Reading a heat loss calculation

The heat loss calculation is the foundation of a heat pump design, and you are entitled to see it. It is usually a spreadsheet, one block per room, and it looks more forbidding than it is. Every line is the same three numbers multiplied together.

What a room block contains

A room's block has a header giving its name, its floor area, its ceiling height and its design temperature, then a row for each surface it loses heat through.

Each row carries an area in square metres, a U-value in W/m²K, a temperature difference in kelvin, and the product of the three in watts. Underneath, a ventilation line: the room volume, an air change rate, and the watts spent warming incoming air, followed by a total for the room.

So a living room might read:

The checks worth doing

Is there a row for every room?

A single whole-house figure is not the method. Room-by-room is what BS EN 12831 requires and what MIS 3005-D references. Without it there is no basis for sizing individual radiators, and no way to find the room that will end up setting the flow temperature for the whole house.

Are the design temperatures right?

Internal temperatures should be at least the tabulated minimums: 22 for a bathroom, 21 for living and dining rooms, 18 for bedrooms, halls, kitchens and toilets. Anything below those has been reduced to make a total look better.

The outside design temperature should be stated and should be plausible for the location, somewhere between −0.2 for Plymouth and −5.9 for Glasgow depending on which column was used.

Are ground floors on the right temperature difference?

This is the single most common technical error. A solid floor in contact with the ground is calculated against the local annual average air temperature, not the outside design temperature. In the table above the walls see 22.7 K and the floor sees 9.7.

If every row in a block shows the same temperature difference, including the ground floor, the calculation is wrong and the house is being made to look worse than it is.

Do the U-values describe this house?

These are the numbers most open to wishful thinking, in either direction. Some rough figures for context:

An assumed cavity fill that was never installed, or an assumed loft depth nobody measured, produces a heat loss well below the truth, and a heat pump that cannot keep the house warm in February. The opposite error, assuming the worst everywhere, produces an oversized machine that cycles.

Are areas counted once?

Window and door areas should be subtracted from the wall they sit in. Counting a 3 m² window as both window and wall adds its area twice at two different U-values.

Is the air change rate honest?

A rate of 0.5 air changes an hour describes a well sealed modern house. A Victorian house with sash windows and a suspended floor is nearer 1.5, and a room with an open flue more still. A whole calculation at 0.5 on an old house is a calculation designed to produce a small answer.

Does the total make sense?

Divide the whole-house total by the total floor area. A well insulated house lands around 40 W/m². An average one is 60 to 80. A poor one is comfortably over 100. A figure of 15 or 200 is not impossible but needs explaining.

Then compare the heat pump with the total. MIS 3005-D requires the heat pump to cover 100 per cent of the heat load at the design conditions, so a 6 kW heat loss wants a heat pump that delivers 6 kW at your outside design temperature and your flow temperature, not at 7 °C and 35 °C. A machine sold as 8 kW may be 5.5 at your conditions, and one sold as 8 kW that reaches it only with a 3 kW immersion heater is not an 8 kW machine at all.

The failure this is all guarding against

Oversizing is the commonest fault in heat pump installations, and its consequences are not obvious to the person living with it.

An oversized machine cannot turn down far enough to match a mild day's load, so it runs in short bursts. Each start costs efficiency, the compressor wears, and the system spends most of its life in its least efficient mode. The house is warm and the bills are higher than they should be, and nobody connects the two.

The heat loss calculation is the discipline that prevents this, which is why the standard requires it and why it is worth twenty minutes of your attention before signing.