Tado <-> Inventum heatpump: Can I modify the OT signal?

[jrs]

Hi there!

I'm very new to this forum and to central heating stuff, so please bear with me if I say something stupid

TL;DR: is it possible to adjust a signal received before it's sent out again?

The situation I'm facing now is that since I've got a heat pump installed in between my Tado and Remeha boiler (central heating system) the 0% modulation signals from the Tado are not always interpreted right. This has the effect that the heat pump does not stop, which is not really what it was meant to do.

I've logged a case with Tado and they really did put some effort into this, but concluded that their system did the proper thing. Unfortunately the help from inventum was negligible, they basically said I should have used an 'approve thermostat' from their list of approved thermostats.

I do have a quick and dirty fix for this issue, namely an automation in my home assistant that detects when the heat pump is active (by measuring the power consumption) while the Tado has 0% total heating request, the automation will send an 'away' signal to Tado which (apparently) triggers an 'OFF' signal that the heat pump does understand and it stops.
Of course this is not ideal, that's why I started searching online and came across the Opentherm Gateway. I immediately ordered a gateway from nodoshop because it's fascinating stuff (not received the OTG yet... )

My grand question to this forum is: when I get a "0% modulation" signal from Tado, is it possible to convert this signal into an "OFF" signal to the heat pump with OTG?

Thanks a lot for your input and thoughts!

- Joris

[hvxl]

I'm not yet convinced that the Tado is doing "the proper thing". The only modulation parameter available to the thermostat in the opentherm protocol is MsgID 14: Maximum relative modulation level setting. The term 'relative' in that description is important. That means that it uses a scale between the minimum and maximum absolute modulation the boiler can do. So 0% doesn't mean off. It means on at the minimum possible level. There is a CHenable bit in MsgID 0: status, that actually tells the boiler to stop heating when cleared (0). From your description I get the impression that bit isn't switched off by the Tado unless you send it the 'away' signal. MsgID 1: control setpoint, may also come into play.

You should be able to determine exactly what is going on once you install the OTGW. The OTGW allows you to modify the CHenable bit. But it doesn't have any built in logic to do that automatically when the max relative modulation level is 0%. You see, it is quite normal for a thermostat to request the boiler to burn at 0% relative modulation when just maintaining the room temperature. So you will have to implement your unusual fix in home assistant or something similar.

When your OTGW arrives, please share a log of the incorrect behavior. I'd be interested to see it and try to determine which side is to blame.

[jrs]

Thanks for your reply hvxl!

I'll upload the logs when the OTG arrives and I captured the strange behaviour!

- Joris

[Yocee84]

Hi!

I have nearly same configuration.

rendszer.jpg (20.08 KiB) Viewed 8069 times

I has 4 tado thermostat, 4 tado trv, 1 wireless extension. One OTGW, one Remeha Tzerra, one Hajdu HPAW ( Midea M-Therma monoblock heathpump, with modbus conroll), and a working Home Assistant system.

My problem the next:
I controll the Remeha boiler with tado perfectly (minus the zone pumps, The zone pumps controlled by Home Assistant from tado's % power signal)
I want to controll the heathpump with tado. (The perfect solution is for me the heathpump boiler signal interrupt otgw gpio input, and the Remeha stop working. Is it possible?) Now, the Home Assistant send CH=0 when i dont want use the boiler, and send the CH=1 for starting. Is it working in service mode, but not working is automatisation. Second, i try set low the setpont. It is almost perfect, but the boiler pump is working always.

thanks

[hvxl]

I'm sorry, I don't understand your question about interrupting the OTGW GPIO input.

With most boilers, the pump will be running when CHenable is 1. It stops after a delay when CHenable is changed to 0. If your Home Assistant keeps CH at 1 when CS is low, that is probably the reason the pump keeps running.

[Yocee84]

Sorry! We have 2 free GPIO. A and B. I need when i pull GPIO A to low (or high...) the CH=0 is activate. And when i open the circuit, the CH=1 is back again. (Like DHWBlock eeg. CHblock...)
If it is work, tha standalone aplication is good, without Home assistant.

[hvxl]

This functionality doesn't currently exist. And I wonder if it will be useful to anyone else. At least you're the first to ask.

There is also the complication that the CH command works in combination with the CS command. So just making the GPIO input mimic the CH command may not be the right thing to do. I will have to think about that.

Of course there's always the option of adding this functionality yourself. The source code is available.

[jrs]

When your OTGW arrives, please share a log of the incorrect behavior. I'd be interested to see it and try to determine which side is to blame.

My OTGW did arrive and I have installed it today and observed the behaviour again, see graph:

Screenshot 2022-11-27 at 22.11.22.png (121.39 KiB) Viewed 7737 times

I've also attached the logs from otmonitor between 19:45 and 22:00. I'm very curious about your conclusion from this log.
Please bear in mind that the graph lines named "Bathroom heating" and "Woonkamer heating" are reading from the Tado API, so they might be delayed.

Thanks!

[hvxl]

As soon as your boiler start heating at minimum power (19:58:16), the water temperature shoots up and quickly exceeds the control setpoint (19:58:31). At least with a regular boiler that is not what I expect. The newly heated water would normally be pumped around the system and cold water returning from the circuit takes its place. As a result, the water temperature only rises slowly, over several minutes.

The behavior captured in your log file seems to me indicative of a flow problem. Maybe your pump is not working properly, or there is a blockage in your system. That way the heated water stays mostly in the boiler. And only that small amount gets heated, causing the temperature to rise very quickly.

Unfortunately, I don't have experience with heat pumps. So I can't say for sure that they behave similarly to regular boilers. But I would expect so.

In any case, the behavior of the Tado seems totally reasonable to me. It requests heating for only 3 minutes. It is not strange that it doesn't immediately end the request when the boiler stops heating after a mere 25 seconds.

[jrs]

Thanks for checking the logs!

The heat pump is active (= heating) during the better part of the logs, the blue line in the graph shows the power usage, which is unfortunately the only thing I can measure from the heat pump. This might explain why the water temp is rising so fast? The heatpump is pre-heating for the actual boiler (Remeha Calenta). I don't know which boiler is sending the "Boiler water temperature", the heat pump or the Remeha.

I did order a second OTGW, so I'll solder it together and put that one in between the heat pump and my boiler. Maybe I'll be wiser after comparing that data. Would you care to see those logs too?

My question remains, why is it active all the time and doesn't it stop heating after the relative modulation level is down to 0.0
The heatpump did stop on it's own just before 22:00, but I can't seem to find any clue in the OT data supporting that, any thought on that one?

[hvxl]

The heat pump seems to stop at 21:55, which is the same time the room setpoint changes from 20 to 18 degrees. At that time also the room temperature suddenly drops from 19.95 to 18.69 degrees, which doesn't sound very believable. But in any case, 21:55 is the first time the room temperature is significantly higher than the room setpoint. Earlier it was never more than 0.1 degrees higher.

So my hypothesis is that the heat pump does its own thing based on the room temperature and room setpoint, rather than listen to the master like it should. It might be interesting to see what it does when you instruct the OTGW to withhold this information: UI=16 and UI=24. If it doesn't like that, you may need a new feature to modify the room setpoint that is sent to the boiler (or in your case, the heat pump).

I am definitely interested to see side by side logs of both sides of a heat pump. That may help me understand what those devices do a bit better.

[jrs]

The heat pump seems to stop at 21:55, which is the same time the room setpoint changes from 20 to 18 degrees.

This is because different rooms have different setpoints, 20 is my livingroom and 18 is my bathroom (which I always like the be a little heated). At 22:00 the living room is set to low by the Tado schedule.

I did also notice that the setpoints send from Tado looks kinda random, during the day it switches from 18 to 21 (my office) degrees, but with your hypothesis in mind there might be more to it.

Screenshot 2022-12-03 at 10.42.08.png (32.34 KiB) Viewed 7163 times

So my hypothesis is that the heat pump does its own thing based on the room temperature and room setpoint, rather than listen to the master like it should. It might be interesting to see what it does when you instruct the OTGW to withhold this information: UI=16 and UI=24. If it doesn't like that, you may need a new feature to modify the room setpoint that is sent to the boiler (or in your case, the heat pump).

I guess you're right and I'd like to request that new feature
Yesterday morning I issued the UI=16 and UI=24 commands and the only time the heat pump did stop was during it's night clock. Even the 'away' signal did not stop the pump, I guess (with the knowledge gathered now) because it only sets the target temp lower. Once I issued the KI commands (during away mode) the heat pump immediately stopped, see graph:

Screenshot 2022-12-03 at 10.24.05.png (63.98 KiB) Viewed 7163 times

Legend:

• blue: heatpump power draw

• purple: Relative modulation level

• cyan: accumulated heating percentage read from Tado API

I am definitely interested to see side by side logs of both sides of a heat pump. That may help me understand what those devices do a bit better.

Will do!

[hvxl]

Yesterday morning I issued the UI=16 and UI=24 commands and the only time the heat pump did stop was during it's night clock. Even the 'away' signal did not stop the pump, I guess (with the knowledge gathered now) because it only sets the target temp lower. Once I issued the KI commands (during away mode) the heat pump immediately stopped

Probably the heat pump stored the last room temperature and setpoint values it received and kept on using those. You could try to send the UI=16 and UI=24 commands when the heat pump is off. Or better yet, restart/powercycle the heat pump after you performed the UI commands.

[jrs]

I issued the UI command's in 'away' mode and restarted my heat pump. It started heating again after a 'relative modulation' level, but didn't stop anymore, so I issued the KI commands again, after which it did stop immediately.

Screenshot 2022-12-06 at 22.23.33.png (46.85 KiB) Viewed 6761 times

Screenshot 2022-12-06 at 22.25.44.png (74.39 KiB) Viewed 6761 times

Any suggestions for a fix I can create with otgw or in combination with homeassistant?

I haven't gotten around to solder the second otgw unfortunately

[hvxl]

It looks like the heat pump only started a minute or two after relative modulation went to 100%. So that probably wasn't the trigger. As usual, logs would be needed for more detail.