Vaillant Arotherm Owners Thread

Setback is its own thing with its own settings.
Room Temp Mod: Normal, Active and Expanded have no bearing on the curves.

Room Temp Mod has 3 options to choose from.

  • Inactive = Pure WC (the Sensocomfort room stat is nothing but a display)
  • Active = WC and Room Influence (depending how far away room temp is from indoor target the system can adjust the flow temp, i.e. increase it if room needs a little more heat)
  • Expanded = WC and Room influence, but also acts as on/off stat switching the heating off if room temp is exceeded.

Have you got “set-back mode” set to Eco or Normal (you want it to be Normal)
(settings, installer level, install config, circuit1)

Then in control, zone 1, heating, you want to set “setback temp” to whatever you desire. ie 18C

Using these setting, setback will only kick in when you don’t have a timed schedule set.

So in this scenario, between 6am and 10pm the heat pump will target 20C internally, using the curve to pick the flow temp. see curve info here:

Then outside of that schedule, so 10pm to 6am, the heatpump will target 18C internally (your chosen setback temp), again, using the curve to pick the flow temp.

ie, flow temp should be around 10C lower when 2C difference in internal temp.

As you can see here on mine, there is a distinct drop in flow temp (and electric in / heat out) between the on periods on the setback times.

My setup is setback 19C, main target 20C.

Does that help?

1 Like

Thanks, I think that all fits with my understanding (some of which is no doubt gleaned from your blog previously - very helpful, thanks!).

System: 12kW AroTherm+, twin zone (up/downstairs) with new big rads throughout, 300l DHW cyl.

Rough history:

  1. Original settings from installer: Expanded, curve 0.75, setback mode Normal;
  2. Edged down to 0.6 in initial “getting to know it” period (still on Expanded), 19°C target with overnight 18°C setback.
  3. Switched to Cosy Octopus in mid-Dec, which has worked well. Boost to 20°C in morning cheap period, 19°C during the day, boost to 21°C in afternoon cheap period, coast across peak period with 18°C setback, back to 19°C for evening, 18°C setback overnight. Phew! This is definitely in the “Cosy incentivises by price, at cost of raw CoP” category. But has given us a 15-20% bill saving.
  4. In recent cold/freezing fog snap in early Jan, noticed undershooting of targets. System doesn’t auto-compensate for heating time lost to defrost cycles.
  5. Trying out suggestion of using Active rather than Expanded. If anything, it’s now overshooting. Edging down the heat curves further (now at 0.4) to see what happens (have been dubious of this before now, as long setback recovery time in the mornings has been detrimental to comfort, so we’ll see).
  6. As mentioned before, will go back to previous settings (as per point 3 above) for now, and revisit once our PV/battery install is complete (half done, got snowed off yesterday, argh!)
  7. Want to get some proper monitoring in place, but may be an “ahead of next winter” project.


1 Like

Is there a setting that would cause DHW to heat outside of a programmed schedule?

I have double checked the schedules in the MyVaillant app this morning and we do not have one configured for Sunday evenings. The state of the heat pump switched from HEATING to STANDBY between 20:52 and 21:06, shortly after we had two showers which had reduced the tank temperature to 12°C

The graph in Home Assistant shows it rising back up to 19.5°C and the flow and return shown in Emomcms reflect a mini DHW cycle.

N.B. The DHW (blue background) feed in Emoncms comes from a HA helper I have created, which is only based on the heating circuit being in standby, flow temp > 20 and a flow rate > 0.3 so that’s not potentially a true reflection of the Vaillant being in DHW mode.

I might be misunderstanding the curve calculations but I think there’s a mistake in Vaillants example and hence in your calculations in the first post.

The figure you posted and your tables claims that for a setpoint room temperature of 21°C, all flow temperatures get shifted by 5°C compared to 20°C setpoint. That value is too large. I think Vaillant just picked that example to have a nice round number (45°C) in their manual.

The 45 degree angle parallel shift of the curve actually leads to a flow temperature change that depends on the outside temperature especially when it’s warm outside. The reason for this is that the shifted curves have to converge at the point Outside Temperature = Setpoint temperature. The original 20°C curves all converge at 20°C and the same must be true for the different setpoints.

An approximate equation that can be fitted to the heat curves (adapted from some German forum where I stumbled over the general idea) is:

TFlow = a * (HC * (Tset - Tout))b + Tset

HC is the heat curve parameter (e.g. 0.3), Tset is the desired room temperature and Tout is the outside temperature. a and b are the parameters of this model that can be fitted to Vaillants curves. The heat curve parameter acts as a multiplier on the indoor-outdoor temperature difference, which is then taken to the b-th power and multiplied by a to get the deltaT the flow temperature needs to be above the set point room temperature.

Using the figure you posted, I extracted the data points using WebPlotDigitizer. I then fitted the model to the data and arrived at

a = 2.55
b = 0.78

The calculated flow temperatures match what I read from my AroTherm plus’ desired flow temperature at a specific outside temperature via eBus and are off by only ~0.07 °C currently. I changed the desired indoor temperature on the controller and at 8.625 °C outside temperature got desired flow temperatures on the heatpump of:

19°C set → 25.25°C flow
20°C set → 26.71°C flow
21°C set → 28.16°C flow

My model yields 25.18 °C, 26.64 °C and 28.09°C respectively.

For a curve of 0.4 at -15°C outside, I get 39.97°C flow which fits Vaillants 40°C in the manual.


Hi folks,

Quick question on Tank temperature, my reported tank temperature doesn’t seem to relate to my tap temperature.
I have a 250l Vaillant tank, I can have a warm (not quite hot) shower when the tank temperature is reporting 15C. So I read the temperature now more like full (50C) or empty (12C).

Any explanations, I’d kind of expected a bit of an offset then direct relationship. On a Home Assistant forum the provider of the vaillant integration has confirmed his tank temperature correlates more directly to tap temperature.

Thanks, Will

The temperature sensor is probably near the bottom of the tank, so you’re seeing the temperature of the cold water that has come in to replace the hot water that has been drawn from the top. The cooler water will stay near the bottom, and the warmer water will stay near the top, known as the “thermocline”.

There’s likely an amount of hot water left in the cylinder, but it’s not a simple task to know how much. Some cylinders have a pocket for a second sensor, I don’t know if the Vaillant one does.

1 Like

Those calculations certainly reflect more closely what I’m seeing with respect to actual flow temperture at various outside temperatures. I have the curve set at 5.5 and target internal at 21C. Only on the coldest day last month did the flow temperature reach 40C.

Using the Valliant app, I’m getting COPs that are too good to be true for my 5kW unit. For example, last month (Jan-24), the app is reporting a COP of 5.4 (592 consumption; 3202 kWh generated). Even on the coldest day (18th Jan) when outside temperature was as low as -6C, the app is reporting a COP of 4.4.

The consumption data look fairly accurate when compared to the meter consumption, but the output data seem too high. Are there settings anywhere that could cause the output to be over estimated?

Thanks for any feedback.

Same for me - have a look at the post I made here: Vaillant inbuilt monitoring vs MID meters - #40 by Andre_K

The problem is that the Vaillant measured flow temperatures are too high - for me around 1.2 K - while the return temperature measures ok. This translates to a too high deltaT for the internal heat calculation. Depending on what your typical flow-return deltaT is, this can be quite drastic. If you’re typically running a deltaT of 10K then the 1 K offset will be only a 10% overestimate, for my scenario with a deltaT of around 3-4K 'm looking at a 30-40% overestimation. You’ll find your individual offset by looking at flow&return temperatures during a time when the compressor is off (between cycles in warmer weather).

So in summary - nothing you can really do about it to get better values to show in the Vaillant app. If you have access to the ebus data you can calculate an approximate correction factor.

1 Like

Thanks for that. I think I’m in the higher category regarding overestimation. Although it would be nice to have accuate efficency mesurements, at the end of end of the day, it’s consumption and cost that matters.

It’s certainly true that cost matters most, but I was able to accurately estimate the actual heating needs of my house just based on my gas usage when I still had a gas heater because I could easily infer the actual heating energy going into the house. It’s much easier to judge improvements to insulation etc. when looking at actual heat energy delivered to the house and not only electrical energy that gets multiplies by a COP that in itself depends on a number of external factors.

1 Like

Interested to know where you found this additional information regarding Room Temp Mod

" * Inactive = Pure WC (the Sensocomfort room stat is nothing but a display)

  • Active = WC and Room Influence (depending how far away room temp is from indoor target the system can adjust the flow temp, i.e. increase it if room needs a little more heat)
  • Expanded = WC and Room influence, but also acts as on/off stat switching the heating off if room temp is exceeded."

The user manual for the SensoComfort is very vague and in my experience i can’t tell any difference between active and inactive, we use ‘Expanded’ to make sure the unit shuts off in the case of solar gain (or other heat sources in the house causing desired temp to overshoot)… HeatGeek seem to advocate pure weather comp. but in reality this would just be wasting money (and overheating the house) in the case of solar gain, esp when you have a lot of glazing and south east facing side of the house

Anyone else have connection problems with the My Valliant app? I’ve had Gateway unavailable eroors for the past 24 h. I’m trying to determine if it’s me or a server problem.

I’ve heard there’s a new version of the app - have you updated it?

1 Like

Thanks, that was it. Updates usually happen automatically on my phone. Fir some reason this one didn’t.

I use Active when it gets cold (sub zero) as this helps minimise defrosts.
Expanded otherwise.

One thing I’ve learnt is that if you go away … ensure on Expanded with 12C min temp.

New 5KW install, level 3 monitoring. Have done quite a bit of reading etc. prior to specifying. OEM has been a useful resource, thank you creators and everyone who runs it.
There was a 10 month gap between our delivery and installation, but hey, can’t do everything all at once :no_mouth: Seems only the SensoNet has been replaced in that time (new, free to a good home :laughing:).

Does anyone run just using the indoor heat pump controller? -no sensoHome/Comfort etc? This only works for weather compensation as there is no room/building 'stat/controller.
The HP controller reconfigures itself if a sensoComfort is connected, and you loose access to some of the functions. Some of these can be tweaked from the SensoComfort.

I was a bit surprised the sensoComfort requires a separate wireless box and a wireless outdoor sensor (both included). I figured the wireless version would connect to the myVaillant box directly and/or have an option for wires, and could take the outside temperature from the installation, the HP intake or already hardwired external sensor. Should probably have gone for the wired sensoComfort :person_shrugging: looks like there maybe ebus pads accessible on the back of the sensoComfort…

Does anyone have any experience with the sensoRoom? (I can’t seem to find it available in the UK without being part of a bundle). Was wondering if it would act as a basic controller or if it also needs the sensoComfort in the system?

What emitters do you have? I’ve tried expanded but simply doesn’t work with UFH in a passive slab as the lag time is too long.

Active will deal with solar gain too. Expanded can be a good idea to reduce idle if i) heat curve is already dialed-in ii) reaction time of house isn’t very long, but otherwise “active” is generally best option if you can put the controller in a location that is central and not directly impacted by solar gain.

@AndyS Did you ever try reducing “Compr.start heat. from” to increase cycles? I’ve tried, but it seems to ignore the new value (-100), even if i restart everything.