Are Air Source Heat Pumps (ASHPs) a Silver Bullet?

According to the Committee on Climate Change heating our homes makes up 40% of our energy use and 20% of our carbon footprint. While there have been dramatic improvements in building standards since 1970 there remains a legacy of poorly insulated homes.

Retrofitting our often old housing stock to reduce heat loss is crucial, but we also need to stop using natural gas as the source of heating if we are to have any chance of meeting our goal of halting global heating. 

It got me thinking about this question – if someone asked about retrofitting their house, and was motivated by the desire to reduce the carbon footprint of heating their home: 

what is the first thing they should do?

It may seem a somewhat artificial question, because in any real world situation, several measures are likely to be advisable, but bear with me.

Many retrofit professionals repeat the mantra “fabric first”, which means, focusing on insulating the building, dealing with leaks, and so forth. This sounds like good advice, given that the cost of some measures, like insulating a loft, are relatively cheap and deliver big savings in carbon emissions.

However, in many cases this is expressed in stronger terms, like “deep retrofit”, which can mean doing everything possible to reduce the heat loss of a building. This could include external wall insulation to homes with solid walls (which cannot benefit from cavity wall insulation), new windows, and dealing with associated issues related to moisture, for example. This school of thought suggests that we should only consider using a heat pump after deep retrofit is complete [Note 1]. 

The mantra “fabric first” then effectively turns into fabric only, because it is not difficult to exhaust a householder’s retrofit budget with changes to the fabric of a building. 

So why should we be considering heat pumps alongside changes to the fabric of a building? 

A heat pump harvests the ambient energy outside a house – either from the air, the ground or water. This ambient energy comes from the sun (when the ground is used as a source it is never deep enough to harvest energy from the core of the Earth, even with a bore hole, and is simply extracting energy from the ground that has been warmed by the sun and stored there). 

For every unit of electrical energy put in to drive the heat pump, it is able to deliver at least 3 units of heat energy into the home. A nice simple explanation of this process is provided here.

Professor David Mackay wrote in his seminal 2010 book Sustainable Energy without the hot air (p.151):

“Let me spell this out. Heat pumps are superior in efficiency to condensing boilers, even if the heat pumps are powered by electricity from a power station burning natural gas. … It’s not necessary to dig big holes in the garden and install underfloor heating to get the benefits of heat pumps”

He was calling for the adoption of Air Source Heat Pumps (ASHPs). He didn’t use the words ‘silver bullet’ but it is clear he was a big fan and frustrated at the low level of take-up. As he wrote 

“heat pumps are already widely used in continental Europe, but strangely rare in Britain”.  

I thought about how to present some information to help explore the question I have posed, and compare fabric related measures to an ASHP. I took data from the Energy Saving Trust website for a typical semi-detached house and plotted the capital cost of different interventions against the annual carbon saving that would result.

The capital costs are indicative and include the parts and labour required.

The only change I made to the Energy Saving Trust data was I reduced the savings for an Air Source Heat Pump (ASHP) from about 4.5 tonnes of CO2 to about 3, because that better reflects the carbon intensity of the national grid in the UK in 2020.

I also indicate the level of disruption involved with colour coding, because this can be a factor in a householder’s decision making. The graph is based on the data tabulated in the Table at the end of this essay [Note 4] (click on image to see higher resolution):

This householder is aiming for the highest carbon saving, and there is only one answer glaring out at you from this graph: the ASHP.

If I change the question slightly we get a more nuanced answer:

what is the first thing they should do, based on carbon reduction ‘bangs for your buck’?

The ratio of the annual carbon saving to the capital cost is a measure of ‘bangs for your buck’.  On this criterion, it makes sense to do the low hanging fruit of loft insulation and fixing drafts, but then once again the ASHP scores very well (this allows for the fact that typically more than half the radiators will need to be upgraded [Note 2], and are included in the cost estimate).

Whereas external wall insulation would typically be similar in cost to an ASHP, but deliver only a third of the annual carbon saving and be a much more disruptive intervention; and in many cases not one that is practical to implement.

A counter argument would be that if we did manage (for a solid walled home) to do all the fabric related measures, we would achieve about 2,400 kgCO2/yr carbon saving (over half the current emissions of 4,540 kgCO2/yr) for an outlay of about £18,500; and then the ASHP could be added and would then ‘only’ need to deal with the remaining half, and that could mean that a lower capacity heat pump could be installed, reducing its cost somewhat.

Each situation will be different and depend on what interventions are possible. If a building is listed and external installation is prohibited (and the alternative of internal insulation dismissed), then the fabric related measures would then total 1,500 kgCO2/yr, leaving two thirds of emissions to be dealt with. 

In either case, in order to maximise the emissions reduction one would require a heat pump.

My argument is not that you must fit a heat pump first, but that you should consider all the available options and to think about a plan (possibly over a number of years). 

I went on to plot another graph where I included the following additional features:

  • To show the ‘best’ reasonable case fabric interventions (which would raise the cost of say, wall insulation, but at the same time increase the carbon saving). 
  • To include a Ground Source Heat Pump (GSHP) option, either with horizontally laid slinkies, or using vertical bore hole(s).
  • To show what happens as the UK electricity grid moves from 2020 carbon intensity levels to being 100% green.

The following illustrative graph was the result:

There are a number of interesting observations based on this graphic.

  • Firstly, as the arrows show, we can increase the carbon savings for each fabric related measure, but these improvements come at extra cost. There will be some trade off for each householder situation as to how far they can go.
  • Secondly, once you have a heat pump, the annual carbon saving will increase every year as the grid gets greener (as it has been in the UK), without the householder having to lift a finger.
  • Thirdly, while a GSHP may have a better performance than an ASHP, it is likely to be quite limited [Note 3], and as Paul Kenny said in his talk to Carbon Coop ‘Heat Pumps – Learning and experiences from Ireland’, if you have extra money spare, why not do the easy thing and spend it on further upgrades to radiators, when you can achieve a target COP without the major disruption and risks associated with a GSHP project (assuming that is even an option).
  • Fourthly, a borehole GSHP is even more costly, and more risky. There are significant risks associated with, for example, drilling into water tables, but the real killer is the cost. For a single householder it makes little sense. Of course, one can imagine scenarios where several houses could share the costs, but these are likely to be exceptional projects; not the basis for mass roll-out of heat pumps.

Some will argue that an ASHP requires supplementary heating during very cold spells in winter. However, in the same talk referred to above, Paul Kenny used data from a significant number of retrofits in Ireland that had ASHPs, using a design parameter of -3°C for cold winters. When the beast from the east came and these houses experienced -6°C, they were all fine and did not require supplementary heating. He wrote a piece on LinkedIn about this experience, which flies in the face of much of the ‘received wisdom’ in the retrofit community.

And in the UK, without the much larger grant that GSHPs enjoy, as compared to ASHPs, it is doubtful there would be anything other than a marginal role for GSHPs. It will be no surprise if ASHPs dominate the heat pump market in coming years and for some installers, this is already the case.

So, are Air Source Heat Pumps a silver bullet to decarbonising the heating of homes? 

One has to say in many ways they are!

But of course, in reality, it makes sense to consider them in the mix of other retrofit measures, and to carry out some improvements to the fabric of a building as part of a ‘whole house’ plan.

We just shouldn’t let the ‘deep retrofit’ mantra put people off considering an ASHP; maybe even as one of the first things you do.

(c) Richard W. Erskine, 2021


1. While I am a huge fan of PassivHaus and similar standards, we must remember that these standards cannot easily be applied to existing stock, and would be hugely expensive. 80% of the homes in 2050 already exist, so 80% of the problem of decarbonising heat in homes is already there; and BRE estimates there are 9 million ‘hard to treat’ homes in the UK.

2. Upgrading radiators is usually required to increase the effective surface area. This is needed because heat pumps operate at a lower flow temperature, and the heat delivered is a function of the temperature of the radiator and its surface area. The surface area can be increased by using 2 or more panels with fins sandwiched between them. This can also help reduce height and width of the radiator that would otherwise be necessary, while making the radiator somewhat deeper / fatter.

3. A GSHP has a better Coefficient of Performance (COP) in winter, an ASHP could do better in Spring and Autumn. The overall Seasonal COP for a GSHP will probably be higher but unlikely to be higher than 15%. We need real world studies to get a good figure here. But the cost of a GSHP using slinkies in 1.2m trenches (for those unusual cases where householders have sufficient land to achieve the area necessary) is something like double that for the ASHP. 

4. Table of Typical three bedroom, solid walled, semi-detached house


Filed under Uncategorized

10 responses to “Are Air Source Heat Pumps (ASHPs) a Silver Bullet?

  1. I am a proud pioneer in The Netherlands with a heatpump and roofsolar in a 1990 house. I followed the route of small investements first. Wall, floor, roof insulation. Roofsolar in two steps. This took 5 years. Combined with energysavings the monthly costs for energy halved. Three years ago we prioritised the investement of a heatpump. Last year we linked a boiler to the heatpump. The gasline has been removed. Now the monthly out-of-pocket energy cost are 20%. The house is on year basis energy neutral.
    In this way we escaped the lock-in position of a ” energy wrong” house.
    The tax system gave support for this route.
    A key factor was the search for a progressive company.
    One last remark is that you can not be a pioneer without a certain risk.


    • Thank you. I guess, to move to massive roll-out, we need heating engineers to unlearn gas and oil and ‘learn’ heat pumps (especially ASHPs). We need it to be less ‘pioneer’ and more ‘new normal’. I think the HPs and standards (such as MCS) mean there is no excuse for this not to be the future.


  2. I have already trodden this path so I may be biased; but it is always feels good to have your biases confirmed!
    We bought a Victorian detached house (early 1850s) in Hereford city nearly 20 years ago. It came with a 1970s/80s single storey extension with cavity walls. There was no EPC rating then but I suspect it would have been E, possibly F.
    Over the first 5 years, we made a concerted effort to improve its energy performance by insulating roof spaces, installing double glazing, fitting draught-free external doors and having the cavity walls filled while also installing PV and solar thermal systems. The 2 gas boilers (one for the old Victorian part and one for the newer 70s/80s extension) were replaced with more efficient ones.
    About 5 or so years ago, we decided to get a local trusted builder to install internal insulation on all the external walls of the Victorian house. Since the rooms needed re-plastering anyway, the cost of having the extra wall insulation was not that great. The large cellar was boarded (floors, walls and ceilings to reduce draughts and insulate the ground floors) and makes for a much dryer storage area.
    The costs of doing the ‘fabric’ bit was achievable by spreading the costs over a number of years and using a local knowledgeable builder while taking advantage of any Government incentives. The house now has an EPC rating of B.
    We had 2 x ASHPs (8.3 and 14 kW) installed last January (2020). We did very little radiator upgrading: we added a radiator to a room that previously did not have one, replaced one leaky radiator and up-rated one other out of a total of 18 radiators. The replacement hot water cylinders were connected up to our existing solar thermal.
    An important point to note was that the Government grant for installation of ASHPs is/was dependent on having a minimum of insulation; possibly EPC of C or greater? Therefore, some costs additional to ASHP installation may be needed if basic insulation measures have not been carried out.
    ASHPs make even more sense if you have PV installed.
    The way you operate the ASHP heating system is different to the way you operate a gas boiler system and may mean you do not need to upgrade the radiators. Your installer should be able to advise. Our installer suggested we stick with our old radiator system and evaluate first because they can always be replaced at a later date. So far we feel no need to do this.
    We still have a wood burning stove in our main living room which is used intermittently – it is basically there in case we have a power cut in the middle of winter!
    Over a year, we used 13500 kWh of electricity (green tariff) for a detached house of 15 rooms. I don’t know how much of this was down to the ASHPs.


    • Phew, quite a journey you have been on. Did the installer of the heat pumps do a room by room heat loss analysis? I am rather surprised that none of the radiators were upgraded (unless they were oversized for gas boiler flow temps. in the first place!). You might like this talk that Chris Wilde gave to Nailsworth Climate Action Network (which I am Sec. of), that you can access here >


      • Richard,

        Just got round to watching Chris’s talk. I have heard him talk before and it was as good as I expected. I think I concur with most, if not all, his points. I stand corrected on the ‘EPC requirements’ needed to claim the RHI.

        We had an independent EPC assessment followed by full room heat loss analysis by the company we used ( Apart from one room which was underpowered in terms of radiator surface area (we had lost a radiator when we insulated the inside of an outer wall so replacing the remaining single with a double made sense) the other two ‘additional’ radiators replaced an old leaking one and a new one in a room that had no radiator. Based on their experience, they said many customers found the rooms too warm using the standard heat loss/requirement calculations so they will sometimes recommend testing out with the current radiator configuration (after correcting for any obvious deficiences) and upgrade only if necessary. This will save money on the installation costs and extend the life of existing radiators. The only ‘underpowered’ room (9 m x 4 m) includes the woodburner & kitchen and is heated by a slightly overpowered 14kW ASHP (the 11.2kW was borderline for this part of the house).

        I would re-emphasise that ASHP heating schedules are different from gas boilers. Our ASHP controls are set for 19 degC during the day and 16 degC during the night. Keep it simple. The house is much warmer than it used to be and there is constant hot water.
        All the best,

        PS I’m now going to have a longer look at the NailsworthCAN blog!
        PPS I came across your website via HotWhopper.


  3. Kevin Boyce

    When we built a new house (mid-Atlantic area of the US) 3 years ago, I went with mini-split air source heat pumps. We need cooling in the summer, so it makes more sense than having heat distributed by water. We have two outdoor units, each with 4 indoor units. I believe our design temperature is -12 °C, and the first winter we got down to -18 one morning. If you reduce the setpoint at night, it can take a couple hours to get back to daytime temperatures inside if it’s that cold outside. The defrost cycle always seems to happen right when I turn up the temperature. But the radiators in my old house took just as long, so it’s not been a problem. And in every other way, they’re fantastic.

    And the units we have claim a COP of >2.5 down to, I think, -25 °C. I don’t know whether that includes defrost cycles, but the lesson to me is that ASHPs are viable in all but the very coldest locations.

    I also enjoy how quiet these are, both inside and outside. The variable speed compressor and fan outdoor makes it difficult to tell whether it’s running. This may only apply to mini-splits; if you’re heating water for distribution, cycling full on to off may carry the efficiency hit that it does if you’re putting the heat directly into living space. That efficiency boost is (I think) the main reason for a variable speed compressor, but the reduced noise is a significant side benefit.


  4. CloverRayner

    As the arrows show, we can increase the carbon savings for each fabric related measure, but these improvements come at extra cost. There will be some trade off for each householder situation as to how far they can go.


  5. Dr Ian Roberts

    Chris Goodall (Carbon Commentary) has written a post on hydrogen vs heat pumps for home heating. I’ve not read it yet but would be interested in your thoughts.


  6. Pingback: When does it make sense to switch to a heat pump? | EssaysConcerning

  7. Pingback: Insulate Britain! Yes, but by how much? | EssaysConcerning

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s