Behavior of auto_regulation_use_device_temp

[jbayart]

Hello,

I am not sure about the behavior of the option auto_regulation_use_device_temp.

So I am using VTerm with a climate entity and basically, without this option, this is what happens : I set a setpoint to 18°C via eco mode of VTerm. Vterm sets 18°C as a setpoint on the underlying. The underlying TRV starts heating when below 18°C and immediatly stops since its internal temperature sensor is wrong. This leads to a room temperature below target.

What I would like is that Vterm periodically checks the offset between room temperature and underlying temperature and set a new setpoint taking into account the new offset.

Here is what I think currently happens with the option on but please correct me if I am wrong:
Without having self regulation on, the target temperature only change when the preset change, and so the offset is only used and applied when preset is changed.
What happened last night is that the preset for the night has been applied with a target temp of 18°C, and so offset has been applied so target temp on the underlying has been set to 19.5°C instead of 18°C. The underlying starts heating and quicky goes above its target according to its sensor, and stops heating. The TRV think its doing its job fine whereas the target temp is far away from the actual temp.

Is this correct? Would it be possible to update the offset even if the actual target temp from VTerm is not updated?

Thank you for all your work,
Jérémie

[jmcollin78]

Hello @jbayart ,

In most case, you should not use the auto_regulation_use_device_temp. I have done this for people who comes from Better Thermostat but it is a very bad way to do temperature regulation.

You should rather use the Strong regulation or very strong regulation (see expert mode) if you need it.

What kind of TRV do you have? If you have some Sonoff TRVZB, you should rather use the 'direct control of valve regulation'. See here: https://github.com/jmcollin78/versatile_thermostat/blob/main/documentation/en/self-regulation.md

[jbayart]

Ok I will disable this option and use self-regulation for now.
Just curious: why is it a bad way to regulate the temperature? Is it because of very frequent adjustements? I feel like strong regulation introduces some kind of inertia I don't really need and it seems like a duplicate to the TRV algorithm in my case, which does already a correct job.

I am using the tado X TRV. I don't have access to the opening of the valve directly, so I don't think I can use 'direct control of valve regulation'

[jmcollin78]

The internal temperature sensor is often bad because it is too near of the heater itself. It is varying too much in too large proportion. That does a bad and unstable regulation.