🔥 Advanced Heating Control

[tco95ttocs (Mario)]

@panhans

Nope, you can specify the step size. Also floats are possible.

thx…i just overread the stepsize

@panhans

It will run calibration with default values. (Except Popp, Danfoss, Hive)

good to know

@panhans

The reference is always the room temperature. 22 is in this case the temperature to hold the room temperature on 16°C, but with time the differences between TRV and room temperature will equalize a little and the calibration offset will be renewed.
You can also play a little with aggressive mode if you want faster reactions.

yep i understand it theoreticaly, but the last days i do some testing with 3 rooms! All 3 have equale valves from AVM (301) and each room has the same wifi-temperature sensor (over tuya). The Automation of all rooms are identical with there own sensors.
There is only one difference between them, is the position of the valves to the windows:

Room 1: valve on the other side of the room
Room 2: valve about 1m away from the window
Room 3: valve directly below the window

The external sensor are about 2m away from the window in alll 3 rooms.
So normaly the automation should work identical in all rooms! When the windows are closed all is fine and worked as exapted.
If the windows are open and the heating stops than only room 1 and room 2 work normal and there is a negativ offset, so that the valves are closed.
BUT in Room 3 the valve closed only a short time, than it open again ad heating goes on! Why? I don’t know. I think that the temperature at the TRV drops to fast, so that the calibration creates wrong values, with the result of heating and not cooling!

But the rest works fine, excelent work :)…i have to do some more tests with the “physical temp change” in the dev version

[robert5785]

Hi
I have 5 valves and external thermometers configured in better thermostat (radiator 1, radiator2, … radiator5)
I would like the automation to activate the “switch.furnace” entity between 5:30 a.m. - 6:30 a.m. and 4:00 p.m. - 10:30 p.m. if any of the temperature sensors detects that it is lower than the set temperature (e.g. bedroom min. 22, living room min. 21, etc. ), heat until the set temperature is reached or until the specified time 6:30 / 22:30) and turn off the switch.stove entity. if the “to” time is reached during the time (6:30 or 22:30), the automation and the switch entity are to be turned off. the furnace, if only “x” valves out of 5 “report” the demand for heat, it should only heat there - the other thermostats should be closed.
in the remaining time intervals 6:30 - 16:00 and 22:30 - 5:30 the temperature is to be maintained everywhere at 19.5 degrees
can you help me?

[hensch93 (Schumacher Hendrik)]

Hey there,
I have installed automation in all my rooms. I have Tado on the radiators. I get the following error message in the logs.

Logger: homeassistant.components.automation.heizung_wohnzimmer
Source: components/automation/__init__.py:615
Integration: Automatisierung (documentation, issues)
First occurred: 11:38:00 (5 occurrences)
Last logged: 11:49:00

Error rendering variables: ValueError: Template error: round got invalid input '-1.5' when rendering template '{% set n = namespace(dict=[]) %} {% if is_calibration and is_generic_calibration == false and entry_calibration %} {% for valve in valves_tado %} {% set offset_old = state_attr(valve, iif(is_metric, 'offset_celsius', 'offset_fahrenheit')) | float %} {% set local_temperature = state_attr(valve, 'current_temperature') | float %} {% set calibration_sensor_temperature = states(input_temperature_sensor) %} {% set offset = float(calibration_sensor_temperature) - float(local_temperature) %} {% set offset_new = ( float(offset) + (offset_old | float(0)) ) | round(rounding) %} {% if (float(offset_old) - float(offset_new)) | abs >= float(input_calibration_delta) %} {% set n.dict = n.dict + [(valve, [{'value': offset_new}])] %} {% endif %} {% endfor %} {% endif %} {{ dict.from_keys(n.dict) }}' but no default was specified
Error rendering variables: ValueError: Template error: round got invalid input '-1.490000000000002' when rendering template '{% set n = namespace(dict=[]) %} {% if is_calibration and is_generic_calibration == false and entry_calibration %} {% for valve in valves_tado %} {% set offset_old = state_attr(valve, iif(is_metric, 'offset_celsius', 'offset_fahrenheit')) | float %} {% set local_temperature = state_attr(valve, 'current_temperature') | float %} {% set calibration_sensor_temperature = states(input_temperature_sensor) %} {% set offset = float(calibration_sensor_temperature) - float(local_temperature) %} {% set offset_new = ( float(offset) + (offset_old | float(0)) ) | round(rounding) %} {% if (float(offset_old) - float(offset_new)) | abs >= float(input_calibration_delta) %} {% set n.dict = n.dict + [(valve, [{'value': offset_new}])] %} {% endif %} {% endfor %} {% endif %} {{ dict.from_keys(n.dict) }}' but no default was specified
Error rendering variables: ValueError: Template error: round got invalid input '-1.1799999999999997' when rendering template '{% set n = namespace(dict=[]) %} {% if is_calibration and is_generic_calibration == false and entry_calibration %} {% for valve in valves_tado %} {% set offset_old = state_attr(valve, iif(is_metric, 'offset_celsius', 'offset_fahrenheit')) | float %} {% set local_temperature = state_attr(valve, 'current_temperature') | float %} {% set calibration_sensor_temperature = states(input_temperature_sensor) %} {% set offset = float(calibration_sensor_temperature) - float(local_temperature) %} {% set offset_new = ( float(offset) + (offset_old | float(0)) ) | round(rounding) %} {% if (float(offset_old) - float(offset_new)) | abs >= float(input_calibration_delta) %} {% set n.dict = n.dict + [(valve, [{'value': offset_new}])] %} {% endif %} {% endfor %} {% endif %} {{ dict.from_keys(n.dict) }}' but no default was specified

Can’t figure out the needed repairs. Can anyone help me?

[panhans]

Thanks for reporting will push a fix in dev version soon. It seems there is a casting error.

//EDIT:

Updated the dev-version

FIX: duration ignorance when automation got triggered by another trigger (only noticeable if you’d using high durations)
FIX: rounding issue with TADO calibration
ADD: warnings in logs if there is something configured wrong, e.g. physical change feature combined with aggressive mode / generic calibration

Also it’s recommended to setup the Uptime integration to make some things work properly like person based presence detection

[tco95ttocs (Mario)]

Hi i done some test with the old and the new dev-version, and the parameter “physical temperature change” is still not working without problems. In some situation it works fine, in other not. I have 5 blueprints, on 2 it works fine, on one its not possible to change the temperatur manually in th ui or at the valve, and the other 2…i dont know, somtimes work sometimes not!

So it is possible to go back to the previous dev? With this i thought the physical temp change worked fine!

[panhans]

I only made one change to this feature in the dev version. This feature isn’t affected by the changed I’ve made in later versions.
I will do some tests in near future. Just one question: How many thermostats do you use in your automations?
I just wish the home assistant developers would fix this old bug. This would save a lot of struggle. The saddest thing about this is that somebody just did a mergeable pull request with fix for this, but no dev had a look into this to give an OK for this and now the automatism marked this pull request as stale just because nobody cares.

[tco95ttocs (Mario)]

i have some different combinations:

1. one thermostat without external sensor (6 automations)
2. one thermostat with external sensor as “better thermostat sensor” (3 automations)
3. two thermostats wiith one external sensor as “better thermostat sensor” (1 automation)

The physical change i tested only with combination 1 and 3! For combination 2 i have a second scheduler, so there is actually no need for the physical change!
But for combination 1 i need that!!

Update: it looks like that only one situation brings the physical change to work: If i set the predefined settings(eco and comfort) in the UI! Then the changes will take effect. If i use the valve or the slider in the UI it go back to the old temp after a short time!

[Timsche2210 (Timo)]

last point at the blueprint “log level”

Where can I find the logs?
I will double check if the events will start correctly with the correct temperature…

I have the feeling that some of my input numbers or so are maybe causes a problem D:

[omgitsdango1 (Dango)]

I saw that you’ve added calibration support for aqara trvs. Are the aqara-srts-a01 supported?

[panhans]

@Timsche2210

last point at the blueprint “log level”

Where can I find the logs?

If you didn’t setup debugging for this automation in your configuration you can select log level warning. The log entries will be shown in the default home assistant log section. (Settings → System → Logs)

@omgitsdango1

I saw that you’ve added calibration support for aqara trvs. Are the aqara-srts-a01 supported?

Yes, its the Aqara E1, right? I’ve testet and implemted this with help of some users. It should work with z2m but there were some issues related to specific z2m versions. At least your device should provide an input_number entity for an external temperature sensor. Some people had problems that the calibration value which was set in home assistant doesn’t appear in z2m but like I said this is some kinde related to the z2m version. Maybe it’s fixed and stable now.

[tco95ttocs (Mario)]

@tco95ttocs

Update: it looks like that only one situation brings the physical change to work: If i set the predefined settings(eco and comfort) in the UI! Then the changes will take effect. If i use the valve or the slider in the UI it go back to the old temp after a short time!

As i can say now, changing to the predefined values will not work as expacted! It works just for some minutes and than it fall back!

@panhans

I just wish the home assistant developers would fix this old bug. This would save a lot of struggle. The saddest thing about this is that somebody just did a mergeable pull request with fix for this, but no dev had a look into this to give an OK for this and now the automatism marked this pull request as stale just because nobody cares.

Is it the bug linked in your blueprint? Is there a way to bring it up in the list? Maybe a new issues with a hint to the old one and as expanation, that it is not only in the “sun”-integration!

[panhans]

@tco95ttocs

Is it the bug linked in your blueprint? Is there a way to bring it up in the list? Maybe a new issues with a hint to the old one and as expanation, that it is not only in the “sun”-integration!

I’ve updated the link in the dev version. But both issues are aiming on sun trigger or events that return results. I think the issue is in a higher abstract layer so much triggers are affected. I’ve also added a comment to the sun trigger issue that the issue also appears when using template triggers.

[tco95ttocs (Mario)]

thats fine, i also added a comment there :)…As interim solution i use a second scheduler with always heating and a switch in the ui, this seems to work. But thats still not optimal

[Timsche2210 (Timo)]

I set the following stats:

Static Eco Temperature: 15 °C
Static Comfort Temperature: 19 °C
Comfort Temperature: input_number.temperatur_bad (18-28 °C) → currently set 22 °C
Scheduler: schedule.heizplan_bad
Party mode: timer.heizen_bad_manuell
Reset Temperature: yes
Off Instead Of Eco: yes
Start Party Timer On Physical Change: yes

why is now heating on at 22 °C if the timer (party mode) is not active?

[panhans]

A trace log would be great to encounter the problem.

[JorgWolski]

In v3 there has been an option to force minimum temperature instead of off.
In my case, the Danfoss TRVs (via ZHA) present the heating on/of feature, but just return an error when turning the TRVs off. They just support minimum temperature.
Due to that behavior the window open function doesn’t work anymore. Strangely this also happens, when enabling the option to use Eco instead of off.
In both cases the blueprint fails with the following error.
Failed to write attribute system_mode=<SystemMode.Off: 0>: <Status.INVALID_VALUE: 135>

[Propboot]

Sorry for the delay. I made a new test run with the latest dev version and the error occurred again. I pasted the full trace here:
https://controlc.com/2b5af716
Thanks for your help!

[Timsche2210 (Timo)]

do you need this one?

79f2d0e3 - Heating 20240229

[zapp7]

Thank you for all the work on this!

Does version 4.0.2 work with the new proximity integration? I saw the v4 RC1 update up thread but not sure which version that is. I have 4.0.2 and proximity doesn’t seem to work for me.

[panhans]

Sry, fist. Now I’m healthy again will check your issues soon.

@JorgWolski

In v3 there has been an option to force minimum temperature instead of off.

Thanks for reporting. Will check this and provide a hot fix for that issue.

//EDIT: Just check out the dev version for a test. I’ve also renamed that option.

@zapp7

Does version 4.0.2 work with the new proximity integration?

Yes, now the new one is supported.

@Timsche2210

do you need this one?

Best way is to set the mode of one thermostat to off or manipulate the temperature manually. Then just trigger the automation. You also can navigate to the last 5(?) traces with the arrows in the graph view. Make sure the graph didn’t end in false conditions.
You also can identify this in the first lines of the log. In your case:

{
  "trace": {
    "last_step": "condition/0/conditions/5",
    "run_id": "2f1ea237a769143a49267afd86d5830d",
    "state": "stopped",
    "script_execution": "failed_conditions",

@Propboot

Sorry for the delay. I made a new test run with the latest dev version and the error occurred again. I pasted the full trace here:
https://controlc.com/2b5af716
Thanks for your help!

Could you post this service call in your developer tools? Just switch to yaml and paste and execute this:

service: number.set_value
data:
  value: -2.5
  entity_id: number.heizung_wozi_lokaler_temperatur_offset

This is the service call that result in that error. In my configuration I also use this kind of calibration. If it results in an error just try an integer like -2. If you execute this and an error occurs the button flashes red and an popup with an error appears otherwise the button flashes green. Waiting for feedback again.