Netatmo and fireplace problems

[Meky]

So, I did test today and moved both radior valves from Bedroom to Living room just to see what will happen. After around 30 min in new enviroment their temperature maxed at 21°C.
Original living room termostat maex at 22°C.
Real temperature was almost 28°C at the place of thermostat (given by high temperature in the room due to fire place running and fact radiator was hot as Netatmo did not recognized this high temperature).



It seems to me valves are just ignoring high temperature jumps for some reason.

[kozmix]

Hi Meky, that indeed looks suspicious. That may indeed be something that Netatmo support could confirm, as there may be some logic built in to "debounce" temperature jumps. I'd expect though that they start to move upwards slowly. I agree this situation doesn't look logical. Hopefully Netatmo support will be able to investigate further based on the above picture!

[Meky]

Well I have only bad experiences with their support. I have addressed this issue twice and got always bullshit answer which pretty clearly showed noone bothered to read my inquiry.

I tried one more thing today - I have disabled the function managing heating while opened window is detected. Thought it might have influence so I'll check more in next days.

[568ml]

For how long were the valves in the room @Meky? Do you think that the valves are 'stuck' at 21/22 degrees, and that they will never go any higher, or could it be that they are extremely slow in sensing temperature changes?

I'm reminded of a thread I read through on here a couple of years ago that discussed this 'lag' in perceiving temperature updates. It was regarding the thermostat (not the valves), but a seemingly very well-informed user posted a number of helpful graphs and commented that the thermostat's temperature sensor is effectively insulated by the plastic casing, which he considered a design flaw.

I wonder if a similar thing could be happening with the valves?

I found the thread, the user I mentioned even drilled holes in the casing of his thermostat to enable air to flow into the internals:
https://forum.netatmo.com/viewtopic.php?f=20&t=7373

Also in my search I came across this which may also be relevant:
https://forum.netatmo.com/viewtopic.php?f=20&t=5301

[kozmix]

It was regarding the thermostat (not the valves), but a seemingly very well-informed user posted a number of helpful graphs and commented that the thermostat's temperature sensor is effectively insulated by the plastic casing, which he considered a design flaw.

I've also thought about this before, as a typical thermostat has ventilation holes so air can easily circulate through. I've also experienced that my Netatmo thermostat responds much slower to temperature changes than other thermometers in the room. I agree this is a design flaw where likely "design" has taken priority over detection effectiveness.

[AcidBurn]

Hi here,

Keep in mind that as the valve is connected directly to the radiator, Netatmo can't use directly the temperature measured on the valve who will not be the same as the whole room when eating.
So It's maybe the algorithm who is responsible to evaluate the real temperature of the room who might be the source of the problem here?

Do you think it's possible?

[kozmix]

Keep in mind that as the valve is connected directly to the radiator, Netatmo can't use directly the temperature measured on the valve who will not be the same as the whole room when eating.
So It's maybe the algorithm who is responsible to evaluate the real temperature of the room who might be the source of the problem here?

It must have something to do with it. I assume the valve's temp sensor is at the "cold end" under the battery lid, but still it means you're measuring very closely to the radiator which isn't representative for the rest of the room. However that is something the PID algorithm should compensate for.

[Meky]

Yes, there will be algorithm considering what room temperature will vs what the temperature around the radiotr is. My above picture shows 22°C valve vs. 28°C close to the valve. If I move the external thermometr to middle of the room I got for example 22°C on valve and ~25° in the middle of the room. So the problem is really somewhere withing the programming there.

[Daegil_Netatmo]

Hello!

There is indeed a estimation of the temperature in the middle of your room because the closer you are to the radiator, the hotter is even if the middle of the room is cold. If you wish to check that the temperature sensors (yes there are two) of the Valves are working well, please compare the estimated temperature with the middle of the room and not next to the radiator.

Cheers!

[Meky]

Hello,
I have done that already. Today example:
Middle of room = 23,9°C
Radiator valve = 22,5°C