Skip to content
OneKitly

Efficiencies Multiply, They Never Add

Published 10/28/2026 · 3 min read · Everyday calculators

Daniel Okonkwo

Daniel Okonkwo — Front-end developer and tech writer at OneKitly

Web performance · File formats

Checked against 2 sources

View profile →
In short

Efficiency is useful output over total input, and when energy passes through several stages each stage takes its cut of what the previous one handed over. Multiply, do not average: a coal plant at 33 %, a grid at 95 % and a switching power supply at 88 % combine to 0.33 × 0.95 × 0.88 = 27.6 %, not to the 72 % an average would suggest. That is why a 9 W desk lamp is really drawing 33 W of primary energy, and why 72 % of what left the fuel never reaches the light. Improving one stage helps in proportion to its share of the losses, which is almost always the worst stage rather than the most visible one.

Advertisement

33 % at the power station, 95 % on the grid, 88 % in the power supply: the chain delivers 27.6 %, so a 9 W lamp costs 33 W at the source.

Long chains punish you twice

Every extra conversion costs something, even a good one. Ten stages at 95 % each — nothing wrong with any of them individually — leave 60 % of what you started with. That is the argument against elaborate energy paths: storing electricity as hydrogen and burning it back, for instance, stacks three or four conversions and arrives around a third of the way. It is also the argument for direct use wherever it is available. A heat pump moving warmth rather than making it can deliver three or four units of heat per unit of electricity, which looks like an efficiency above 100 % and is really a different quantity — a ratio of moved to spent, not of produced to consumed.

Where the number is quoted dishonestly

Almost always by choosing where the chain starts. An electric car quoted at 90 % efficiency is being measured from the battery terminals, which is true and silent about the power station. A gas boiler quoted above 100 % is using a convention that ignores the energy recovered from condensing water vapour in a way the older convention did not. Neither figure is fabricated; both answer a narrower question than the reader assumes. The defence is to ask what went in at the far left of the diagram — fuel dug out of the ground, usually — and to compute the whole product yourself.

Running total
Coal to desk lamp, stage by stage
StageEfficiencyRunning total
Power station33 %33.0 %
Grid and transformers95 %31.4 %
Power supply88 %27.6 %
Lost overall—72.4 %

Worked with our own calculator

Efficiency calculator

Given

Useful output energy
80
Total input energy
100

Result

Efficiency
80%

These figures are produced by the calculator below, not typed in by hand — they are recomputed whenever the tool changes.

Run it on your own figures →
Advertisement

Frequently asked questions

Which stage should be improved first?
The worst one, by a wide margin. Taking the power station from 33 % to 40 % lifts the whole chain from 27.6 % to 33.4 %, a gain of nearly six points. Taking the power supply from 88 % to 95 % lifts it to 29.8 %, barely two points, because that stage was already keeping most of what it received. The rule of thumb is that a stage can only give back what it is currently losing, so ranking the stages by their losses ranks the opportunities. Attention usually goes the other way, to the stage that is nearest and cheapest to change.
Why can no heat engine reach 100 %?
Because an engine that turns heat into work must dump some heat somewhere colder, and the fraction it is forced to dump depends only on the two temperatures involved. A turbine running on steam at 550 °C and rejecting to a river at 20 °C cannot exceed about 64 % even if built perfectly, and real losses take a coal plant to roughly half of that ceiling. The limit is not an engineering shortfall to be designed away; it is a statement about heat itself, which is why efficiency figures near 40 % represent genuinely good machines rather than sloppy ones.

Articles you may find interesting

All guides →
ExplainerWater Takes Nine Times More Heat Than Steel to Warm UpTwo kilos of water need 83.72 kJ for ten degrees. The same two kilos of aluminium need 18, and of steel 9. That single number explains kettles, radiators and coastal weather.ComparisonHeat Pump or Gas Boiler: the SCOP Your Local Prices DemandBefore any grant, the price of electricity divided by the price of gas sets a seasonal efficiency the heat pump must beat just to break even on running cost. In late 2025 that number was 1.53 in Portugal and 2.79 in Germany — and on old radiators a German heat pump can cost more to run than the boiler it replaced.ComparisonInsulate or Replace the Heating: Which Repays First, and Why the Answer InvertsPut both on one axis — euros paid per kilowatt-hour a year of permanent saving — and the comparison becomes decidable. The same loft job costs 0.83 euros per kilowatt-hour a year in a 1965 house and 5.00 in a 2005 one, a factor of six. That, and not the age of the building by itself, is what inverts the verdict.How-toHow to Estimate the Running Costs of a HomeAdd up energy, water, heating, and maintenance with simple rules of thumb so you know what a home really costs to run each month.ExplainerFive Points of Learning Rate Turn Fifty Repetitions Into Four HundredAt an 85 % learning rate, a task that takes 20 minutes first time reaches 8 minutes after 50 attempts. At 90 % it needs 415 — the same task, five points apart.ExplainerPSI to Bar: Convert Tire Pressure Both WaysConvert PSI to bar and back with the exact factor, learn the 1 bar ≈ 14.5 PSI shortcut, and set tire pressure right in either unit.

Related tools

Sources

Spotted a mistake in this article?