This blog post is a comprehensive review of the third winter living with our Vaillant Arotherm heat pump.
For the purposes of this article I class “winter” from the 1st of November through until the end of April.
I hope you find this real world account useful in some way.
Solar, Batteries and Heat Pump = The Holy Trinity
Solar, Batteries, Heat Pump, EV and Time of Use tariffs = The Renewable Quintessentials
Find more details about the heat pump installation here:
This is our third winter with the heat pump. You can read my two previous write-ups here where you can see mistakes along the way and lessons learnt.
Combined heating and hot water performance again hit SCOP 4 across the coldest 6 months of the year in this third winter.
Month / Year | Electric Input kWh | Heat Output kWh | COP | Outside Low | Outside Avg | Outside High | Avg Room |
Nov-2024 | 343 | 1434 | 4.2 | -4.5 | 6.6 | 13.8 | 19.7 |
Dec-2024 | 356 | 1617 | 4.5 | -0.6 | 6.4 | 12.8 | 19.8 |
Jan-2025 | 571 | 2055 | 3.6 | -5.4 | 2.6 | 10.0 | 19.9 |
Feb-2025 | 453 | 1721 | 3.8 | -2.4 | 4.4 | 14.0 | 20.2 |
Mar-2025 | 331 | 1332 | 4.0 | -2.6 | 7.0 | 16.5 | 20.3 |
Apr-2025 | 180 | 723 | 4.0 | -0.3 | 10.2 | 23.2 | 20.1 |
Winter 2024-2025 | 2234 | 8882 | 4.0 | -2.6 | 6.2 | 15.1 | 20.0 |
Electricity Figures
I have a month by month breakdown of heat pump performance stats going back to November 2022 here of you want even more detail:
These stats show a whole year SCOP in 2024 of 4.23 (heating 4.45 and hot water 3.84)
Friday 10th January 2025.
The COP was not bad considering it was almost -3C all day!!
You can view the data from that day here:
https://emoncms.org/energystatsuk?mode=power&start=1736467200&end=1736553600
In both Winter one and Winter two articles I mentioned TRVs (Thermostatic Radiator Valves).
In Winter one I talked about how I experimented with Tado smart TRV and shutting down various rooms in the evening and when they got warm from solar gain.
This was ultimately a failed experiment as this led to poorer heat pump performance as you’re taking away valuable system volume (water) that the heat pump needs when you’re closing off radiators.
In Winter two I started to remove those Tado TRV and went back to conventional mechanical ones, but still kept a couple of them as limiters in certain south facing rooms (ie, put them at 22C, ie above my target heat pump temperature, so they’d only be triggered by external factors).
For the whole of this third winter I didn’t bother using any TRV. On the rads that still had the older mechanical TRV, I just cranked open fully so they were never activated. Preferring to keep the whole system loop open.
When there is solar gain in winter, I see this as free heating and I’m happy to couple this with low temperature water in the radiators. With it being in the winter I find the rooms eventually come back down to target temperature when the sun goes back in because it’s chilly outside.
With the bonus that you’re still keeping full system volume (water in the radiators) in play for the system to use, which means you’ve got more chance of getting better performance from longer system runs (less cycling).
So at all times now, my system is fully open loop. All 100 litres of water in all the radiators and pipework is available for the heat pump to use.
TRV Tips
Note: You can reduce the temperature in certain rooms by reducing the flow rate into that particular radiator.
Mario Dodić from the HVAC Education Hub explains this concept in this great video
In both winters one and two my Arotherm heat pump was running a main circuit board with firmware revision 351.06.07
As we described in depth in this article, this version has a software bug that affects normal operation of the heat pump and makes it cycle more often than it should.
Vaillant Arotherm Firmware 351.06.07 Problems (Energy Integral)
In the summer of 2024, so before the third winter began, I had the PCB swapped to a new one running firmware 351.09.02
I had to pay for this myself as Vaillant does not acknowledge this issue as a warranty problem, despite my exhaustive investigation.
The good news is that my heat pump is working much better now, like it should do, with cycling vastly reduced.
I have always known our dining room has been the worst performing room in the house. Even before I got all nerdy and started placing sensors everywhere you just knew it was the coldest room.
There is little surprise though as the dining room is a single storey extension with pitched roof at the back of the house on the north side that was built and added around 2000. So whilst there was ‘some’ insulation as per building regs at that time, it wasn’t to a great standard. And with the extension being on the north side it rarely gets much solar gain.
It also doesn’t help that the extension also has two windows AND a large double door patio window either, all three of them only double glazed. If you can see through something you know it’s not a very good insulator.
This room is the very epitome of what Heat Geek calls the “reference room” in your house. The one that dictates the design flow temperature for the rest of the house.
In the previous two winters, despite adding radiators, I had to run the whole system flow temperature a little higher to get the dining room up to temperature, but at the expense of running the other rooms in the house a little warmer than needed to.
The other option was to run the rest of the house at the right temperature, but be okay with the dining room being a little cooler.
A few summers ago I pulled down the dining room ceiling and re-insulated the roof space with new PIR insulation and taped it all, which helped.
In terms of other upgrades, there’s probably not much point going from double glazing to triple glazing when you look at the U values. It doesn’t seem that way from a financial point of view anyway.
And there isn’t much I can do about the original wall construction. There is a cavity and it appears to be filled. Everything is decorated, so there is no opportunity to internally insulate the walls at this stage. External wall insulation (EWI) isn’t possible either.
So that only leaves the suspended floor. Having had the flooring up a few times in the past I knew there was room for improvement. I’d seen what looked like the original polystyrene insulation with the concertina ends that just sat in between the joists, some of which appeared to have fallen to the floor over time.
So in the summer of 2024, with the help of my friend Mike from Practical Property Services, who built my plant room for me, we ripped the whole floor up and did it properly.
The joists were 100mm deep, so we used 100mm PIR insulation between the joists.
Mike had some slats of wood spare from a project so we put some batons underneath the joists to stop the insulation ever dropping.
The PIR was cut very tight and banged into place. We did foam any larger gaps before taping the whole perimeter with Pro Clima Uni Tape as well as taping along every joint between joist and PIR and where pieces of PIR butted up against each other.
This created an airtight seal between the very cold space below and the room above.
Here’s a photo of the work in progress.
The best indicator of “did it make a difference” I can find is the amount of heat generated for space heating only in both winters (less being better).
Here ia breakdown of those months showing the amount of heat in kWh and the average outside temperature for the month.
Month | Winter 23/24 | Winter 24/25 |
November | 1032 kWh (avg 6.5C) | 1038 kWh (avg 6.6C) |
December | 1322 kWh (avg 6.0C) | 1189 kWh (avg 6.4C) |
January | 1955 kWh (avg 4.3C) | 1649 kWh (avg 2.6C) |
February | 1090 kWh (avg 7.1C) | 1359 kWh (avg 4.4C) |
March | 1047 kWh (avg 7.3C) | 967 kWh (avg 7.0C) |
April | 694 kWh (avg 9.2C) | 439 kWh (avg 10.2C) |
Note: we went away for a week in April 2025, hence the total heat required is less that month.
So I’m really happy with the outcome of the work.
Despite this though, the dining room is STILL my reference room. But as explained earlier, I’m not sure what else I can realistically do now to reduce the heat loss in this room any further?
Throughout 2024 and into the 24/25 winter I’ve always struggled getting to grips with the hot water runs on our 5kW Arotherm. I never seemed to choose the right mode.
In the first winter I was in Normal mode (fast recharges, terrible efficiency) and for all of 2024 I ran in Eco mode (slow recharges, fantastic efficiency).
You can read about the different modes here
But you can also read about the struggles I had with Eco mode here:
TL:DR of that article is that a 5kW heat pump running in Eco / slow mode and a 250L cylinder with high hot water usage is a challenging combination.
The bit I struggle with is how much heat the house loses when the hot water is being reheated.
An Eco hot water run for me can be anywhere from 90 minutes to almost 3 hours depending on how empty the cylinder is and how cold it is outside.
Note: The heat pump doesn’t kick out as much power when it’s colder outside.
So if you’re running a setback temperature during the day whilst at work or overnight whilst sleeping AND you also do your hot water runs, it’s then like you’ve got two setbacks occurring.
Which then means when you come out of the hot water reheat and it’s time to bring the house back up to temperature, the house temperature could have dropped for 3-4 hours.
You can see exactly that in this article. When the hot water run started we were already in a setback situation so the room temperature was dropping. This was then compounded by the 3 hour Eco hot water run.
You can link to this hot water reheat yourself here
https://emoncms.org/energystatsuk?mode=power&start=1736123220&end=1736138120
This meant the room temperature was pretty low once the hot water run had finished.
Heat pumps are way more efficient when maintaining a room temperature than having to increase it. They really don’t like working hard to bridge a large gap between current room temp and target.
Anything above 2C between actual temp and target temp presents a real challenge which can take hours to overcome.
And house comfort really suffers when you’ve got those big temperature drops before the room eventually gets back to target temperature.
Much of this problem on the Vaillant is because the load compensation controls aren’t aggressive enough when there is a big difference between internal room temperature and target room temperature.
In the ‘second winter’ article I detailed our heating schedule. But for this third winter I made a slight change based on the setback/hot water findings above, which I’ll describe and detail below.
Here’s how our Weekly Planner looked for this winter.
Weekends
Weekdays
Hot water Runs
As you can see on the heating, I now run a very minimal overnight setback (just 1C) because of the issue with the setback and hot water combination.
I also now don’t bother running a setback during the day on weekdays.
As stated before, my wife only works part time and I occasionally work from home. Coupled with the hot water run at 1PM (a natural setback), we are much preferring “Comfort over COP”
I will happily admit I’ve been chasing COP for the past two winters, many times at the expense of comfort, especially lowering the setback temperature and switching to Eco hot water runs.
But honestly, running it in this new third winter way, with just a slim setback has been way nicer.
It’s got to the stage where you have no idea what temperature it is outside because it’s just so constant and nice indoors.
My wife now calls the house “The Eden Project”
Not because of any high temperature, but because it’s like a controlled and constant environment.
On the Vaillant Sensocomfort thermostat I have that in “Room Influence Mode (active)’. Rather than just sticking rigidly to the weather curve, active mode takes into account the temperature of the room and subtly adjusts the flow temperature of the water into the radiators to reach or maintain room temperature.
You can read about the internal controller modes here:
I spent the winter flipping between 0.45 and 0.50 on the weather compensation curve. This gets me 36.5C or 38C flow at -3C outside for target 20C indoors, which seems to work well in conjunction with ‘active’ room influence.
You can read about weather compensation curve here:
I felt I needed curve setting 0.5 when 2C and below, whilst 0.45 worked okay when 3C and warmer outside. I think it’s something to do with reduced heat output when defrosts start to occur. This is usually around 2C outside and below.
The final “Comfort over COP” choice has been with hot water.
As stated earlier, I ran the first winter in Normal mode and the second winter in Eco mode.
From November to the end of January in this third winter I ran in Eco, to be able to complete the year long Eco mode experiment of over 700 hot water runs, which I will reference later.
In February of 2025 I started experimenting with what I call “Sweet Spot Hot Water Mode.
This uses Noise Reduction mode to give a halfway house between the fast but inefficient Normal mode and the slow but efficient Eco mode.
You can read about that here
As hot water runs complete in about half the time as Eco mode equivalents, this means the indoor temperature isn’t dropping as much as before during hot water, especially coupled with the less drastic overnight setback temperature.
So again, “Comfort over COP”.
Yes, my Sweet Spot hot water runs aren’t quite as efficient as they would have been in Eco, but the house is more comfortable as a result.
The only other hot water change is that I’m now taking the whole tank to 45C on each re-heat.
In winter two, I was taking the whole tank to 48C, but after a successful trial I’m now happy to leave the target temp at just 45C.
You can read about cylinder sizing and mixing water here:
One of the golden rules for efficient hot water production with a heat pump is to schedule reheats at the highest outside temperature.
Here’s a recap of the four golden rules.
The elephant in the room of course is that most off-peak electricity tariffs offer cheaper rates overnight, when it’s usually colder.
Another highlight from the 700 hot water runs investigations was that it’s more than likely always cheaper doing your hot water runs overnight to take advantage of off peak electricity pricing than it is to rely on a better COP from higher afternoon temperatures.
Way more details in the article:
But here is a snippet
Here is a comparison of every hot water run in 2024, summarised and averaged by AM and PM.
Time | Avg Temp | Electric Input | Heat Output | Avg COP |
AM (1am) | 9.1 | 438 | 1588 | 3.63 |
PM (1pm) | 12.9 | 412 | 1590 | 3.86 |
First observation is that the outdoor temperature golden rule holds true; a warmer outside temperature leads to better COP.
But the headline takeaway for me is there isn’t much to choose between the AM and PM outdoor temperatures when viewed across the whole year.
Certainly not that much to gain enough COP to outweigh the financial gain by running overnight on cheaper electricity.
For example, say we need 10 kWh of energy to heat the water (this would be approx 215 litres heated from 10C to 50C)
For the majority of winter 24/25 the price cap for gas and electricity was:
I wrote a whole article on how heat pump COP / SCOP in conjunction with electric unit price are used to determine running costs for a heat pump.
With me having an EV, Batteries and Solar I have been able to reduce my average unit cost paid each month.
I am able to load shift a lot of usage into cheaper (and greener) overnight periods of the Octopus Intelligent Go tariff. EV charging, battery top ups and hot water runs for example. So big ticket items. This drives down average monthly import unit rates.
When it gets colder the heat pump uses more electricity, which is pretty obvious.
And the amount of solar we generate changes each month, with lowest amounts in December and January.
On Octopus Outgoing fixed, each kWh sent back to the grid is worth 15p.
So 15p x 549 kWh = £82.35 in credit.
Note: I have monthly stats going back to 2018 for our solar generation if you are interested
Here are some tables of costs based on the monthly breakdown of figures above.
Electric Used by Heat Pump kWh | Unit Cost (p) | Monthly Cost | |
November | 343 | 9.88 | £33.89 |
December | 356 | 9.82 | £34.96 |
January | 571 | 11.11 | £63.43 |
February | 453 | 10.37 | £46.98 |
March | 331 | 8.75 | £28.96 |
April | 186 | 8.60 | £16.00 |
Total | 2,240 | Total | £224.22 |
Electric Used by Heat Pump kWh | Solar Gen kWh | Electric Remain kWh | Unit Cost (p) | Monthly Cost | |
November | 343 | 121 | 222 | 9.88 | £21.93 |
December | 356 | 83 | 273 | 9.82 | £26.80 |
January | 571 | 89 | 482 | 11.11 | £53.55 |
February | 453 | 161 | 292 | 10.37 | £30.28 |
March | 331 | 450 | 0 | 8.75 | £0.00 |
April | 186 | 602 | 0 | 8.60 | £0.00 |
Total | 2,240 | 1116 | Total | £132.56 |
The removing the solar export payments (£82.35), leaves the total at just £50.21
Electric Used kWh | Unit Cost (p) | Monthly Cost | |
November | 343 | 24.5 | £80.04 |
December | 356 | 24.5 | £87.22 |
January | 571 | 24.5 | £139.90 |
February | 453 | 24.5 | £110.99 |
March | 331 | 24.5 | £81.10 |
April | 186 | 24.5 | £45.57 |
Total | 2,240 | Total | £544.82 |
So the house needed 8,882 kWh of heat across the 6 months for both heating and hot water.
At a COP of 4.0, that means we need 2,220 kWh of electricity to generate that heat (8,882 kWh / COP 4.0 = 2,220 kWh)
To create 8,882 kWh of heat we would need to buy 10,450 kWh of gas.
10,450 kWh gas x 85% efficiency = 8,882 kWh of heat.
10,450 kWh gas x Price cap gas (6.24p) = £652.08
Note: all prices above exclude standing charges on both fuels. Obviously, if you ditched gas (like I have), you wouldn’t have to pay the £100/year gas standing charge.
You can see Octopus Tracker pricing for both Electricity and Gas here:
As well as doing this breakdown for just this winter of 2024/2025, I did a full breakdown of all house running for the whole of 2024 in this article.
Once again, delighted with our heat pump.
Achieving a combined winter SCOP of over 4.0 (400%) in a near hundred old house (heating 4.2, hot water 3.5)
Retrofit, no underfloor, all rads showing that system design is key to making this happen.
If you want to read about what’s involved in proper system design look at this article about the tremendous “heat pumps unlocked” book.
Comfort is off the charts compared to an on/off gas boiler system.
No messing with the thermostat. It’s just a constant lovely internal temperature leaving the Vaillant weather compensation to do its thing.
Our running costs are minimal due to the investment in Solar, Batteries etc (the Holy Trinity and the Renewable Quintessentials).
But as the analysis above shows, running costs would still be cheaper than gas even on conventional electricity tariffs. But running costs could be dropped further by making use of smart tariffs. I.e. doing your hot water runs in cheaper periods.
We would not have been able to dive this deep with Open Energy Monitor installed.
If you can afford an Open Energy Monitor system when you have your heat pump installed, I would highly recommend it.
I see 2025 (and fourth winter of 25/26) being much quieter as my wife has asked for “no more projects”. LOL
Although there will be nothing stopping me looking at non-destructive projects.
I have already started looking at and trialling Havenwise remote system control to see if it can do a better job than the very static Vaillant Weather Compensation / Load Compensation controls.
Will keep you posted how that goes.
If you find the content of this website useful and are thinking of joining Octopus Energy, please consider using my referral code or ‘buy me a coffee’ to help support running costs of the website (hosting fees / cloud servers etc).
Homeowners using this referral code will gift you £50 of free Octopus Energy credit after signup: https://share.octopus.energy/linen-pearl-869
Business users that signup using the same code with be granted £100 of free credit.
If you would like to ‘buy me a coffee’ to show support, please use this link: http://buymeacoffee.com/SVHgIbUYM
Energy Stats can also be found on Twitter / X. Please follow us @energystatsuk for daily tariff pricing graphs and summaries.
We have also started posting the same graphs over on Mastodon. You can find us here: https://mastodonapp.uk/@energystatsuk
And now over on Bluesky too. Find us here: https://bsky.app/profile/energystatsuk.bsky.social
And we are even trying to get the info out via Instagram. Find us here: https://www.instagram.com/energystatsuk/
Note: The current and past performance of energy pricing is not necessarily a guide to the future.
This website uses cookies.