Why a Kettle Is Almost Perfect and a Power Station Isn't
Boil a litre of water in an electric kettle. Almost every joule of electricity that enters becomes heat in the water — a small amount escapes through the walls and as steam, but the conversion is well over 90% efficient.
Now generate that electricity in a gas-fired power plant. The best combined-cycle plants convert somewhere around 60% of the fuel's chemical energy into electricity. Older plants manage considerably less.
The kettle isn't better engineered than the power station. It's doing an easier job. And the reason has nothing to do with engineering quality — it's a hard limit written into thermodynamics.
Two Kinds of Energy Conversion
Converting to heat is easy. Any form of energy will degrade into heat given the chance. Electricity through a resistor, friction, chemical combustion — all of it ends up as thermal energy without much effort. Resistive electric heating is essentially 100% efficient, because heat is where energy goes naturally.
Converting heat into work is hard. This is the direction that runs against the grain, and it's governed by the second law of thermodynamics.
Any heat engine — a steam turbine, a petrol engine, a jet engine — takes heat from a hot source, converts some of it into mechanical work, and dumps the rest into a cold sink. The maximum possible fraction it can convert is fixed by the two temperatures:
Carnot efficiency = 1 − (T_cold / T_hot)
Both temperatures in kelvin.
A steam plant running at 550°C (823 K) exhausting to a condenser at 30°C (303 K):
1 − (303 / 823) = 0.632
63% is the theoretical ceiling. No design, no material, no amount of investment gets past it. Real plants achieve less because of friction, heat loss, and pumping losses.
This is why power generation is stuck in the 35–60% range while a kettle is at 95%. They're not comparable tasks.
Where the Energy Actually Goes
A petrol car
Of the chemical energy in the fuel:
- Roughly 20–30% becomes mechanical work at the wheels under typical driving
- The rest leaves as exhaust heat and radiator heat
That's why a car's cooling system is so substantial and why the exhaust is hot. Most of the petrol you buy is heating the atmosphere.
Diesel engines do somewhat better, mainly because they run at higher compression ratios and therefore higher peak temperatures — back to the Carnot relationship.
Electric motors, by contrast, don't convert heat to work at all. They convert electrical energy to mechanical energy directly, which sidesteps the heat-engine limit entirely. Efficiencies above 90% are routine. The thermodynamic penalty doesn't disappear — it moves upstream to wherever the electricity was generated — but a motor itself is not a heat engine.
An incandescent bulb
Around 2–5% of the electricity becomes visible light. The rest becomes heat.
An LED converts a much larger fraction to light. The improvement isn't marginal — it's the difference between a device that happens to emit light while heating a room and one that actually does the job you wanted.
A gas boiler versus a heat pump
A condensing gas boiler can exceed 90% efficiency — it converts chemical energy directly to heat, the easy direction.
A heat pump routinely achieves a coefficient of performance of 3 or 4, meaning it delivers three or four units of heat for every unit of electricity consumed.
That looks like it breaks conservation of energy. It doesn't. A heat pump isn't creating energy — it's moving heat from outside to inside. The electricity powers the pumping, not the heating. Since it's transporting energy rather than converting it, it isn't bound by the same limit.
Its performance does fall as the outdoor temperature drops, because moving heat uphill against a bigger temperature difference takes more work. Which is, again, the same thermodynamic relationship showing up from the other direction.
Comparing Across Units
Efficiency comparisons only work when everything is in the same unit, and energy is quoted in a lot of different ones depending on the field:
| Unit | Typical domain | Approximate equivalent |
|---|---|---|
| Joule (J) | SI base unit | — |
| Kilowatt-hour (kWh) | Electricity billing | 3.6 million J |
| BTU | Heating, cooling (US) | ~1,055 J |
| Therm | Gas billing | ~105.5 million J |
| Calorie (cal) | Chemistry | ~4.184 J |
| Kilocalorie (kcal) | Food labelling | ~4,184 J |
| Electronvolt (eV) | Particle physics | tiny |
The Energy Converter handles the translation:
- Enter the value.
- Select the source unit.
- Read equivalents across every supported unit.
This is genuinely necessary for practical comparisons. If your gas bill is in therms or cubic metres and your electricity bill is in kWh, you cannot compare heating costs until both are in the same unit — and then adjusted for the efficiency of each appliance.
Worked example. Suppose gas costs €0.08/kWh and electricity €0.28/kWh.
- Gas boiler at 92% efficiency: €0.08 / 0.92 = €0.087 per kWh of delivered heat
- Heat pump at COP 3.5: €0.28 / 3.5 = €0.080 per kWh of delivered heat
- Resistive electric heater at 100%: €0.28 / 1.0 = €0.28 per kWh of delivered heat
The heat pump and the boiler are close; the resistive heater is more than three times either. The exact answer depends heavily on local energy prices and on the heat pump's actual seasonal performance in your climate, so run the numbers with your own figures.
Primary Energy and the Chain
A single efficiency figure can be misleading if it ignores what happened upstream.
Consider heating with resistive electricity where the grid runs on gas plants at 45% efficiency, with around 8% lost in transmission:
Gas energy in → 0.45 → electricity → 0.92 (grid delivery) → 1.0 (heater)
Overall: ~41%
Versus burning the gas directly in a 92% boiler at home: 92%.
By that measure, resistive electric heating is thermodynamically poor. Now redo it with a heat pump at COP 3.5:
0.45 × 0.92 × 3.5 = 145%
Over 100% relative to primary fuel, because the pump moves ambient heat that didn't come from the fuel at all.
And on a grid with substantial renewable generation, the upstream conversion figure changes entirely, since wind and solar aren't heat engines.
The point isn't that any one technology wins. It's that efficiency figures are only meaningful when you're clear about which boundary they're drawn around — the appliance, the whole chain, or the primary fuel.
Practical Tips
Ask what boundary an efficiency figure covers. Device-level, system-level and primary-energy figures for the same setup can differ enormously.
Convert everything to one unit before comparing costs. Preferably kWh, since electricity is already billed that way.
Divide cost by efficiency to get cost per useful unit. That's the number that matters, not the headline price per unit of fuel.
Remember that COP above 1 isn't a violation of anything. Heat pumps move energy; they don't manufacture it.
Waste heat isn't always waste. Combined heat and power plants use the rejected heat for district heating, pushing total fuel utilisation well above what the electrical efficiency alone suggests.
Insulation beats efficiency. Reducing the amount of heat you need is generally cheaper per unit saved than improving the efficiency of producing it.
FAQ
Why can't a power plant be 100% efficient? Because it's a heat engine, and the second law of thermodynamics limits how much heat can become work based on the temperature difference between the hot source and cold sink.
How can a heat pump be more than 100% efficient? It isn't converting energy — it's moving existing heat from outside to inside. The electricity powers the transport, not the heat itself.
Is electric heating efficient? Resistive electric heating converts nearly all electricity to heat, so it's efficient at the appliance. Whether it's efficient overall depends on how the electricity was generated and what the alternative is.
What's the difference between a calorie and a kilocalorie? A kilocalorie is 1,000 calories. Food labelling uses kilocalories, though they're often written as "Calories" with a capital C.
Why do BTUs still appear on air conditioners? Historical convention in the US HVAC industry. One BTU is roughly the energy to raise a pound of water by one degree Fahrenheit.
Does a higher efficiency always mean lower cost? No. Cost per unit of delivered energy depends on both efficiency and the price of the input fuel. A less efficient device running on cheap fuel can be cheaper to operate.
The Takeaway
Efficiency isn't a measure of how well something is built — it's mostly a measure of what physical process it's performing. Converting energy to heat is nearly free; converting heat to work is capped by temperature. Once you know which category a device falls into, its efficiency figure stops being surprising.
Convert between joules, kWh, BTU, calories and more free with the Energy Converter at sadiqbd.com — no sign-up, instant results.
