Efficiency
A heat pump is sized from a heat loss calculation, not from the old boiler
The survey that should precede a low-temperature heating system works out how fast the building loses heat room by room, and skipping it is how systems end up too large.
By Manish Trivedi4 min read
The boiler was never a measurement of the house
The obvious way to size a replacement heating appliance is to look at what is currently fitted and match it. This fails, because the existing boiler was almost certainly chosen with generous margin, often sized around hot water delivery rather than around space heating, and quite possibly installed before the loft was insulated or the windows changed.
A gas or oil boiler tolerates being oversized reasonably well. It modulates down, it cycles, and the inefficiency is real but modest enough that nobody investigates. Carrying that same margin into a low-temperature system produces a machine that spends most of the year unable to run gently enough for the load it faces.
So the starting point has to be the building rather than the equipment it currently contains. Measure the house. The appliance follows from that, and not the other way round.
What the calculation actually computes
A heat loss survey works out, room by room, the rate at which heat escapes when it is cold outside and comfortable inside. Two mechanisms are added together: conduction through each element of the fabric, calculated from its area and its thermal performance, and ventilation loss from air exchanging with outside at an assumed rate.
Each of those inputs is an estimate, and the ventilation figure is the softest of them, because nobody knows the air change rate of a particular house without testing it. Wall construction is often uncertain too, especially in older buildings where what is behind the plaster is a matter of inference.
The output is a rate of heat loss for the whole house and for each room, and the second of those is what sizes the radiators. It’s a calculation with error bars, and a surveyor who presents it as exact is overselling it.
The design temperature is a choice, and it moves everything
The calculation is run against an outside design temperature — a cold condition the location experiences, but not the coldest it has ever seen. Choosing a lower design temperature produces a larger system that copes with rarer events and runs at part load for the rest of its life.
This is a genuine trade rather than a technical detail. Sizing for the worst night in a decade means every other night is served by a machine larger than needed, and modern systems handle a brief shortfall by running longer and letting the house cool a fraction rather than by failing.
Standard practice uses published local design temperatures for exactly this reason. Asking which figure was used, and why, is a fair question to put to a surveyor.
Oversizing costs more than it protects
A heating system substantially larger than its load reaches temperature quickly and then has to stop, and a machine that starts and stops repeatedly runs less efficiently than one running steadily at low output. It also costs more to buy, may require a larger electrical supply arrangement, and can be noisier in operation.
The temptation to oversize comes from a reasonable fear of the system failing to keep up on a cold night, and it is understandable. It is also the wrong response, because the correct protection against that is a properly chosen design temperature and adequate emitters rather than a bigger appliance.
Undersizing isn’t harmless either, and this is the honest counterweight. A system that cannot reach comfort in cold weather will be supplemented by electric heaters, which is the most expensive heat available.
A survey that does not visit every room is not a survey
The room-by-room element is the part most often skipped, and it is the part that determines whether the emitters in each space are adequate. A whole-house figure divided across existing radiators is not the same exercise, and it is how a house ends up warm downstairs and cold in one back bedroom.
A proper survey involves measuring rooms, identifying construction, noting insulation where it can be seen, counting external walls and windows, and asking about draughts and use. It takes hours rather than minutes and it should produce a document you can read.
If a proposal arrives without one, the number in it came from somewhere else. Ask where.
Fabric first is not a slogan here
Because the appliance is sized from the heat loss, reducing that loss before the survey shrinks the system that gets specified. Insulation and draught-proofing carried out afterwards leave you with a machine sized for a house that no longer exists.
That said, waiting indefinitely for a perfect building is its own trap, and many houses are improved enough already. The practical sequence is to complete whatever fabric work is definitely going to happen, then survey, then specify — rather than doing them in whatever order the trades become available.
Common questions
Can I size a heat pump from my annual gas consumption?
It is a useful cross-check and not a substitute for a room-by-room calculation. Annual consumption tells you roughly how much energy the house needs across a season, which helps validate a survey that produces a wildly different picture, but it says nothing about how the load is distributed between rooms. That distribution is what sizes the emitters.
What if the survey says my radiators are too small?
That is a normal outcome rather than a problem with the survey, because existing radiators were sized for much hotter water. The options are to enlarge the emitters in the rooms that fall short, to reduce the heat loss in those rooms, or to accept a higher flow temperature with the efficiency penalty that brings. Which is best depends on the room and the cost of the work.
Is a bigger heat pump safer in a cold snap?
Not really, and it carries continuous costs to buy that insurance. A correctly sized system handles cold weather by running for longer periods, which is what it is designed to do, and a brief shortfall on an exceptional night shows up as the house cooling slightly rather than as a failure. Persistent inability to reach temperature is a sizing or emitter problem, and that is a different diagnosis.
Deputy editor, Power Your Roof
Manish has been reporting on solar basics, batteries, bills & tariffs since long before it was fashionable and would rather show the working than assert the conclusion.
