Modulating boilers eg OpenTherm

[Si-Hill]

Description

Boilers that support modulation alter the temperature of the water going to the radiators based on various factors such as the difference between current and target temperatures or the % opening of the TRV. Some also take the outside temperature into account when calculating the output flow temperature. I'm sure the algorithms are far more complex than that too, eg Tado.

I imagine this could affect the RoomMind learning if it predicts the time to heat a room based on a previously different flow temperature.

Could RoomMind have a global "boiler flow temperature" option set by an existing entity and take that into account in the model? It would be global and not per room as all rooms would receive the same temperature water at any one time.

Use case

I used Tado on an OpenTherm boiler which varies the flow temperature based on demand. I do have an entity from Tado CE that reports the current flow temperature.

Alternatives considered

No response

[Jaz666]

This information is mainly for @snazzybean , who might not be familiar with the eccentricities of Tado.

I'm one such user, who fully endorses the "low and slow" approach to hot water radiator heating.
I've been running RoomMind since public launch alongside the official Tado integration.

A week ago I switched to the Tado CE integration with both Weather and Room Compensation (which is light years ahead of Tado's default Room Compensation).
The integration has the ability to control the boiler flow temp, and I've observed my boiler running at much lower temps than with the default Tado Modulating setup.
Whilst Tado CE could take over the scheduling and calls for heat, I'm sticking with RoomMind for "the best of both worlds".

For example, it's currently 7 degrees outside, RoomMind is calling for heat from 5 rooms, but as each of them are only just below the target threshold, Tado's Modulating control is calling for a max of 11% heating power.
Temperature Compensation has set the boiler down to 45 degrees (out of a max of 65), and Room Compensation has bumped that up slightly to 47.5 to factor in the number of rooms calling for heat.

I've not noticed any incompatibilities between RoomMind and Tado used in this manner yet.

Some observations:

• With Tado, set the climate entries in RoomMind to "Direct (device regulates)" to retain Tado Control of the temps (if you don't, RoomMind will set Tado to 25 degrees, overheating the rooms and negating Modulating efficiency gains).
• RoomMind seems to work best with regular temperature readings, I'm lucky enough to have the max Tado API calls, if you are on the 100 calls a day limit, it might not be enough.
• It's common for RoomMind to call for heat for extended periods of time, but the boiler is not actually firing.

This is yesterday's graph

[Si-Hill]

@Jaz666 The problem I have with setting the TRV's to Direct mode is that RoomMind sets the TRV to the target temp and doesn't take my additional temp sensors into account. So RoomMind sees a room at 18.5c, sets the TRV to 19c, but the TRV's own temp is 19c already so doesn't call for any heat or even activate the boiler pump. I've looked at Tado CE but haven't figured out how to use it alongside RoomMind, although I am using Tado via HomeKit so avoid a lot of API calls there. It would be good to contact you directly somehow if that's ok, fb messenger?

[dns13]

I back this request as I also control the feed temperature from my gas boiler with home assistant depending on outdoor temperature, target temperature, difference between current and target temperature, weather and sun irradiance on the main room. This results in quite volatile heat output of my radiators which I assume will decrease the accuracy of the model (could not test this, as currently not really much heating is required, and only using the integration for a few days). I have target and current temperature available and I think that would help the model to have this information available.