Air Source Heat Pumps: more evidence of problems in cold weather

Air source heat pumps are a risky choice for householders trying to save money and CO2 emissions. This piece looks at the experience of one householder in the south of England who has kept detailed meter readings over the last few weeks. The findings are disturbing. The recent low temperatures (early February 2012) have shown that the costs of running a heat pump can be unacceptably high in cold weather. Anybody considering this new - and apparently eco-friendly technology – should be very wary indeed about their energy bills in deep winter. In fact, they should consider turning off the pump and going back to electric radiators when temperatures drop. The date in this article come from a home of about 90 sq metres (approximately 1000 sq ft), which is about 20% larger than the UK average dwelling. Because the house is detached, with a larger exposed wall area, energy bills are likely to be higher than a terraced house or a semi-detached of the same size.  But the householder has done substantial eco-renovation on the house, including filling the cavity wall and insulating the floors and loft. The windows are double-glazed. His final action was to install a new air source heat pump, put in place by specialists. He knew that a heat pump could only possibly be effective in a well-insulated house but he thought his work would mean that his family would benefit finacially from the new heating system. So far, this hasn't been the case

My rough calculations suggest that this well insulated house probably loses about 200 watts per degree of temperature difference between the inside and the outside. That is, if it’s 10 degrees outside and 20 inside, it will need a heating system that provides 2000 watts, or 2 kilowatts. The key question : is a heat pump a good way of providing this?

The big advantage of this relatively new technology is its potential ability to use relatively small amounts of electricity to create larger amounts of heat. (No – this doesn’t break the laws of thermodynamics, see here).  The effectiveness of using heat pumps to cut our energy bills depends crucially on how much heat you get out for every unit of electricity you put in. Manufacturers will usually quote ratios of three or four. This householder’s experience suggests that the real figure may be as low as 2 or below.

At that level it makes no sense in cash or carbon terms to use a heat pump. Even for homes with cheap rate meters (‘Economy 7’) for night electricity, the average 24 hour price of power is about 8.5p per kilowatt hour at the moment.  Mains gas - which isn’t available around the home whose electricity usage I am reporting here - is about 3.5p per kilowatt hour. In other words, a heat pump which converts one unit of electricity into only two units of heat costs more than 2 units of gas. The carbon dioxide emitted at the average power station to produce a unit of electricity is also over twice as much as the direct emissions from burning gas in a home boiler. If the figures at this home are typical, heat pumps don’t work well in the UK. (This is a strange finding – they really do work well in some other countries such as cold Sweden and nobody seems to be sure why things aren’t the same in the UK).

The failure of many air source heat pumps to save money in Britain must, I suspect, be down to poor expertise among installers. Heat pumps are fiddly to operate and require delicate adjustments. Unfortunately, until this problem is solved, no householder will be prepared to be the guinea pig for a technology that often seems to struggle in (relatively) cold weather. Some sources suggest that the problem arises because the pump ices up - but this doesn't explain why the same problem doesn't occur in colder countries

The numbers

We’ve had a wide range of external temperatures over the last couple of weeks. It started quite warm but the last few nights have been very cold by UK standards, with the thermometer dipping to as low as minus 7 degrees in the local area.  As the chill worsened, the efficiency of the heat pump dropped dramatically.

Dates of measurements Average amount of heating required over 24 hours* Estimated heating need for the house over 24 hours** Typical daily  electricity use*** Implied ratio of electricity input to heat output (‘coefficient of performance’)
Around 23rd January 10 degrees 50 kilowatt hours 25 kilowatt hours 2
Around 28th January 15 degrees 75 kilowatt hours 50 kilowatt hours 1.5
Around 3rd February 20 degrees 100 kilowatt hours 100 kilowatt hours 1


*The difference between the average external and internal temperatures

** The average heat loss from the house’s walls, windows, door, floors and roof per degree of temperature difference multiplied by the average temperature difference.

** The metered use of electricity over a typical 24 hour period

In the early part of this short study period, the electricity consumption figures were poor but not excessively so. The family was getting 10 degrees of heating of his house from the pump for about 25 kilowatt hours a day. This meant the ratio of heat output to power input was about 2, well below the level promised by the manufacturer but still nearly enough to justify using a heat pump. But as the thermometer fell, the bills went up. He was getting about 100 kilowatt hours of heat for each 100 kilowatt hours of electricity he used. This means that in cold weather the unlucky householder is spending eight or nine pounds a day on electricity (multiplied up, £250 a month) but, even more strikingly, he would be better off if he simply installed a few electric heaters in the main rooms. In fact, if I were advising him, I’d say he should turn off the pump whenever outside temperatures fall below about 7 degrees.

The householder has been worried about the performance of his expensive new heat pump since it was put in. He’s had the people who installed it round, as well as the main contractors for the insulation improvements, just  in case they could find out whether the house had major temperature leaks. His concerns seem warranted because his pump is costing far more than it should do. This story  is repeatedly heard across the UK – it’s now time to really find out why many of the heat pumps installed in houses come nowhere near achieving the benefits claimed by manufacturers.