/cover-rf-time-based

Time-based cover with customizable scripts or entity to trigger opening, stopping and closing. Position is calculated based on the fraction of time spent by the cover traveling up or down. State can be updated with information based on external sensors.

Primary LanguagePython

NOTE

Since I moved the time-based functionality to an ESPHome node, which works much better as it's being done in hardware, I'm archiving this repository and stop maintaining it.

Anyone feel free to fork it and continue development as needed.

Cover Time Based script/entity

Cover Time Based Component for your Home-Assistant based on davidramosweb's Cover Time Based Component, modified for native cover entities, covers triggered by RF commands, or any other unidirectional methods.

Note: Since ESPHome v1.15.0 (September 13, 2020) it is possible to implement a Time Based Cover entirely in the Sonoff RF Bridge hardware, which excludes the necessity of this component (at least of the RF part). Jump to the bottom of this readme for an example how to set it up.

With this component you can add a time-based cover. You either have to set triggering scripts to open, close and stop the cover or you can use an existing cover entity provided by another integration which does not have timing or status feedback. Position is calculated based on the fraction of time spent by the cover travelling up or down. You can set position from within Home Assistant using service calls. When you use this component, you can forget about the cover's original remote controllers or switches, because there's no feedback from the cover about its real state, state is assumed based on the last command sent from Home Assistant. There's also a custom service available where you can update the real state of the cover based on external sensors if you want to.

You can adapt it to your requirements, actually any cover system could be used which uses 3 triggers: up, stop, down. The idea is to embed your triggers into scripts which can be hooked into this component via config. For example, you can use RF-bridge or dual-gang switch running Tasmota or ESPHome firmware integrated like in the examples shown below.

When you use it to extend functionality of an existing cover implementation, this component will generate a new cover entity with these new features.

The component adds two services set_known_position and set_known_action which allow updating HA in response to external stimuli like sensors.

The entities generated by this component have assumed states, which are remembered between Home Assistant restarts.

Support forum

Installation

hacs_badge

  • Install using HACS, or manually: copy all files in custom_components/cover_rf_time_based to your /custom_components/cover_rf_time_based/ directory.
  • Restart Home-Assistant.
  • Create the required scripts in scripts.yaml.
  • Add the configuration to your configuration.yaml.
  • Restart Home-Assistant again.

Usage

To use this component in your installation, you have to either set RF-sending scripts to open, close and stop the cover (see below), or use an existing cover entity provided by another integration, which is missing the features provided here.

Example configuration.yaml entry

Add the following to your configuration.yaml file:

cover:
  - platform: cover_rf_time_based
    devices:
      my_room_cover_time_based:
        name: My Room Cover
        travelling_time_up: 36
        travelling_time_down: 34
        close_script_entity_id: script.rf_myroom_cover_down
        stop_script_entity_id: script.rf_myroom_cover_stop
        open_script_entity_id: script.rf_myroom_cover_up
        send_stop_at_ends: False #optional
        always_confident: False #optional
        device_class: garage #optional
        availability_template: "{{ is_state('binary_sensor.rf_bridge_status', 'on') }}" #optional

OR:

cover:
  - platform: cover_rf_time_based
    devices:
      my_room_cover_time_based:
        name: My Room Cover
        travelling_time_up: 36
        travelling_time_down: 34
        cover_entity_id: cover.myroom
        send_stop_at_ends: True #optional
        always_confident: True #optional
        device_class: curtain #optional
        availability_template: "{{ not (is_state('cover.myroom', 'unavailable') or is_state('cover.myroom', 'unknown')) }}" #optional

All mandatory settings self-explanatory.

Optional settings:

  • send_stop_at_ends defaults to False. If set to True, the Stop script will be run after the cover reaches to 0 or 100 (closes or opens completely). This is for people who use interlocked relays in the scripts to drive the covers, which need to be released when the covers reach the end positions.
  • always_confident defaults to False. If set to True, the calculated state will always be assumed to be accurate. This mainly affects UI components - for example, if the cover is fully opened, the open button will be disabled. Make sure to set the current state after first setup and only use this entity to control the covers. Not recommended to be True for RF-based covers.
  • device_class defaults to shutter if not specified. See the docs for availale cover device classes.
  • availability_template if not specified will make the cover always available. You can use a template returning True or False in order to toggle availability of the cover based on other entities. Useful to link with the connection status of your RF Bridge or relays device.

Example scripts.yaml entry

RF covers

The following example assumes that you're using an RF bridge running Tasmota or ESPHome open source firmware to integrate your radio-controlled covers. The command scripts pass the rfraw_data parameter to a general transmitter script which takes care of queuing the transmission of the codes and keeping an appropriate delay between them:

'rf_transmitter':
  alias: 'RF Transmitter'
  mode: queued
  max: 30
  sequence:
    # With Tasmota
    - service: mqtt.publish
      data:
        topic: 'cmnd/rf-bridge-1/rfraw'
        payload: '{{ rfraw_data }}'
    # With ESPHome
    - service: esphome.sonoff_rf_bridge_send_rf_raw
      data:
        raw: '{{ rfraw_data }}'
    # add a little delay
    - delay: 00:00:01

'rf_myroom_cover_down':
  alias: 'RF send MyRoom Cover DOWN'
  mode: single
  max_exceeded: silent
  sequence:
    - service: script.turn_on
      target:
        entity_id: script.rf_transmitter
      data:
        variables:
          rfraw_data: 'AAB0XXXXX....XXXXXXXXXX'

'rf_myroom_cover_stop':
  alias: 'RF send MyRoom Cover STOP'
  mode: single
  max_exceeded: silent
  sequence:
    - service: script.turn_on
      target:
        entity_id: script.rf_transmitter
      data:
        variables:
          rfraw_data: 'AAB0XXXXX....XXXXXXXXXX'

 'rf_myroom_cover_up':
  alias: 'RF send MyRoom Cover UP'
  mode: single
  max_exceeded: silent
  sequence:
    - service: script.turn_on
      target:
        entity_id: script.rf_transmitter
      data:
        variables:
          rfraw_data: 'AAB0XXXXX....XXXXXXXXXX'

Note that for RAW data you also need the Portisch firmware to be flashed on the EFM8BB1 embedded RF-transmitter chip of the bridge unit.

For the scripts above with Tasmota you need a small automation in automations.yaml to set RfRaw back to 0 to avoid spamming your MQTT server with loads of sniffed raw RF data. This trigger is checked every minute only so set > 40 set in the value_template to be a bit bigger than your biggest travelling_time:

- id: rf_transmitter_tasmota_cancel_sniff
  alias: 'RF Transmitter Tasmota cancel sniffing'
  trigger:
    platform: template
    value_template: "{{ ( as_timestamp(now()) - as_timestamp(state_attr('script.rf_transmitter', 'last_triggered')) | int(0) ) > 40 }}"
  action:
    - service: mqtt.publish
      data:
        topic: 'cmnd/rf-bridge-1/rfraw'
        payload: '0'

Switched covers

The example below assumes you've set send_stop_at_ends: True in the cover config, and you're using any two-gang switch running Tasmota open source firmware to integrate your switch-controlled covers:

'rf_myroom_cover_down':
  alias: 'Switches send MyRoom Cover DOWN'
  mode: single
  max_exceeded: silent
  sequence:
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER1' # open/close
        payload: 'OFF'
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER2' # power
        payload: 'ON'

'rf_myroom_cover_stop':
  alias: 'Switches send MyRoom Cover STOP'
  mode: single
  max_exceeded: silent
  sequence:
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER2' # power
        payload: 'OFF'
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER1' # open/close
        payload: 'OFF'

'rf_myroom_cover_up':
  alias: 'Switches send MyRoom Cover UP'
  mode: single
  max_exceeded: silent
  sequence:
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER1' # open/close
        payload: 'ON'
    - service: mqtt.publish
      data:
        topic: 'cmnd/myroomcoverswitch/POWER2' # power
        payload: 'ON'

Note how you don't have to configure these as switches in Home Assistant at all, it's enough just to publish MQTT commands strainght from the script (credits to VDRainer for this example). Of course you can customize based on what ever other way to trigger these 3 type of movements. You could, for example, turn on and off warning lights along with the movement.

Services to set position or action without triggering cover movement

This component provides 2 services:

  1. cover_rf_time_based.set_known_position lets you specify the position of the cover if you have other sources of information, i.e. sensors. It's useful as the cover may have changed position outside of HA's knowledge, and also to allow a confirmed position to make the arrow buttons display more appropriately.
  2. cover_rf_time_based.set_known_action is for instances when an action is caught in the real world but not process in HA, .e.g. an RF bridge detects a stop action that we want to input into HA without calling the stop command.

cover_rf_time_based.set_known_position

In addition to entity_id and position takes 2 optional parameters:

  • confident that affects how the cover is presented in HA. Setting confident to true will mean that certain button operations aren't permitted.
  • position_type allows the setting of either the target or current posistion.

Following examples to help explain parameters and use cases:

  1. This example automation uses position_type: current and confident: true when a reed sensor has indicated a garage door is closed when contact is made:
- id: 'garage_closed'
  alias: 'Doors: garage set closed when contact'
  trigger:
  - entity_id: binary_sensor.door_garage_cover
    platform: state
    to: 'off'
  condition: []
  action:
  - data:
      confident: true
      entity_id: cover.garage_door
      position: 0
      position_type: current
    service: cover_rf_time_based.set_known_position

We have set confident to true as the sensor has confirmed a final position. The down arrow is now no longer available in default HA frontend when the cover is closed. position_type of current means the current position is moved immediately to 0 and stops there (provided cover is not moving, otherwise will contiune moving to original target).

  1. This example uses position_type: target (the default) and confident: false (also default) where an RF bridge has interecepted an RF command, so we know an external remote has triggered cover opening action:
- id: 'rf_cover_opening'
  alias: 'RF_Cover: set opening when rf received'
  trigger:
  - entity_id: sensor.rf_command
    platform: state
    to: 'open'
  condition: 
  - condition: state
    entity_id: cover.rf_cover
    state: closed
  action:
  - data:
      entity_id: cover.rf_cover
      position: 100
    service: cover_rf_time_based.set_known_position

confident is omitted so defaulted to false as we're not sure where the movement may end, so all arrows are available. position_type is omitted so defaulted to target, meaning cover will transition to position without triggering any start or stop actions.

cover_rf_time_based.set_known_action

This service mimics cover movement in Home Assistant without actually sending out commands to the cover. It can be used for example when external RF remote controllers act on the cover directly, but the signals can be captured with an RF brigde and Home Assistant can play the movement in parrallel with the real cover. In addtion to entity_id takes parameter action that should be one of open, close or stop.

Example:

- id: 'rf_cover_stop'
  alias: 'RF_Cover: set stop action from bridge trigger'
  trigger:
  - entity_id: sensor.rf_command
    platform: state
    to: 'stop'
  condition:[]
  action:
  - data:
      entity_id: cover.rf_cover
      action: stop
    service: cover_rf_time_based.set_known_action

In this instance we have caught a stop signal from the RF bridge and want to update HA cover without triggering another stop action.

Icon customization

For proper icon display (opened/moving/closed) customization can be added with option device_class set either in the cover's config, based of what type of covers you have.

Can also be done in configuration.yaml:

homeassistant:
  customize_domain: #for all covers 
     cover:
      device_class: curtain
  customize: #for each cover separately
    cover.my_room_cover_time_based:
      device_class: curtain

More details in Home Assistant device class docs.

Some tips

When using this component with Tasmota RF Bridge in automations

Since there's no feedback from the cover about its current state, state is assumed based on the last command sent, and position is calculated based on the fraction of time spent travelling up or down. You need to measure time by opening/closing the cover using the original remote controller, not through the commands sent from Home Assistant (as they may introduce some delay).

Tasmota RF bridge is able to send out the radio-frequency commands very quickly. If some of your covers 'miss' the commands occassionally (you can see that from the fact that the state shown in Home Assistant does not correspond to reality), it may be that those cover motors do not understand the codes when they are sent 'at once' from Home Assistant.

This can be handled in multiple ways:

  • try increasing your RF range. Make sure the wire antennas of the covers are not tied close to the power cables or to big metallic surfaces. For 433MHz, the antenna length should be around 17cm (this may include the part going inside the tube motor). Sonoff RF Bridge has two copper helical antennas near the PCB, you can unsolder them and simply solder in place two straight hard wires of 17.3cm, which can go out through some small holes on the sides of the unit. You need to solder only one end of each wire, to the points where the helical legs were shorter (points U7 and U8). This will increase the range substantially, to the cost of aesthetics.
  • avoid backlogs with rfraw AAB0XXXXX....XXXXXXXXXX; rfraw 0 if you need multiple covers opening and closing at once. Switching the sniff on and off quickly for every cover movement may cause issues. It's enough to send rfraw 0 only once with some delay after all procedures related to cover movements finished, the example scripts above take care of that.
  • if you are sending 0xB0 codes (decoded with BitBucketConverter.py) you can tweak those to be sent with repetitions (multiple times) by changing the repetition parameter (5th byte) of the code. For example 20 repetitions can be achieved by changing 5th byte from 04 to 14. Also BitBucketConverter can be run by specifiying the required repetitions at command line before decoding. Some covers might not like this, though.
  • alternatively, you can further reduce stress by making sure you don't use cover groups containing multiple covers provided by this integration, and also in automation don't include multipe covers separated by commas in one service call. You could create separate service calls for each cover, moreover, add more delay between them:
- alias: 'Covers down when getting dark'
  mode: single
  max_exceeded: silent
  trigger:
    - platform: numeric_state
      below: 400
      for: "00:05:00"
      entity_id: sensor.outside_light_sensor
  action:
    - service: cover.close_cover
      entity_id: cover.room_1
    - delay: '00:{{ (range(1,10)|random|int) }}:00'
    - service: cover.close_cover
      entity_id: cover.room_2
    - delay: '00:00:02'
    - service: cover.set_cover_position
      data:
        entity_id: cover.room_3
        position: 20
    - delay: '00:00:01'
    - service: cover.set_cover_position
      data:
        entity_id: cover.room_4
        position: 30

Implementation with ESPHome

Use the following configuration for ESPHome on Sonoff RF Bridge:

substitutions:
  device_name: sonoff-rf-bridge
  friendly_name: "RF Bridge"
  device_ip: 192.168.81.22

esphome:
  name: ${device_name}
  platform: ESP8266
  board: esp01_1m
  esp8266_restore_from_flash: true

wifi:
  ssid: !secret wifi_ssid
  password: !secret wifi_password
  manual_ip:
    static_ip: ${device_ip}
    gateway: 192.168.81.254
    subnet: 255.255.255.0
  use_address: ${device_ip}

logger:
  baud_rate: 0
  level: INFO

uart:
  tx_pin: GPIO01
  rx_pin: GPIO03
  baud_rate: 19200

rf_bridge:

api:
  reboot_timeout: 15min
  password: !secret api_password
  encryption:
    key: !secret encryption_key
  services:
    - service: send_rf_raw
      variables:
        raw: string
      then:
        - rf_bridge.send_raw:
            raw: !lambda 'return raw;'

ota:
  password: !secret ota_password

web_server:
  version: 2
  local: true
  port: 80
  auth:
    username: !secret web_server_adminuser
    password: !secret web_server_password

sensor:
  - platform: wifi_signal
    name: ${friendly_name} WiFi signal
    update_interval: 60s
  - platform: uptime
    name: ${friendly_name} Uptime

status_led:
  pin:
    number: GPIO13
    inverted: true

binary_sensor:
- platform: status
  name: ${friendly_name} Status

button:
- platform: restart
  name: ${friendly_name} Restart
  
cover:
- platform: time_based
  name: 'My Room Cover'
  device_class: shutter
  assumed_state: true
  has_built_in_endstop: true

  close_action:
    - rf_bridge.send_raw:
        raw: 'AAB0XXXXX....XXXXXXXXXX'
  close_duration: 34

  stop_action:
    - rf_bridge.send_raw:
        raw: 'AAB0XXXXX....XXXXXXXXXX'

  open_action:
    - rf_bridge.send_raw:
        raw: 'AAB0XXXXX....XXXXXXXXXX'
  open_duration: 36s

For multiple covers to be operated simultaneously, a delay is needed between sending out the codes, this can also be done within ESPHome, see detailed configuration example.