I’m told that I need to run my radiators at a lower temperature if I have a heat pump fitted, but how will I still be warm?
This could seem a confusing topic.
Let’s think firstly of how radiators are normally controlled. A boiler system is traditionally designed so that radiators turn themselves on or off to achieve the desired room temperature. E.g. in mid-winter they may come on at 6am, and maybe 1 or 1.5 hours later, the house is warm. To be able to achieve this, the radiator may need to be very hot. Hot enough to not only ‘beat’ the heat loss from the building, but to heat it up quickly too. Once the house is warm, the radiator goes on and off as dictated by a room thermostat or the flow is reduced with trv radiator valve.
Heat pumps will be far more energy-efficient if they work at lower water temperatures, and there is a lot that can be done to achieve the same room temperature with lower radiator temperatures.
a) Instead of turning on and off, the radiator can be kept at a constant lower temperature that matches the heat demand.
b) To reduce the warm-up load, the room can be never allowed to get too cold. E.g. a night temperature of say 18°C. The heating can also be turned on say an extra hour (or more) earlier in the morning.
It may seem wasteful to have the heating enabled when we are asleep or out. However, the heat pump’s efficiency gain can be significant if we can drop the water temperature flowing through it, so even if the total heat delivered to the building is greater, the heat pumps power input can be less.
In well- insulated buildings that are occupied most of the time, it can be found most effective to keep the system constantly enabled 24/7.
1. I have heard stories that some heat pumps are expensive to run. Can this be true?
I will answer this simply using an analogy with a car.
Heat pumps are different from any other heating method – they transfer heat from outside. The energy-efficiency (heat output compared to electrical cost to run ) can vary greatly, and is mostly affected by the temperature of the water being circulated. For example –
Producing lukewarm water is easy for a heat pump. (e.g. 4kW of heat for every 1kW input).
Producing very hot water is harder work for a heat pump. (e.g. Only 2.5kW of heat for every 1kW input).
The same applies to a car and fuel efficiency
Driving on the flat, good fuel economy 50 MPG (5.5 litres/100 km)
Driving uphill, poor fuel economy 30 MPG (9.4 litres/100km)
Back to the heat pump. If your system is set up well so that the temperature of water circulating through the heat pump is as low as possible for your particular system, but still giving adequate heat output to the room, then you should be getting good ‘fuel economy’
However, some heat pumps are in practice making the water far hotter than needed. This is akin to driving a car up an incline all the time, with resulting high running costs.
Here are some examples of what could go wrong.
- A heat pump system configured to operate for only a few hours a day. In order to get adequate heat to the building, the radiators need to be very hot.
- A system that has a buffer cylinder and mixing valve that is badly set-up. Here the water that the heat pump produces could be far hotter than the temperature that the underfloor heating or the radiators actually need.
For both these examples, the analogous car is being driven uphill at high throttle, making it uneconomic.
So, to answer the question, yes, the running cost could vary considerably. But by ensuring that the heated water temperature is not too high, your heat pump is likely to be operating efficiently so it should not be expensive to run.
Another reason for high running costs could be an oversized heat pump heating say only half the house on milder days. Not all heat pump systems (including inverter types) deal with ‘light load’ conditions well.
2. Which is best, air-source or ground-source?
The choice depends mainly on the application, so the answer to this question requires some thought – Before we delve into that, ……one initial thought – in small well-insulated houses, it might be hard to justify the extra cost and upheaval of ground source. The difference in running cost between air or ground would be small. In general, larger houses are more likely to lend themselves to ground source.
Advantages of Ground Source (GSHP)
The main advantage with Ground Source is that the ground below about 5 ft. (1.5m) deep maintains its heat during mid-winter. This is a particular advantage in climates with dramatic changes (swings) in air temperature. GSHPs are great during severe winters. It is also more of an advantage at the beginning/middle of winter when a lot of the summer’s heat is still in the ground.
Disadvantages of Ground Source
Digging horizontal trenches in your garden can be very disruptive (though the growth does recover). Boreholes may be less disruptive, but are likely to be more expensive. The whole GSHP installation is likely to be costly and take time. You also need a suitable position for the heat pump inside the house that is not adjacent to a bedroom or a quiet room. The ground collector also needs to be large and deep, else the energy-efficiency will tend to reduce, particularly towards the end of a long winter.
Advantage of Air Source
Air source is far easier and cheaper to install, and the unit would be sited outside the house. This might be welcome for smaller houses where there is no room for equipment inside. During autumn and spring, a good ASHP is likely to be equally as energy-efficient (same running cost) as Ground Source. In fact, an ASHP can be more efficient than a GSHP in spring.
Disadvantages of Air Source
At the coldest times of the year, the efficiency of the device is at its minimum. This is just the time when maximum output is required. So during a severe winter, some alternative supplementary heating may be needed. That said, the best and newest models can operate well in very cold conditions.
ASHPs have a defrost mechanism that clears frost build-up on the heat-exchanger when the air is below around 6°C (43°F). However, units have advanced considerably over the last decade or so, so this issue is minimised. The defrost mechanism typically reduced the output by only about 10% (at worst times).
The unit will need outside space which is not always available. The operating noise can be relatively low, but this could be unwelcome for some. Being outside, the unit will be affected by the weather. It is therefore not likely to last as long as it’s kept-in-the-warm GSHP counterpart.
Type of building
Interestingly, older solid-stone (high thermal mass) buildings could utilise a relatively small ASHP quite well by partly utilising the thermal storage of the building’s fabric. Due to the slow response of these ‘heavy’ buildings, the heat pump could be set to run a little more during the warm day than the cold night. The average efficiency of the heat pump is improved whilst the outside air is warmer. The average inside temperature may remain stable due to the heat storage in the stone. That said, it can be extremely difficult to configure this with currently available controls, and it could be wasteful to have continuous full-heating unless it’s needed, i.e. if the building is occupied all day.
An ASHP would match well with some sort of supplementary heating device to help cope with times of very cold weather. Many heat pumps can be used alongside boilers (hybrid / bivalent). The ASHP can provide all heating on the average winter’s day, but can ‘call upon’ the boiler for assistance during very cold times.
The most environmentally-sound option for out-of-town places could (arguably) be a combination of ASHP and wood fuel. If the wood stove is used during the coldest times, the ASHP could automatically reduce itself – hence the heat pump’s load is reduced at the times when it would be least energy-efficient. That said, it is current thinking that wood-burning is actually fairly polluting and not good for health.
See slightly old article on Air Source – It’s worth reading https://aecb.net/download/air-source-heat-pumps-friend-or-foe-2/
3. How do I ensure that my system is energy efficient?
It is very important to understand that the hotter the water produced by a heat pump, the poorer the energy-efficiency. So, running at a lower temperature can save a lot of energy. The following ratings, for a typical ground source heat pump system, illustrate this point:
(note – the larger the COP, the better)
Water heated to 55°C (95°F), COP = 2.4 ( e.g. a 6kW heat for 2.5kW input)
Water heated to 45°C (113°F), COP = 3.2 ( e.g. a 6kW heat for 1.9kW input)
Water heated to 35°C (131°F), COP = 4 ( e.g. a 6kW heat for 1.5kW input)
(COP is the energy efficiency ratio. See Glossary for better description).
By keeping the heat-output temperature low, the energy-efficiency will be high. This is generally achieved by having large radiators or good underfloor heating.
That said, I see many heat pumps that are badly set-up so that the circulating water is hotter than needed. This causes the system to be less efficiently than it could. The operating water temperature can sometimes be dropped by attending to various things – For example – by operating for longer periods at lower temperatures, by ensuring that the water is circulating adequately, e.g. floor loops should not be throttled too much, and by ensuring that any mixing valves are operating correctly.
4. Why is underfloor heating coupled with a heat pump so highly rated? Can I use radiators?
The lower the temperature of the heated water, the better the heat pump’s efficiency, and lower the running costs.
Radiators traditionally used hot water (60 deg.C, 140°F) or higher, but with better-insulated building and larger radiators, this can be dropped to around 45°C, (113°F), or considerably lower for much of the year.
In simple terms, the bigger the heat-emitting surface area, the ‘cooler’ the water needs to be. Indeed, a good underfloor heating system may require water that actually feels cool to touch. However, such low-temperatures are only achieved with a good underfloor design. In simple terms the more pipe the better (it’s not rocket science), so long as the pressure drops and circulation pump power is kept within limits.
For energy-efficiency, a concrete screed systems with tiled surfaces promise to be the best – better than a timber floor that act partly as an insulator. Carpets can be used, but reduce the system efficiency when a heat pump is used. However, the pipe design can be altered to part-compensate for the timber or carpet covering.
One thing to be mindful of is the underfloor surface temperatures. For old poorly-insulated buildings, the floor might need to be uncomfortably warm on the feet. By contrast, in a highly insulated new-build, the heated floor could actually feel cold. For this reason, some choose a timber floor in these super-insulated houses simply to keep feet feeling warmer.
So, back to the question, can radiators be used? Yes, however, they might need to be very big. If the house is highly insulated, it is far easier to achieve nice low efficient radiator operating temperatures. Furthermore, given the improvements to energy efficiency of heat pumps, it may be more acceptable to operate radiators a little warmer than it was previously, so they may not need to be as big as they were in former years.
Given that heat pumps generally operate at lower temperatures, they tend to require higher flow rates. Make sure your pipe work is not too restrictive.
5. Why are heat pumps better suited to well-insulated houses?
One might think that there would be more to be gained (saved) by fitting a heat pump to an old building with very high fuel bills. However, in poorly-insulated buildings, it is more difficult to dissipate sufficient heat without running the radiators or underfloor at a relatively high temperature (hence lower COP and higher running cost).
By contrast, low radiator or underfloor temperatures are far easier to achieve in well-insulated houses, so the COP can be considerably higher, which is desirable.
However, mains electricity has de-carbonised considerably over the last few years. This shifts the goal posts, and means that heat pumps in old buildings, with mediocre COP, may now show much better carbon savings compared to common methods of heating. Added to this, the generous RHI payments can mean that it’s now affordable to install expensive and efficient emitter systems (e.g. underfloor) that can achieve good COPs in older buildings.
Contrary to popular belief, this issue has nothing to do with heat being lost through walls etc. It’s all down to the water temperature that the heat pump operates at.
Also see my YouTube video
6. I am getting conflicting information about sizing. Why is this?
Traditionally, boiler systems have been sized to be plenty big enough so that they can raise the temperature of a building in a relatively short period of time. However, as discussed in the items above, low operating water temperatures are advantageous for heat pumps, so it is far better to operate a small heat pump for much longer hours with lower water temperatures. Furthermore, big heat pump installations are expensive to buy compared to boilers. All-in-all, this ‘steers’ us away from large-capacity heating systems using heat pumps.
A small heat pump system, that may operate almost continuously 24/7 on the coldest day in winter, could be cost effective all-round. This system may however have an exceptionally slow response time such that it is incompatible with traditional time/temperature control methods, and requires a different user attitude. The slow response issue may become less of a problem in fully-occupied houses and well-insulated buildings where the temperature can be maintained constantly at one level. As you can see, the compromise between the traditional methods and more heat pump friendly methods are open to differing opinions, and all hinges on good controls and knowing how to operate them. It’s no wonder that sizing can be a contentious issue.
One further issue to be mindful of is the possible need for direct-electric back up heating to supplement an overly small heat pump. Direct electric heating is a bad thing in many ways, but from my experience, it is sometimes a lack of understanding of control techniques that is responsible for excessive use of electric back-up heaters, rather than the fault of heat pump being under-sized.
In Sweden and other countries, they have successfully used small heat pumps, but the UK’s MCS scheme (Microgeneration certification Scheme) had a requirement for the system to be large enough so as to supply all of the building’s needs on the coldest day.
In my personal opinion, a lot of heat pump units are too big, and some heat emitter (e.g. radiators) circuits too small.
7. If I leave my heat pump on continuously, will it cost a lot to run?
Firstly, Heat pump systems are (hopefully) self-controlling, so they automatically turn themselves on and off (or up and down). If we leave the unit ‘enabled’ (switched on), it is not necessarily using power all the time; especially if it is turned to ‘low’.
Here is one factor to consider – “the circulating water temperature has a very considerable impact on energy-efficiency. The hotter the water, the lower the heat pump’s energy-efficiency”.
Here is another factor to consider -“there is no point keeping a room heated all day if it is unoccupied”.
Let us consider this further – a radiator may achieve the same room-comfort by being either ‘warm’ all day, or ‘hot’ for just a few hours. Given that the energy efficiency of a heat pump (COP) will improve considerably if the radiator temperature can be kept low, it follows that continuous low-temperature radiators (or underfloor) should be better. BUT, the building will lose more heat if it is warmer for longer. On balance, the extra energy required (with hotter water) to rapidly heat a room can be greater than the energy to keep it warm all the time (with lukewarm water)… to keep it ticking over. All this depends on the type of building (heavy or light-weight) and the occupancy, but experience shows that leaving heat pumps on a low continuous setting can give lower running costs for many situations. They are also far simpler.
8. How should I set Thermostatic Radiator Valves (trv)?
trv valves were developed with boilers in mind. If too many radiator valves are ‘throttling’ the flow, the energy efficiency of a heat pump can be impaired in some cases. It’s best to keep the main rooms TRVs set high, and use trv valves to limit temperatures in bedrooms and extremity rooms. You might find it worth doing some tests – if you open most TRVs up, the house should become hotter than you need. If so, try adjusting the water temperature on the heat pump controller down. You will need to reduce the ‘heating curve’ or simply reduce the heated water setting. This will reduce the temperature of the circulating water. If you can get the house to desired temperatures, all well and good. Experiment with the heat pump’s control to find the right house temperature. Now use the trv’s to ‘trim’ the temperatures of bedrooms and any rooms that are warmer than needed.
9. What about underfloor heating zoning?
This is a similar issue to trv valves. It is normal to have a thermostat in every room (zone), and this controls the water flow to the floor loops. Room zones tend to open/close at irregular times, but generally, a heat pump would ideally prefer to heat most rooms together at one time. Given that there is a degree of natural self-regulation with under-floor heating, it is possible to heat well-insulated buildings by keeping many of the zones open all the time. This is again an issue where installers have differing opinions. However, keeping your setting on the heat pump (the heating curve) low, not only help increase the COP, but also helps to keep zones on for longer – this is particularly beneficial for systems with no buffer cylinder – as many system now are.
10. Can a heat pump also heat the domestic hot water?
It certainly can, but whilst heating to the high temperatures required, the energy-efficiency reduces. However, even low efficiencies are far better than an electric immersion heater. Most of the latest heat pump units have the hot water option built in, so it is usual to use this facility to heat a cylinder. As the insulation levels in buildings increases, the room-heating demand drops, but the hot water demand is, if anything, increasing. It is therefore becoming more important to optimise the hot water function. i.e. the size and design of the hot water cylinder is very important.
Don’t overlook the detail of distribution to your taps. This can be very wasteful.. i.e. the time taken for the tap to run-hot. See DHW
11. Do I need a buffer tank?
The manufacturers do not always agree on this point, but it is suggested that you go with their specific recommendations. A buffer tank is simply a quantity of water that can help to reduce the number of times the heat-pump has to ‘cycle’ (i.e. times it has to stop and start). It is particularly necessary in a larger property where many heating zones are involved. In well-insulated and open-plan houses a buffer tank may not be needed. In these cases, the floor itself can act as the buffer. However, the floor must have sufficient pipe in it with good thermal contact within a thick screed. High water-content radiators can also act, in part, as a buffer, and may be a benefit in some buildings.
Air source heat pumps will have a defrost mechanism. This reverses the heat flow for just a few minutes to melt the ice. This can momentarily cool the radiators. large water volume radiators could minimise this effect. If ‘light’ radiator are used, then a small buffer tank might be advisable.
The relative size of the heat pump also has a bearing here; if the heat pump is relatively large, it is more likely to need a buffer cylinder than a continuously-enabled small one. Furthermore, many systems now have variable ‘inverter’ compressors. These ‘modulate’ their output, so rarely need a buffer cylinder.
In summary; having a buffer tank is playing-safe, and recommended if the radiator or underfloor system is unknown, or un-matched. With well-designed house and well-designed emitter circuits, you might be better off without one. I try to design the buffer tank out, such that it is not needed. It’s always best to go with the manufacturer’s recommendations.
12. What is ‘Weather Compensation’?
As you now know, it is important to keep the heated water as low as possible if high efficiencies are to be attained. It is sensible, if not vital, to adjust the water temperature depending on outside conditions. i.e. your system might require water for the floor at 40°C when at -5°C outside, but require only 32°C when it is +5°C outside. This can be adjusted manually on the heat pump over the seasons. However, Weather Compensation does this automatically, and is an integral part of almost all ground source heat pumps. This is the ‘heating curve’ setting. Not only can this control save energy, but interestingly, a well-adjusted control like this can sometimes achieve accurate-enough room temperatures without any room thermostats.
For ASHPs, the advantage of this facility is less clear. On a cold night, the setting will automatically rise. This effectively causes the heat pump to ‘rev up’ during cold nights. The COP will be poor at this time since the air is cold.
If the building is stone/brick and has high mass and very slow response, it might be more energy-efficient to operate on a fixed daily water temperature based on the average daily outside temperature. This will ‘shift’ the run-time a little towards a time when the air is warmer. That said, this issue can be dealt with by using a night set-back temperature.
It’s worth noting that a well designed emitter system can help here in systems with no buffer. If the emitters are large, they can ‘hold down’ the water temperature naturally, but still be giving out the same average heat quantity. In mild weather, the room thermostat is ‘satisfied’ before the radiators have reached their water set-point. This is effectively natural weather compensation.
13. Is a vertical borehole better than a horizontal pipe trench system?
The energy collected by either of these systems is mostly from stored solar heat. Whilst it is a little warmer at say 100m deep, in reality, either system will produce similar results. The warmer (or ‘less-cold’) the fluid in the ground pipes, the better the performance. However it is too costly to place a large quantity of pipe vertically. Horizontal trenches however can have have more/better pipe contact, but in will be an inferior place. Choice of collector type is usually a matter of cost and practicality. e.g. if land is available a horizontal trench system will usually be cheaper to install than a borehole.
Boreholes become cheaper if a lot are installed, so are likely to be more viable for big projects.
Since excavator costs are generally not excessive, it should be cost-effective to size trenches well (large), and make them deep. Large deep trenches can sometimes out perform boreholes.
If summer cooling is required, then the borehole may prove to be a better option. Free-cooling or passive cooling is likely to be far better from a borehole.
13. How much ground do I need?
The more the better, but this strategy could be un-economic. The average garden is often too small to get efficient heat output from a horizontal trench. However, as houses become better-insulated, then ground collectors don’t need to be so big. Do not underestimate the upheaval of digging trenches, but when the grass and plants have grown back it will all seem worthwhile. This component of the system should last well over 50 years, so extra pipe loops should eventually pay for themselves since the efficiency of the system will be better. Ground conditions will also affect the performance. For example, wet ground conditions assist the heat transfer process. Whereas dry sandy ground is inferior, requiring more land area and far more buried pipe.
There is some debate about the depth, and 2m may be ideal if you are trying to achieve the highest efficiencies. However, excessive cost and health & safety regulations usually mean that shallower trenches are used, and 1.2m to 1.5m seems typical. During deep cold-spells in mid-winter, deeper trenches would be beneficial, but arguably, the extra cost of deeper trenches may not be cost effective. If the collector area is relatively small, there may be less advantage by going deep. The reason being that a small collector will be colder, so does not benefit so much from deeper- potentially warmer ground.
Some claim that shallow trenches recover better in spring. This is true, however there is a counter argument here. Most run-hours are clocked up in mid winter and at this time shallow trenches will be adversely affected by the winter temperatures above.
14. How long will a heat pump last?
Most good ground source types will far outlast even the best quality boiler. They should operate for over 20 years, and with minimal maintenance.
Air source systems are usually exposed to the elements and have a slightly harder life, so may have a similar life to a gas boiler.
Sadly heat pumps can break down, and occasionally are very expensive to repair. Thankfully such instances are very rare.
Unfortunately, we are finding now that spares for some heat pumps, say 15 years old, are becoming hard to get or prohibitively expensive. Many components are generic, but controllers are usually very specific. Manufacturers with a good track record are more likely to provide spares for many years into the future.
15. Can I power a heat pump from a renewable energy source like a windmill or solar panels?
Yes you can, but it seems sensible to consider things separately at present. The middle of winter, when most heat is needed, coincides with times when days are short and solar-electric (PV) output is low. It is unlikely that a PV array would contribute much for the heat pump in the depth of winter. Conversely, in summer, the PV output can be vastly greater than any heat pump’s needs at that time, so usually the mains-grid is used to export PV electricity. There are of course a lot of cold sunny days when PV output could power your heat pump, but dont underestimate the difficulty of matching these two up. In time, software driven systems will no doubt control this nicely
Again, wind is variable, so would be mains grid-linked. Whilst it may be preferable not to have all your eggs in one basket, it can be expensive having several technologies, and debatably better to concentrate on one properly.
There is a further thought here – most PV electric panels are only up to 20% efficient, so even when coupled with a heat pump, they have similar efficiency’s to solar thermal. (solar thermal at low temperature can be over 60% efficient). That said the electricity from solar PV is far more versatile than warm water from solar thermal. PV also have other advantages, e.g. they can be positioned anywhere, unlike solar thermal that needs to be close by
There is also interest in re-charging a ground collector using Solar thermal. This could allow the ground collector to be smaller in area. The advantage could however be minimal if it’s not done well, and it could sometimes be akin to running your rain gutters into the sea to collect water. Unless the heat pump actually ‘sees’ a higher source temperature in winter due to re-charging in summer, there would be no advantage. This method could be an excellent way of improving a system that was originally fitted with collector loop that is too small (as was common on some older systems). You could use low-cost un-glazed pool solar panels here.
16. Is there a real environmental benefit?
A decade of more ago, when a large chunk of our electricity was derived from coal, the environmental benefits of some heat pump systems were marginal. At that time, it was only highly energy-efficient heat pumps that showed very good environmental benefits. However, over recent years, the electricity used to power them has got far ‘greener’. This is in part due to highly efficient gas power stations, and in part due to the increase in renewables. This means that you don’t need to have an exceptionally high COP to be able to show a good Low Carbon advantage.