Monday 29 November 2021

Real-World Heat Pump Running Costs

Solar PV is a Necessary Enabler of the Transition to Electric Heating 

Much of the discussion of the transition to electric heating has focussed on the installation costs of heat pumps, but what do the running costs look like?  It’s all very well handing out £5,000 grants to make heat pump installations more affordable for consumers, but if people take this government incentive only to discover that the cost of energy bills become cripplingly expensive, the resulting negative coverage could stop the transition to clean heating before it gets going.  Conversely, if energy bills fall for houses with heat pumps, it will make it much easier to convince people to ditch their gas boilers.  To get a sense of costs, we need to know two things - how much does electricity cost compared to gas and what efficiency can we expect from heat pumps and gas boilers.

How Much do Heat Pumps Cost to Run?

Advocates of heat pumps regularly claim that a ‘well designed, well installed and properly run heat pump will cost no more to run than a gas boiler’.  A careful re-reading of this sentence will show you that three things have to go right for heat pumps to cost no more than gas heating. 

One thing we know for sure is that in the UK electricity costs much, much more per unit than mains gas.  Nottingham Energy Partnership has the average standard rate for electricity in September 2021 at 23.3p/kWh (kilowatt-hour), and mains gas at 4.39p/kWh.

The efficiency of a modern condensing gas boiler is often said to be around 90%, but since we are interested in real-world heat pump performance, we should compare like for like.  A field trial of the seasonal efficiency of 60 boilers by the Energy Saving Trust in 2009 gave a value of 82.5% for combi boilers.  

Using this efficiency one unit of gas heating costs 4.39/0.825 = 5.32p/kWh.

For heat pump heating bills to cost no more than a gas boiler, the efficiency of the heat pump would need to be higher than 100% x 23.3/5.32 = 438%, but what efficiency do heat pumps achieve in practice?

Real World Heat Pump Performance

The Energy Savings Trust and the Department of Energy and Climate Change (now called BEIS), set out to answer this question in 2008.  The first large-scale heat pump field trial in the UK aimed to determine how heat pumps perform in real-life conditions. The year-long field trial monitored technical performance and customer behaviour observed at 83 domestic properties across the UK.

The resulting report (Getting Warmer: a field trial of heat pumps), published in 2010, found that the average efficiency for an Air Source Heat Pump (ASHP) was 220% (page 16), although this was revised down to 182% by a subsequent analysis published in 2012.  This second report corrected errors and removed data provided  by ‘Manufacturer A’ which were felt to be from systems that had been hand-picked, carefully optimised and installed in the homes of the manufacturer's own staff.  (See Detailed analysis from the first phase of the Energy Saving Trust’s heat pump field trial, pages 19-25)

Note: I have focussed only on Air Source Heat Pumps because most people expect that this is the technology that will be deployed in the greatest number.  They are lower cost and more convenient to install than more efficient Ground Source Heat Pumps which require a deep bore hole to be drilled or trenches to be dug.

Image: System Efficiencies of Air Source Heat Pumps reported in “Detailed analysis from the first phase of the Energy Saving Trust’s heat pump field trial”

Despite the best efforts of the authors to put a gloss on things (“the best performing systems show that well-designed and installed heat pumps can operate well in the UK”), the results were highly disappointing.  

Real-World Heat Pump Performance - Try Again

The UK heat pump industry responded positively to the issues identified in the trial and significant changes were made to the regulatory scheme for UK heat pump installers. The Microgeneration Certification Scheme (MCS) rewrote its MIS3005 installation standard for heat pumps to better control the quality of system design, installation practices and householder training that had been shown to affect heat pump performance.

Consequently, a second phase of the study was initiated.  38 of the heat pumps in the first trial were selected for interventions to improve their performance. Interventions ranged from major (swapping an over or under-sized heat pump), medium (changing radiators, adding a buffer tank, replacing circulating pumps with variable speed DC pumps) or minor (changes to controls, refilling the ground loop, adding insulation). Householders also received improved guidance on how to operate the heat pumps properly.  Six new heat pump systems installed to the new MCS standard were added to the sample and all were monitored from April 2011 to March 2012.

The results for the second attempt were published in a summary and detailed form:

As a result of all these interventions, the average efficiency of ASHPs in the new study rose to 245% 

Note: this performance improvement Phase 2 and Phase 1 included a change of the definition of efficiency – on a like for like basis the increase was from 183% to 211%.  However the preferred efficiency measure in Phase 2 (SPF H4) is in my opinion a better comparator with boiler efficiency than the System Efficiency measure used in Phase 1.  System Efficiency includes losses between hot water tank and taps/showers, whereas the SPFH4 boundary stops at the hot water tank.   


Real World Heat Pump Performance - Third Time Lucky? 

Around 14,000 Heat Pumps were installed with funding from the RHPP, and 700 of these (around 5% of the total) were subject to a detailed monitoring study.  The study reports an average efficiency based on SPFH4 for the ASHP in the sample of 241%. 

However it also reveals that the heat meters used in the study were calibrated for water and not the antifreeze-mix with which most would be installed .  The estimated 4-7% over-statement of performance was not corrected in the published result.  Applying a mid-range 5% correction, would make the true average SPFH4 nearer 229%.

Reassuringly, this is still closer to the second EST study than the first and suggests that the changes made to the industry standards in response to the disappointing performance of systems in the first study had fed through into a higher general performance, across a reassuringly large sample of installations.

Taking efficiency from this most recent study of 229%, the annual energy costs for a house heated by a heat pump will be (23.3/5.32) x (100/229) = 1.91 times higher than the same house heated by a gas boiler.

So, even after industry steps to eliminate design errors, carefully optimising the installation and coaching the householder how to use the heat pumps, running costs are still double those of a gas heated property. 

What hope do we have when we scale up to install heat pumps in the huge numbers envisaged by UK policy makers?  If installations increase from 30,000 a year currently to the 300,000 a year called for by the government will the heat pumps perform as well as those in the second study, or is it more realistic to anticipate performance closer to the first study?

Adjusting the Price of Gas & Electricity

One approach to make heat pumps more appealing is to make gas more expensive and electricity cheaper.  Government indicated in its recently published Heat in Buildings Strategy that it would consider this approach:

we will look at options to shift or rebalance energy levies (such as the Renewables Obligation and Feed-in-Tariffs) and obligations (such as the Energy Company Obligation) away from electricity to gas over this decade” Heat in Buildings Strategy p16.

What impact might this have?  According to OFGEM  Environmental and Social Obligation Costs at 25% of the price of electricity, whereas it’s only 2.5% of the price of gas.

Infographic Bills, prices and profits, 27 Oct 2021, Source OFGEM

The cost of a unit of electricity might come down to 75% x 23.3p = 17.5p/kWh

By how much would gas need to increase to replace the lost revenue?  Again, according to OFGEM typical dual fuel domestic consumption values as of 1st April 2020 were: 12,000kWh for gas and 2,900kWh for electricity.  (Source - see footnote)

For an annual use of 2,900kWh for electricity, the social tariffs come to 25% x 2,900 x £0.233 = £169

For a gas use of 12,000kWh to replace this social levy, the price of gas would have to rise by £169/12,000 = 1.4p per kWh, taking the price of a unit of gas heating after boiler efficiency up to 6.72p/kWh

If this were to happen, we can adjust our calculation for the difference in running costs 

Under this scenario, an ASHP might have running costs (17.5/6.72) x (100/229) = 1.14 times higher than gas heating, but only in 10 years’ time as government makes clear that any transition would have to be gradual to avoid pushing people into fuel poverty.

Heat Pumps and Solar are a Perfect Combination

Even when heat pumps are ‘well installed and properly operated’, even by taking 25% off the cost of electricity and shifting it over to gas, it seems likely that consumers are going to be paying more for the shift to electric heating long after the bill for the installation cost has been settled.

For an average dual fuel bill with 12,000kWh of gas use at 4.39p/kWh, the heating cost is £527/year.  Taking the ASHP efficiency from the most recent study, with ASHP heating bills 1.91 times higher than gas, the extra cost to the householder is £479/year.

One way to make the transition to zero carbon heating cost neutral on running costs is to insulate the property and reduce its heat demand.  If heat demand could be halved, running costs would end up at the same level.  However, this might be a tall order for households that have already taken the basic steps of loft and cavity insulation and double glazing, and also taking into account that the hot water demand cannot be insulated away.

If a 3kWp solar system is installed with the heat pump, generating say 2,550 kWh a year of electricity, and if 50% of that generated electricity is used on site to offset electricity use at 23.3p/kWh (£297) and 50% is exported to the grid under the Smart Export Guarantee at 5p/kWh (£64) then we’ve saved the resident £361 a year from their energy bill.  If we combine this with battery energy storage and push the self-consumption of solar electricity up to 80%, then the corresponding saving becomes £500 a year.

The solar doesn’t need to be generating at the same time the heat pump is operating for the savings to be there – remember that any electricity use in the property can be offset (for example appliances, heating hot water and even charging electric vehicles), and every unit not bought from the grid is a saving on that household’s electricity bill.

Social Landlords, housebuilders and policy makers facing the challenge of how we are going to get our homes to zero carbon while bringing tenants, homebuyers and voters along for the ride need to start thinking of solar PV and other smart energy technologies as enabling technologies for zero carbon heating.  Otherwise the real-world running costs for heat pumps could prove to be an inconvenient barrier to mainstream adoption of electric heating.


2.12.21 - it was pointed out to me that the original version of this blog used gas boiler efficiencies of 90% (which are representative of laboratory test) and unfairly compared these with actual performance in the field for heat pumps.  The blog was updated to use field test results of combi-boilers from Final Report: In-situ monitoring of efficiencies of condensing boilers and use of secondary heating, 2009 The Energy Saving Trust, with annual efficiency of boilers re-set to 82.5% instead.