Continuing

We found a heatpump installer who seemed to have some experience with installing into older properties - to this day not sure if he was winging it or really did know what he was doing??  however my own calculations suggested that a single 14kw heatpump - in combination with the AGA and the woodburner should be sufficient so we proceeded on this basis.

I would say initial process was not without stress - being G2 listed we had to apply for listed building consent which took 8 weeks to get - they were ok with it providing the heatpump was not visible from the street , which meant locating it behind a hedge approx 5 meters from the house - with underground pipework to contend with.

Had to hound supplier for an install date - eventually they turned up and completed the install in a week- generally decent guys but I got the impression a bit amaturish and not 100% convinced that they new what they were doing. At this point I was still a bit lacking in knowledge myself to challenge them too much. Pointed out some snagging items at the end of the job and they got a bit Narky tidied up a bit then disappeared , leaving the system running but clearly not optimised.

I set too trying to understand what we had been left with - which was broadly as follows

1. Ecodan 14kw heatpump

2. 5 M length of 28mm plastic primary pipework underground between heat pump and house - not proper stuff but just standard polypipe with standard insulation in land drainage pipe.

3. Wilo Primary pump running flat out

4. a Vaillant Unistore cyclinder

5. Hydraulic separation in the form of a close coupled T

6. Two secondary pumps connected in series both running flat out and connected to existing heating pipework via 22mm copper existng pipework.

7. Most of the radiators were replaced with larger units - but no heat loss calculation was provided - so I dont know if the sizes were an educated guess or calculated - in the end they seem to have been about right, good luck or what??

8. Controls were set up with a combination of weather compensation but with a third party Nest thermostat presumably acting as a limiter?? , The mitsubishi display was mounted in the cylinder cupboard. I quickly binned the Nest and set to run on pure weather comp.

9. System filled with Glycol for freeze protection.

So in general a lot of the standard mistakes that seem to be made. 

I fairly soon installed a basic version of open energy monitor in order to be able to see what is going on - I just monitored primary flow and return temp , Secondary Flow and return temp , outside temp and energy consumption - from this fairly good estimation of COP can be made.

Despite everything not be optimal according to best practice on this forum -  the system worked not too bad actually probably what you would call average - by luck the primary and secondary temperatures were only about 1 degree different so not too much mixing going on. We ran like this for about 2 years - house comfortable I would say, DHW OK -  We had supplementary heating from AGA and Woodburner.

Our Index circuit was a study at the far end of the house from the heatpump and this struggled to be warm - no surprise as 3 radiators at the end of a long run of 15mm pipe - despite the 2 secondary pumps trying to squeeze enough water down the pipes.

Fast forward to last year - I have now retired from work so in the house a lot more - also my big retirement project last year was to build a garden room attached to the study , with underfloor heating - clearly the existing system was not going to cope - So I decided to bite the bullet and correct all or as much as possible of the things that were bugging me about the installation. So I did the following (Note by now the origional installer has gone bust so not too worried about any warranty though probably worthless anyway)

1. Ditched the Glycol and fit antifreeze valves

2. Did a full system design from first principles - using heat engineer software for heat loss calculation , and manual calculation of flow rates and pressure drops - my aim being fully open loop with no hydraulic separation and a single circulation pump.

3. Upgrade primaries to 32mm MLCP from heatpump to first floor landing where the system splits into 4 x 22mm circuits

4. Installed an MID approved heat meter from OEM so I can properly measure the heat produced.

5. Complete new run of 22mm copper through to the study and garden room underfloor heating manifold

6. Moved the Mitsubishi controller from the cylinder cupboard to the dining room (center of house)

7. Installed underfloor heating in the new garden room and study , 3 loops at 150mm spacing (total area 40M2), with no mixing valve on the manifold - I was a bit unsure about the balance between radiators and underfloor heating - but in practice it works well as the screed is quite thick so we can run the underfloor at the same flow temp as the rads.

8. Set the controls up using Auto Adapt mode on the ecodan now that the controller can see a realistic room temp. We are currently on a fixed rate tariff paying 19p/unit and just run low and slow 24/7. Tarif ends soon so thinking about TOU etc since we recently finally got a smart meter that works.

Results:

Very pleased so far - Despite what you may think I believe heatpumps can work extremely well in an old 18th century stone house  - even though on paper it is a thermal disaster area - the heat capacity of the walls is huge - we were on holiday last week and I turned the heating down to 16 Deg from normal 20 Deg C , It took 2 days for the room temp to drop to 16. 

If you live in a house like this I think the important thing is to make it as draughtproof as possible , we have old sash windows which we would not and are not allowed to replace , but we do have perspex secondary glazing which is very effective and unobtrusive.

It is also I think important to follow all of the advice for old buildings such as only using lime mortar and plaster so that the walls can stay as dry as possible, I think a dry stone wall has much better performance that given credit for by EPC etc.

Some Energy metrics in next post.

@travellingwave 

Looking forward to hearing the rest of your story. We just installed a heat pump in our house, so far it has been exceeding expectations and it makes us think about adopting the technology on a wider scale in our family.

My paternal house is subject to restrictions and approvals for anything that affects its appearance, external insulation is not an option.  The oldest deed we have dates from 1912, so it might have been build in the 19th century. It features 60cm solid stonemason walls, and ceilings that are 4.2m high. My parents installed an oil boiler and radiators in the ‘80’s which however has exceeded its designated life span.

So I’d really like to find out how you made it!

@mk4 Thanks for your interest. If the existing oil boiler heating is still functional I think it is worthwhile to do a test as a sense check on any heat loss calculation that is done.

Oil boilers are essentially a fixed output and the only control they have is by switching the burner on and off based on the temperature of the heating water (boiler thermostat controls this). What I did was to wait for a reasonably steady cold spell of weather ideally freezing or below  - I turned on the heating and left for some hours to let everything stabilise at a comfortable temperature  - then I timed the on and off time of the boiler burner - just manually using a stop watch - For about an hour or so. From this you can calculate the percentage of the time the burner is running - if you multiply this by the rating of the boiler it will give a rough idea of how much heat the boiler is producing. Also bear in mind that there may be a room thermostat turning the heat on ond off so ideally this should be turned right up to take it out of the equation - if the house overheats turn down the boiler stat.

I also turned the boiler temp down to about 50 DegC to see how well the existing rads and heating system might cope with lower flow temps - which it seemed to do.

Obviously this is a very approximate test but does give a bit of insight into likely heatloss and gives a bit of confidence moving forward.

Another thing is to look at annual oil consumption - oil giving c 10kwhr of heat per liter - if an old boiler maybe 60% efficient - so each liter maybe gives 6 kwhr of heat into the house. I think lots of people have done this - how you then use the information to challenge is another story?? I guess if it shows a big difference to a heat loss calculation at least you can then ask more questions and dig a bit deeper.

I do think low temperature heating systems can suit older properties very well - the continuous nature of the heating keeps the building fabric warm and helps to stave of condensation - we have never had any hint of condensation or mold - I see these horror  stories about houses which have been over insulated and are infested with mold. Of course it will always cost more to heat a poorly insulated house than a modern eco-home whatever heating system is used.

I don't know if any studies have been done on real world heat-loss v calculated heatloss in older houses - I guess modern builds with modern insulation are relatively easier to get right since the U value of insulation materials is well known, stone walls of varying thickness and construction and dryness  less so.  I will share my own results in next post.

Your ceilings sound very high - our study/garden room has vaulted ceilings into the roof structure  so at the highest point are a similar height , the underfloor heating seems to cope fine with this.

There are many insulation products better suited to older properties - such as wood fiber board, hemp/lime plaster etc - you won't make modern building regs but can still make meaningful improvements. And I would always try to fix any draughts - in my experience this has the biggest impact on how warm you feel.

So our the total heat output from our heat pump since Oct 6 2025 has been 15293 kwhr , with an electricity input of 4179 kwhr giving a scop so far of 3.66

At a cost of £0.19 per unit it will have cost us £794 so far . I don't think this is bad given the size and nature of the property and the fact that we are perpetually warm with the heating running 24 hours 7 days per week.

To be fair we also have some heat from the AGA - not sure what heat this actually produces but it typically consumes 900 liters of oil per annum - turn it on in autumn and switch it off about April time. I suspect that quite a lot of the heat goes up the Chimney but SHMBO likes it plus living in a rural location its good not to be completely relying on electricity for heating. We also have a log burner in the living room but this is only used optionally if we want to be really warm - ie 23 +  iinstead of 20  and again as a backup in case of power cut.

Below is the OEM trend from one of the cold  days in January this year , the average outside temperature was -1 Deg C with a min of -3 Deg C  - So around about the design temperature , and the flow temperature average is 41 Deg C  with a DT of 4.6 DegC and average heat pump output power of 8.3 Kw.

So It would appear that our 14 Kw heat-pump is actually oversized?? - interesting given that one proposal had us needing 28 Kw!!

My own heat loss calculation using heat engineer came out I think at about 16Kw - but this included heating everyroom in the house whereas we dont heat the attic room. So still somewhat over stating the heat loss.

Here is the trend from today - It is much milder and the heat pump is ticking over at near its minimum output - If it gets much milder it will probably start cycling a bit. Along with others I have noticed that the Ecodan is much better at steady running without cycling when it is used in Auto Adapt Mode.

We Do have a couple of high limit stats - one smart TRV in the bedroom as we like it at 18 DegC - and a temp limiter stat in the study where the underfloor heating is - this is the only reason why the flow temperature is not a perfectly flat line as these stats do cause some minor disturbances - I should probably get rid of them and balance it a bit better.

Most of the bits I removed - the two pumps alone accounted for a continuous 0.14Kw of power consumption.

New 32mm MLCP pipe - pre-insulated

The original underground pipework to the heat-pump was only 28mm plastic  laid in land drain pipe which was certainly too small for a 14Kw unit - In fact it did seem to work OK probably as the actual heat demand was much lower than 14Kw - plus the extra pumps to squeeze the water round. In the end I bit the bullet and replaced it with 32mm MLCP - in fact even this is a bit small but I took a chance. To buy the proper underground heat pipe is quite expensive and being as I was already doing building work I decided to make a ducting to lay the pipes in. I used 50mm Thick Polystyrene insulation which is reasonably water resistant - I put a layer of sand in the bottom of the trench to aid drainage - with ceramic tiles  to keep rodents out - All topped over with paving slabs. A lot more work but my labor is free and I had all the stuff laying about  v maybe £600 for the proper stuff. Also has the benefit of being reasonably accessible if any future maintenance or repair needed.

@travellingwave — Thanks for your very detailed account of what you have done. I have a very similar situation, 200-300 year old stone building, solid floors, Listed Grade 2, leakier than a sieve, and a successful heat pump installation. The idea that these building can't be heated with a heat pump is nonsense. What is not nonsense is that they will cost more to heat, but they will always cost more, whatever the heat source. The emitters are heat source agnostic, they don't care where the heat comes from, only that it is sufficient.

I fully endorse just about everything, if not everything, that you say. Some comments from my experience:

(1) if you have oil heating running before you plan to install a heat pump, measuring the heat loss as you did (timing the boiler on/off periods etc) may well give you the most accurate heat loss estimates. Spreadsheet based heat loss calculations are only as good as the assumptions that go into them, and they are often way out, particularly with old buildings. I did come across some research on heat losses and old solid stone walls and if I recall correctly the spreadsheets tend to over-estimate the loss. I will try and find the papers. On the other hand, I didn't find my historical records of heating oil use much use. Knowing I had used 1000L of oil over a heating season didn't tell me much about actual heat loss, because I was running a timed system. You need a steady indoor air temp to do an empirical heat loss, and on timed heating that doesn't happen.

(2) secondary double glazing (I use PET-G with magnetic holding strips) is an absolute game changer. It can absolutely be done (DIY if you want, as I did) on G2 Listed Buildings, and it works wonders, not just by being double glazing itself, but also by draught proofing. It can be fiddly, many of my windows aren't anywhere near square, part of the character if the building, but it is very well worth the effort.

(3) one thing I have found is that I do in practice use more energy, because a heat pump pretty much compels you to run it 24/7 (or something close to that), whereas the oil boiler was on a timer, I think mine was one hard for around twelve hours a day, off at night and during the day when everyone was out working - but then again I am now retired, so an oil boiler if I had one would be on more of the time. The positive flip side of continuous running is, as you have also found, much greater comfort, and, with the aid of secondary glazing, all but no condensation. With the on/off oil boiler, the windows in autumn in particular used to almost run with condensation, now they are dry. This in turn is better for the fabric, less risk of rot in wooden frames etc. But this potential extra cost is something to be aware of.

(4) Listed means needing Listed Building Consent and that in turn triggers a need for planning permission. This can be tiresome in the extreme. At one point the planners decided I needed a bat survey (a small lean to once outside loo which had more recently housed the oil tank needed to be removed to make space for the heat pump outdoor unit). As those who have enjoyed the pleasures of bat surveys will know, the whole process is designed to grind you down until you give up. As I had good reason to believe no bats would be harmed during the works (there were no bats nor any evidence of bats in the lean to), I asked why I needed a bat survey. The answer was they had a secret unpublished bat map that showed I had bats. I managed to get them to send me a copy of the bat map, and it was immediately obvious it had been drawn using whatiffery (modelling). The give away was the north end of the ~6ft long lean to had bats, whereas the south end did not, a most unbelievable proposition. An orchard had bats in one half, not in the other half. When I pointed this out, and added there was no physical evidence of bats past or present in the lean to, and furthermore I was replacing a fossil fuel boiler with a heat pump, they finally agreed to do without a bat survey. The point of this anecdote is to show you may need considerable reserves of energy and an ability to do your homework and challenge unreasonable assumptions during the process to get planning permission/Listed Building consent.

PS I see you have added some more posts while I was typing this, will read them shortly, and add any comments I may have. Meanwhile keep the posts coming! 

Posted by: @travellingwave

↑

So It would appear that our 14 Kw heat-pump is actually oversized?? - interesting given that one proposal had us needing 28 Kw!!

28kW - that is what spreadsheet based heat loss assessments do. I had similar but not quite as large variations in assessments. 

I expect I am telling you what you already know, or can work out, you can do a very accurate heat loss assessment with a heat pump. Plot the hourly OAT against the hourly energy use (this will be kWh but no matter) for as many OATs as possible (weeks at least) while the IAT is stable and the point where the regression line crosses your low end design temp is the design heat loss. In your case you will also obviously need to turn off your auxiliary sources of heat (or possible leave them on, if they are always on, but then your result won't be a total heat loss, just the heat loss that needs to be made good by the heat pump).

@cathoderay Thanks for your detailed reply - and also for your input over the years - as I say this resource is invaluable.

Fully echo everything you say - I too have recently retired and not having to worry about the heating being on or not is a real boon especially when the weather has been so miserable lately.

I think if you buy a listed building you have to accept the restrictions that come with it - though I agree that some of the things that get triggered seem a bit daft at times - I have generally found our local Conservation Officer to be quite pragmatic especially if you build a relationship and some trust that you are doing the right thing by the building. - though this is getting harder I know with many councils only engaging when money changes hands for planning apps etc. 

We too have the secondary glazing with magnetic strips and it works really well for us too - have had to add some extra clips as it can be a bit top heavy and come adrift sometimes.