How much of the electricity you pay for reaches the battery?

The gap between the meter and the battery is not a percentage. It is a constant, and that changes what you should do about it.

9 September 2026 · updated 10 September 2026 · 7 min read

Charge at home and two numbers disagree. The car says it added a certain number of kilowatt-hours to the battery; the electricity you were billed for is more than that. The difference is usually described as an efficiency percentage — 90%, or 85%, or whatever a forum thread settled on — and that framing is what makes the practical advice come out wrong.

Across 23 home AC sessions on a 2023 Model Y, the gap averaged about 11%. But it ranged from under 5% to over 25%, and the pattern in that spread turns out to be more useful than the average.

The loss is a constant, not a percentage

Expressed as power rather than as a proportion, the same sessions look completely different. Instead of a percentage that swings by a factor of five, there is a draw of roughly 0.2 kW that barely moves whatever the charging speed:

That is what a fixed overhead looks like. While a car is charging it is awake: the battery management system is running, the coolant pump may be running, the 12 V systems are being fed. None of that scales with how fast the electricity is arriving. It is simply on.

So the percentage is not a property of the charger or the car — it is arithmetic. Two hundred watts is 17% of a 1.2 kW charge and 5% of a 3.7 kW one. Same loss, different denominator.

Other owners' numbers agree on the mechanism

This was put to a UK EV forum and a Spanish Tesla group, and the responses converged on the same explanation from three directions.

Any time the car is charging it is on so there is a base load of maybe 300W regardless of what is going into the battery.

Over the years I have seen this topic raised quite a few times and often with recorded data similar to yours… they all bracket around 15% for granny charging set lower than 10 amps and 8% for normal UK charge points at 32 amps.

A separate owner with a metered Tesla Gen3 wallbox — a real meter, rather than the car's own reporting — measured 3% lost on a session averaging 3.9 kW. That works out at about 0.12 kW of overhead: lower than our figure, from the better instrument. Another put the base load at 300 W independently, in a different country, without having seen the first.

The constant is two things, not one

Asked where his 300 W came from, that second owner had measured it directly: diagnostic software reading the high-voltage battery on an older Model S, 300 W at peak and never below 200 W. That figure is worth separating from ours, because it is not measuring the same thing.

What he measured is the car's own housekeeping — the electricity the vehicle draws for itself while it charges. What a wall meter measures is that, plus the energy lost turning household AC into something the battery can take. Both are per hour rather than per kilowatt-hour, so both behave as a constant, but only the first is the car being awake.

That also explains the spread. An older Model S with more to keep running draws 200–300 W for itself before any conversion loss is counted; a newer car reading 0.12 kW at the meter is paying less for both. So the honest range for the constant is something like 0.12 to 0.3 kW, and where a given car sits in it depends on its age as much as on its charger. The mechanism is not in dispute. The exact number is, and anyone quoting a single figure to three significant places is guessing.

What to do about it

Because the loss is per hour and not per kilowatt-hour, the cheapest way to put a given amount of energy into a car is to put it in quickly. Eight hours on a 3-pin lead spends four times as long paying the same overhead as two hours on a wallbox, for the same kilowatt-hours in the battery.

• Charge at the highest current your installation safely allows, rather than trickling.
• If you are on a cheap overnight rate, the rate you actually pay per kWh in the battery is 5–17% higher than the number on your tariff. On 7p that is nearer 8p on a wallbox, and 8.5p on a 3-pin lead.
• Very short top-ups look terrible as percentages and do not matter — the overhead is real but tiny in absolute terms.
• Solar charging is free in euros but not in kilowatt-hours: the overhead eats surplus that could have gone into the battery or been exported.

The exception that reverses it

Below about 10 °C, charging faster can cost more rather than less. Several owners reported that their cars will heat the battery once the charge rate goes above roughly 3.5 kW in the cold, and battery heating is not a small draw.

I know my Ioniq will heat the battery if it's below about 10C and you charge at 3.5kW or more. So ironically in winter charges can actually be more efficient on the granny charger… At 0 degrees for example the battery heater will run for maybe 40-60 minutes, "wasting" around 2kW of a 7kW charge.

Another owner measured around 700 W of battery heating near freezing on a Renault Zoe, running for about half an hour. So in a cold climate the advice inverts: the slow charge avoids waking the heater at all, and can be the efficient one. In southern Spain, where these measurements were taken, that case never arises — which is exactly why it needed someone else's winter to find it.

How to measure it on your own car

• The number you want is energy from the wall against energy into the battery, for the same session.
• For the wall figure, a metered wallbox or a plug-in meter is the only trustworthy source. Power reported by the car is a fair approximation for AC charging, but it is not a meter and it does not include what the wallbox itself consumes.
• Then divide the difference by the session's duration, not by its energy. That gives you the constant in kilowatts, which is the number that predicts your next session.
• Note the outside temperature. It is the variable most likely to explain a session that does not fit.

These figures come from evlogr, which logs a Tesla’s drives and charges automatically and prices each charge against your own time-of-use tariff. Free on one car.