Thanks for a great integration. Really excellent work and thank you for taking the time to reply.
Yes I have been using opening degree AND had set the calibration offset. I have not using the internal temp of the device (unless by accident setup error or bug), I’m using sensors elsewhere in the room (mixture of Awair, Sonoff and Aqara). These are the ones I have specified under Room Temperature in the “Main” VTherm config page for each. I have never had “use underlying internal temperature of the underlying” ticked for any of my TRVs.
I think there may be a fundamental logic issue the Sonoff TRVZBs firmware which prevents them from responding to certain valve set requests when the underlying hvac_action state conflicts with the command. They have to be in “heating” state to accept an increase in the valve percentage value and not heating state to accept a set request to a decreased percentage value.
It was this post in another thread that made me aware of this ‘feature’ of TRVZBs, and I’ve confirmed that I am seeing the same behaviour with latest firmware:
I have implemented a similar workaround to the above user. I have automations that set the target of the underlying to 30C whenever hvac_action of the VTherm changes to heating. Similarly I have automations that set the target of the underlying to 10C whenever hvac_action of the VTherm changes from heating to anything else.
This way when the VTherm attempts to adjust the valve opening % this actually gets applied and I hear the valve motor change. This seems to work well with sensible fine adjustments made by the VTherm to the opening percentage to hit and maintain the temperature around the set point. As only the VTherm is shown on my dashboard, it doesn’t matter to me what the target temperature of the underlying is set to as it’s not seen anywhere.
However if anyone has Sonoff TRVZB valve control working reliably without requiring a workaround similar to the above I’d appreciate seeing their specific settings. If the underlying internal temp perfectly mirrored the external temp (i.e. instant replication tracked to within 0.1C) this workaround wouldn’t necessary as there’d never be a conflict. I’ve never been able to get this to happen satisfactorily, either with my own automations via Z2m commands, Better Thermostat or Versatile Thermostat.