Common heat loss survey mistakes

The most widespread shortcut is "100 W per square metre" or some variant. Multiply the floor area by 100, and you have your heat loss. Simple, fast — and wrong.

The problem is that 100 W/m² ignores everything that actually determines heat loss: the number of external walls, the window area, the construction type, the insulation level, the ventilation rate, the design temperature difference, the room height. Two rooms with identical floor areas can have wildly different heat losses depending on these factors.

A well-insulated mid-floor flat with one small external wall might lose 30 W/m². A poorly insulated end-terrace living room with a large bay window might lose 120 W/m². Applying 100 W/m² to both rooms gets neither right.

Rules of thumb exist because they are easy. But ease is not the same as usefulness. If you are sizing radiators or assessing whether a heat pump will work, you need room-by-room figures, not a guess based on floor area.

Even when a whole-house heat loss figure is calculated properly, it is not enough for radiator sizing. A house with 8 kW total heat loss does not mean every room needs a proportional share of 8 kW based on its floor area.

Heat loss varies room by room depending on the number and size of external surfaces, the construction, the ventilation and the target temperature. A bathroom with one small external wall and a high target temperature has a different heat loss profile from a large living room with two external walls and a bay window.

The whole-house figure hides the problem rooms. It is perfectly possible for a house to have "enough" total radiator output while one room is under-emitted and permanently cold. This matters most when switching to a heat pump, where lower flow temperatures mean every radiator has less headroom.

Radiator catalogues typically quote outputs at Delta T 50 — that is, with a mean water temperature 50 C above the room temperature. For a 21 C room, that implies a mean water temperature of 71 C, which corresponds roughly to flow at 80 C and return at 60 C.

Modern condensing boilers run more efficiently at lower temperatures. Heat pumps typically operate at flow temperatures of 35 to 55 C. At these lower temperatures, the same radiator produces significantly less output than the catalogue figure suggests.

A radiator rated at 1,500 W at Delta T 50 might only deliver 750 W at Delta T 30 — roughly half. Comparing a room's heat loss against the catalogue output without correcting for the actual flow temperature and Delta T gives a false sense of comfort. The radiator looks adequate on paper but is undersized in practice.

A Victorian solid brick wall has a U-value of roughly 2.0 to 2.1 W/m²K. A modern insulated cavity wall might be 0.3 W/m²K or better. That is a seven-fold difference in thermal performance.

Mistakes happen in both directions. Sometimes an old solid wall is assumed to be a cavity wall because "it looks like a normal brick wall from inside." Sometimes a wall that has been retrofitted with internal insulation is assumed to be original construction. In older buildings, the wall build-up is often unknown, and a wrong guess makes a large difference to the calculated heat loss.

The U-value of the external walls is typically the single most influential assumption in a domestic heat loss calculation. Getting it wrong — even by a reasonable amount — can shift the total heat loss by 20% or more.

Fabric heat loss gets all the attention — walls, windows, insulation. But in many older houses, ventilation heat loss accounts for 30% or more of the total.

Draughty windows and doors, unsealed suspended timber floors, open chimneys, poorly fitted loft hatches and background trickle vents all allow warm air to escape and cold air to enter. A house that has been well insulated but not draught-proofed still loses substantial heat through air movement.

Many quick heat loss estimates focus entirely on U-values and ignore the airtightness side entirely. The result is a figure that underestimates total heat loss — sometimes by a large margin in draughty buildings.

A heat loss calculation produces a single number for each room — say, 1,247 W. That level of precision can create a false sense of accuracy. In reality, the figure depends on assumptions about wall construction, air leakage, thermal bridging and design conditions, many of which carry genuine uncertainty.

A room calculated at 1,247 W might realistically be anywhere from 1,000 to 1,500 W depending on which assumptions turn out to be right. The calculation is still useful — it gives a structured, reviewable estimate that is far better than a guess. But treating the output as though it were a laboratory measurement leads to bad decisions.

The right approach is to understand which inputs are measured, which are assumed, and where the biggest uncertainties lie. That way you can make sensible judgements rather than treating a single number as gospel.

When there is uncertainty in the heat loss figure, the temptation is to "just add a bit" to the boiler or heat pump size. If the calculation says 8 kW, fit a 12 kW generator "to be safe."

The problem is that oversizing creates its own problems. An oversized boiler will short-cycle because it cannot modulate low enough to match actual demand. An oversized heat pump cycles too, and also costs more upfront. Neither runs efficiently, and both wear faster than a properly matched unit.

The better response to uncertainty is not to round up — it is to understand the uncertainty. Which assumptions have the biggest effect? Can any of them be improved with better survey evidence? What is the plausible range? A transparent calculation with visible assumptions beats a hidden margin every time.