Curious if anyone was aware of any unexpected behavior when using light transition automations on MQTT-driven lights such as Tasmota or Shelly devices.
The following automation is used as part of the morning routine. Essentially, it starts a lighting transition, as well as turning up the heat, in our bedroom 30 minutes before we plan on getting out of bed (it also kills a noise machine).
- id: '16102317315567'
alias: Wake up Routine
trigger:
- platform: template
value_template: '{{ states(''sensor.time'') == (states.input_datetime.wakeup_time.attributes.timestamp
| int - 1800) | timestamp_custom(''%H:%M'', False) }}'
condition:
- condition: state
entity_id: input_boolean.wakeup_enabled
state: 'on'
- condition: or
conditions:
- condition: state
entity_id: input_boolean.wakeup_weekend
state: 'on'
- condition: time
weekday:
- mon
- tue
- wed
- thu
- fri
action:
- service: climate.set_temperature
data:
entity_id: climate.thermostat_bedroom
temperature: 75
hvac_mode: heat
entity_id: climate.thermostat_bedroom
- delay: 00:00:05
- type: turn_on
device_id: 330448934e9740d899bffb2d592f0c37
entity_id: light.headboard_lights_headboard_lights
domain: light
brightness_pct: 1
- delay: 00:00:05
- service: light.turn_on
data:
brightness_pct: 100
transition: 1800
entity_id: light.headboard_lights_headboard_lights
- delay: 00:35
- device_id: 9523966de4494bdcafd8a44e9ad96f8d
domain: switch
entity_id: switch.noise
type: turn_off
mode: single
The ‘headboard lights’ are a Tasmota-flashed LED strip. When they are used in this automation, they turn on at 1% and then, rather than transitioning, immediately jump to 100% after the delay. All other components of the automation work fine. Additionally, if I replace the headboard lights with a Shelly 2 dimmer tied to dimmable bulbs (no integration - straight MQTT), I have the same issue (1% -> 100%, immediately).
If I replace the headboard lights with a group of dimmable Zigbee bulbs that are controlled through a Hubitat (HACS integration using the Maker API), the light transitions work as intended(thus, the entire automation works as intended).
Any one aware of any issues with transition lighting over MQTT? Given that the Tasmota/Shelly devices are controlled via MQTT, while the Hubitat bulbs are over HTTP, I can only assume there’s something going on there.