DOB: "1990-01-01" # DOB (in yyyy-mm-dd format)

      USER2_LT: 35            # If the weight is less than this number, we'll assume that we're weighing User #2
      USER2_SEX: female
      USER2_NAME: Serena      # Name of the user
      USER2_HEIGHT: 95        # Height (in cm) of the user
      USER2_DOB: "1990-01-01" # DOB (in yyyy-mm-dd format)

      USER3_SEX: female
      USER3_NAME: Missy       # Name of the user
      USER3_HEIGHT: 150       # Height (in cm) of the user
      USER3_DOB: "1990-01-01" # DOB (in yyyy-mm-dd format)

Running script directly on your host system (if your platform is not listed/supported):

Install python requirements (pip3 install -r requirements.txt)
Open wrapper.sh and configure your environment variables to suit your setup.
Add a cron-tab entry to wrapper like so:

*/5 * * * * bash /path/to/wrapper.sh

NOTE: It's best to schedule via crontab at most, every 5 min (so as not to drain the battery on your scale):

*/5 * * * * python3 /path-to-script/Xiaomi_Scale.py

Home-Assistant Setup:

Under the sensor block, enter as many blocks as users configured in your environment variables:

  - platform: mqtt
    name: "Example Name Weight"
    state_topic: "miScale/USER_NAME/weight"
    value_template: "{{ value_json['Weight'] }}"
    unit_of_measurement: "kg"
    json_attributes_topic: "miScale/USER_NAME/weight"
    icon: mdi:scale-bathroom

  - platform: mqtt
    name: "Example Name BMI"
    state_topic: "miScale/USER_NAME/weight"
    value_template: "{{ value_json['BMI'] }}"
    icon: mdi:human-pregnant

Acknowledgements:

Thanks to @syssi (https://gist.github.com/syssi/4108a54877406dc231d95514e538bde9) and @prototux (https://github.com/wiecosystem/Bluetooth) for their initial code

Special thanks to @ned-kelly (https://github.com/ned-kelly) for his help turning a "simple" python script into a fully fledge docker container

m0rphy / xiaomi_mi_scale