TRVs horribly inaccurate - any suggestions?
Answers
-
Hi
I've wondered both those things, too!
Currently updating every 5 minutes, but considering only updating if more than 0.5 degrees out.
I'm only just starting with HA, not sure how to script the updates, any chance you could share your script?
Jonathan0 -
Deployed and updated on github
allowed variation can be set by user in the code
0 -
Hi guys,
This is an amazing topic. However, I would like to throw something else in the mix: Have you played with OpenHab ? It has a Tado biding. That binding would definitely welcome this parameter added. I am struggling with the same offset issue, but all my other data I already have it in OpenHab.
At the minute I am struggling with understanding and changing the OpenHab Tado binding (JAVA) and I would appreciate if any of you would be interested in adding the possibility to update the offset parameter via the OpenHab Tado binding...
0 -
@Jon_On_A_Bike, thank you for making the code available. I currently have it running on a Wemos D1 mini clone. I did find it necessary to increase the memory to the dynamicjsondocument otherwise it would fail with the 'nomemory' error in just under two hours.
I have now added the code for only updating if more than 0.5 degrees out, to your code for the ESP8266 and that has been running without issue for nearly a week now. I am staying with the 8266 for the moment since I have more of them on order.
Keep up the great work!
1 -
I installed tado and 8 TRVs in December. All TRVs are vertical.
So far, I've found no issues with the temperature being influenced by the proximity of the valve to the rad. I'll post a photo of one of the rads later but I think the type of rad may play quite a big part in the accuracy-1 -
Can't seem to be able to post pics
The end plate of ours is just a cover so doesn't produce as much heat as the one in the OP's post. I wonder if that could be why I don't see the same problem?0 -
I’d like to add another use case for this API, and ask for some openhab help.
My Tado valves are unrepresentative of the room temp and need a big offset when the heating is on, but they’re reasonably accurate with heating off. So they’re always wrong for some of the day.
The offset needs to change depending on whether the heating is on or not. This seems a simpler solution than needing additional sensors, and would at least get the valves closer, if not quite perfect.
I can see heating mode, heat demand, temperatures, etc in openhab, so I can set up a rule to make an api call to set the offset when the heating mode changes in any particular room.
What I need help with is making this api call from openhab (running on an rpi). Does anyone have any experience with openhab and how to achieve this?
Offset is not available through the existing openhab-Tado integration.
Side question: is there a reason Tado couldn’t implement this as an optional software setting? I’m thinking a toggle switch underneath the offset value. Dynamic changes to offset based on heating mode would potentially solve a lot of problems people are having with their valves.
0 -
@Yog, I have written a python script that does exactly that; unfortunately it is not as simple as checking if the heating is on or off; that is because, if you turn the heating on for half an hour, you still need to keep the offset active for some time, without abrupt changes; i.e. in order for this to work, I found I need to keep the history of the heating demand of the past x hours
I started working on it with some decent results (although not as good as using an externl thermometer) but I needed to record data for the room for a coupe of days and then retrospectvely calibrate 3 or 4 parameters
0 -
I’d thought of the same issue. Openhab can record the heating demand over time, so managing or adding a delay is not a problem.
Just need to get the openhab binding updated...
0 -
I have managed to get the smart TRVs to work correctly by fitting an Oventrop adapter between the radiator and Tado valve. I have created a new post with more details in to make it easier for people to find, see here:
1 -
I have managed to get the smart TRVs to work correctly by fitting an Oventrop adapter between the radiator and Tado valve. I have created a new post with more details in to make it easier for people to find, see here:
0 -
Hey
Just came across this so thought I would post a comment to say I have been working on this too.
I use Home Assistant so have targeted that as a solution and did a PR to PyTado and to HA Tado integration. So now the offset can be updated via automations in Home Assistant and I wrote an automation for it as a kind of template for anyone using Home Assistant to copy and paste, which manages all devices in the house (including a battery saver option).
My Home Assistant automation write up is here: https://github.com/north3221/home-assistant-config
It was discussed on Home Assistant here: https://community.home-assistant.io/t/tado-integration-put-offset/175964
and here: https://community.home-assistant.io/t/battery-level-and-temperature-offset-for-tado/81068
If you are not using Home Assistant, you could still take advantage of the PyTado library: https://github.com/wmalgadey/PyTado
0 -
I have bought oventrop adapter and it helped a lot in stabilization, there is still spikes but not as bad as before (more reliable).
What also helped further is speedcomfort radiator fans (you can find them here https://www.speedcomfort.nl/) These are great devices that dissipates heat accumulated on top of the radiator to the room by pushin air from below. A positive side effect is heat generated by radiator not affecting the smart radiator thermostat as it is being diverted faster so better temperature measurement happens.
I would suggest both of these (speedcomfort fans actually saves money as well, in case you want to learn the science behind this is a detailed test report https://www.speedcomfort.nl/media/downloads/Validation_Report_SpeedComfort___Final_30.08.2019_(signed).pdf they found that up to 20% additional radiator capacity and 11% cost reduction can happen with these fans)
1
