Efficiency in burning gas, is the amount of heat energy you get out, relative to what you put in. It’s usually expressed as a percentage:

**Efficiency = energy out / energy in **

** **

You can’t get out more that you put in – not all the energy available is captured to do something useful – some is lost *(this is known by some as an example of the Tanstaasfl effect – “there ain’t no such thing as a free lunch”) *– so efficiency is always less than 100%.

**Gross input** is the total amount of heat released by burning gas at a certain rate, and measured in kilowatts (kW). There’s no more available, and if all of this was put to use, and none was lost, the efficiency would be 100%. Some is always lost – in the case of a boiler, principally via the flue.

If we burn a volume of gas, energy is released, measured in joules (J) or megajoules (MJ). If we burn gas at a certain rate (in cubic metres per hour m^{3}/hour), energy is produced at a certain rate or power, expressed in watts (W) or kilowatts (kW). The relation between volume and energy is the calorific value of the gas. According to National Grid figures, if we burn the gas fed into the UK network at the rate of 1 m^{3}/hour, the gross input will be in the range 10.42 to 11.9kW. The figures differ in other territories. |

When natural gas (nearly all methane) is burned completely, the products of combustion are carbon dioxide CO**2** and water vapour H2O. It used to be accepted that the water vapour was lost via the flue, carrying its latent heat of vaporisation into the atmosphere. This latent heat was never available as a heat input to the boiler. What __was__ available was the **net input**, and the figure for this is obtained by dividing the gross input by 1.11.

Enter the condensing boiler. In one of these, some of this latent heat is recovered by condensing the water vapour into liquid water (the condensate – the stuff that leaves via the plastic pipe – usually white). This means that while the boiler’s useful output can never reach the gross input figure (100% efficient), it can exceed the net input, and if we compare output to __net__ input, we can get a figure of more than 100%. It’s not a meaningful measure of efficiency, it’s just comparing what we get out when we condense the water vapour, with what we would have put in without any condensation – comparing an apple to an orange if you like.

Nothing beats a worked example so here are some manufacturer figures for an actual boiler, a Baxi Duotec Combi 33 HE A, for central heating in condensing mode at full power:

Gross heat input: 32.1 kW

Net heat input: 28.9 kW (a factor 1.11 lower)

Heat output: 30.3 kW

The efficiency (heat output/gross input) is 94.4%.

The heat output divided by the net heat input expressed as a percentage is 105%. This manufacturer has been helpful in giving both gross and net figures so we know where we are. These days, often only net input is given, without being stated as such, so daft figures like our 105% can be obtained.

Even more helpfully, the manufacturer has given the output for the boiler in non-condensing mode (return water temperature too high to allow condensation to occur) – 28kW, giving an efficiency of 87%.

Manufacturers quote efficiency figures as SEDBUK values – Seasonal Efficiency of Domestic Boilers in the UK. This is an attempt to give a meaningful measure under realistic operating conditions. We’d expect it to be lower than the optimum case of the boiler condensing all the time, and higher than for the boiler not condensing at all, and so it is – in this case 91.1%.

Additional sources of confusion:

- Older boilers give input figures as gross, but don’t necessarily tell you.
- Newer boilers give input figures as net, but don’t necessarily tell you. If you divide output by input and get a daft efficiency number of more than 100%, you know you’ve been given the net figure. Multiply the
**input**by 1.11 to restore sanity. - In relating boiler input to gas rate, manufacturers use different values for the calorific value of the gas. 10.5 is a popular figure (you’ll find a range of 10.4 to 10.6 used for different power versions of the same boiler), 10.75 another. That range is enough to turn a boiler that’s burning 2% above nominal rate into an apparent 5.5%. To check whether a boiler’s healthy, where possible, I don’t bother calculating power, I just compare the measured gas rate in m
^{3}/hour with the manufacturer’s figure.

Clarity is hard to achieve, but I hope after reading this you understand how daft figures for efficiency are easy to arrive at, and now know how to avoid them.