EV Charging Speeds Chart 2026: 376 Combinations, Rated Against Real Delivered kW

The 2026 master charging speeds chart#

The 2026 master chart covers 376 combinations of connector, cable rating, supply type and vehicle onboard limit, and real delivered power across them runs from 2.08 kW to 362 kW. Delivered power sat a mean of 8.9% below the rated figure, and the limiting factor is named on every row.

The chart comes first because it is what the page is for. Four inputs describe every row. The connector decides what can physically be plugged in and, in the Type 1 case, sets a hard ceiling of its own. The cable or charger rating is the figure printed on the equipment. The supply is what the property or the site can actually deliver. The vehicle onboard limit is what the car will accept, fixed when it was built. Two output columns follow: rated kW, which is the lowest of those electrical ceilings rather than the largest number among them, and real delivered kW, which is what EV Cable Hub's 2026 measurement recorded at the vehicle inlet.

The rule that governs every one of the 376 rows is that charging speed is the lowest of the ceilings, never the highest. A 22 kW cable on a 100 A single-phase supply feeding an 11 kW car does not charge at 22 kW, at 11 kW or at anything in between: it charges at 6.76 kW, because the single-phase supply is the floor under all three. This is why a driver can spend money on a faster cable and see the same number on the dashboard the following morning.

The column that matters most is the last one. A row that reports 6.42 kW without saying why is a row that tells a reader nothing they can act on. The limiting factor column names the ceiling that was actually binding when the measurement was taken, and it is the difference between a chart you look at and a chart you use. Where two ceilings sat within 0.02 kW of one another both are named, which is why some rows read "Cable rating and Vehicle onboard charger" rather than picking one arbitrarily.

Two scope notes belong with the chart rather than in a footnote. The master grid models four supply types, and the three-phase case in it is a 100 A per phase domestic supply, so its ceiling is 22.0 kW; the 43 kW three-phase figure quoted elsewhere on this page comes from the 24 sessions recorded on a 63 A per phase supply and sits in Table 8 rather than in the grid. And the slowest setting EV Cable Hub measured anywhere in 2026, 1.24 kW on a Mode 2 lead at its 6 A eco setting, appears in Table 35; the master grid starts at the 10 A setting, which is where the delivered figure of 2.08 kW comes from.

The 2026 headline findings#

Real delivered charging power came in a mean of 8.9% below the rated figure across 3,164 measured sessions in 2026. On AC the household supply was the binding ceiling on 41.2% of combinations; on DC the vehicle was the limit on 58.3%.

The whole chart rests on one principle, and it is worth stating before any of the numbers. Charging speed is the lowest of four ceilings, not the highest of them, and the rating printed on any single component is a design maximum rather than a delivery promise. A 22 kW cable is a cable that will not overheat at 22 kW. It is not a cable that produces 22 kW, and nothing about buying one changes the fuse in the meter cupboard or the charger built into the car.

The four ceilings bind at very different rates, and knowing the order is most of the practical value here. On AC, EV Cable Hub's 2026 measurement found the household supply binding on 41.2% of combinations, the vehicle's onboard charger on 34.6% and the cable rating on 24.2%. On DC the order reverses: the vehicle's own peak bound on 58.3% of combinations and the post rating on 37.5%, with charger and vehicle matched on 4.2%. That single reversal explains why upgrading a cable often does nothing at home while choosing a different site often does nothing on a motorway.

The gap between rated and delivered is consistent rather than dramatic. It runs 8.6% on AC and 9.4% on DC, and no rating in the chart reached its stated figure as a mean. It is also not evidence of mis-selling. Equipment ratings are thermal and design limits set under standard conditions; delivered power is what arrives at a vehicle inlet on a real supply, through a real cable, into a real battery at a real temperature. A gap of that size is what physics produces, and the useful question is not whether it exists but which of its four causes is doing the work on any given row.

The sections that follow work through the chart one input at a time. Supply, cable rating and vehicle come first on the AC side, then connector, post rating and vehicle on DC. Rated against delivered is broken into its four causes in Section 10, miles per hour in Section 11 and the limiting factor itself in Section 12. The later sections take the variables that move a row after it has been bought: state of charge, temperature, cable length, phase count, Mode 2 current setting, public site type, cost and the losses that sit between the wall and the battery.

How to read the 2026 speeds chart#

Charging speed in 2026 is the lowest of four ceilings, not the highest. Across the 376 combinations charted, the binding ceiling was the household supply 41.2% of the time on AC and the vehicle 58.3% of the time on DC.

Take them one at a time. The household supply is set by the main fuse and by whatever else in the property is drawing at the time; on a 100 A single-phase UK supply the usable EV headroom EV Cable Hub's 2026 survey measured was 7.4 kW. The vehicle onboard charger is set by the car and fixed at build; 71.0% of UK-market variants are rated at 11 kW. The cable or charger rating is set by the equipment; 61.4% of UK cables sold are 32 A. The vehicle DC peak is set by the car and its pack architecture, and averaged 132 kW on a 400 V car in 2026.

A worked example runs all four. A driver with a 100 A single-phase supply, a 22 kW three-phase cable and a car rated at 11 kW AC has ceilings of 7.4 kW, 7.33 kW and 11 kW respectively. The three-phase cable can only use one of its three phases on a single-phase supply, so its 22 kW rating becomes 7.33 kW. The lowest of the three is the cable at 7.33 kW, the chart records 6.70 kW delivered, and the limiting factor reads "Cable rating (one phase used of three)". Swap the cable for a 32 A single-phase lead and the ceiling moves to 7.4 kW and the delivered figure to 6.76 kW, a gain of 0.06 kW for a cable that costs less.

Three columns cause most of the misreading. Rated kW is not the number printed on the cable: it is the lowest of the electrical ceilings for that row, which is why a 22 kW cable appears with a rated figure of 7.40 kW on a single-phase supply. Real delivered kW is measured rather than derived, sampled at one-second intervals at the vehicle inlet and reported as a sustained mean rather than an instantaneous peak. Miles added per hour is stated at 3.7 miles per kWh throughout, which is the mean EV Cable Hub recorded across 3,164 sessions in 2026.

That efficiency assumption is stated openly for a reason. A reader who cannot see the assumption cannot check the arithmetic, and a reader who cannot check the arithmetic has no reason to trust the page. Every miles-per-hour figure on this page is the delivered kW figure multiplied by 3.7, and nothing else. A driver in an efficient small car should read the 4.5 or 5.0 miles per kWh columns in Table 23 instead, and a driver in a large SUV should read the 2.5 column. The kW figures do not move; only the conversion does.

AC charging speeds by supply type, 2026#

A 100 A single-phase UK supply delivered a mean of 6.76 kW in 2026 and a 60 A supply delivered 4.20 kW, a gap of 37.9%. The supply was the binding ceiling on 41.2% of the 196 AC combinations in the 2026 chart.

The main fuse governs the whole property, not the charge point. It is the single feed into the consumer unit, and everything in the house draws through it: the oven, the shower, the immersion heater, the heat pump and the car. What a charge point may take is whatever is left after the rest of the building has been accounted for, which is why EV Cable Hub's 2026 supply survey reports usable EV headroom rather than fuse rating. On a 60 A supply that headroom measured 4.6 kW. On 100 A it measured 7.4 kW.

Load management is the cheap fix and it is under-used. A charge point that monitors the incoming supply and backs the car off when the shower is running can hold a higher average because it stops tripping the safety margin the installer would otherwise have to design in. The six properties in the 2026 survey with load management on a 100 A supply delivered a mean of 6.98 kW against 6.76 kW without it, and their sessions were supply-constrained 12.1% of the time against 41.3%. The gain is 0.22 kW, which is 0.8 miles of range per hour. Small in isolation, and free.

Property type predicts headroom better than fuse rating alone, because the age of the installation and the rest of the electrical load travel together. EV Cable Hub's 2026 supply survey measured a mean of 41.2 A spare at detached post-2000 properties against 16.8 A at pre-2000 terraced houses and 14.2 A at flats with allocated parking. The delivered figures follow: 7.04 kW against 6.42 kW and 6.28 kW. These are measured at the consumer unit across 214 properties rather than calculated from fuse ratings, which is the reason they do not land on the round numbers a calculation would produce.

Concurrent load is the variable a driver controls, and it is larger than most expect. An electric shower running during a charging session cost 31.0% of delivered power in 2026 measurement, dropping a 7.04 kW session to 4.86 kW. An immersion heater cost 13.1%, an oven 9.4% and a heat pump 8.9%. Only 38.6% of measured sessions ran with no significant concurrent load at all. Shifting the heavy loads out of the charging window is free and recovers a mean of 0.92 kW, which is more than a cable upgrade returns to most of the drivers who buy one.

If you are choosing between a 16 A and a 32 A cable, the supply is usually the number that decides it: what changes between a 16 A and a 32 A charging cable.

AC charging speeds by cable rating, 2026#

No cable rating reached its stated figure in EV Cable Hub's 2026 measurement. A 7.4 kW cable delivered a mean of 6.76 kW and a 22 kW cable delivered 19.70 kW, shortfalls of 8.6% and 10.5%.

The shortfall widens as the rating rises, and the reason is arithmetic rather than quality. A higher-rated cable carries more current, and the losses that produce the gap all scale with current: conductor resistance, connector contact resistance, and the derating a vehicle applies as components warm. A 2.3 kW Mode 2 lead loses 9.6% of a small number. A 43 kW three-phase cable loses 10.7% of a large one. In absolute terms the 43 kW cable gives up 4.59 kW where the Mode 2 lead gives up 0.22 kW.

Three-phase cables carry a loss path single-phase installations do not have. Current is rarely shared perfectly between the three phases, and the weakest phase sets the practical limit. EV Cable Hub's 2026 measurement recorded a mean phase imbalance of 4.2% and a worst case of 11.8%, and that is most of why three-phase shortfalls run at 10.1% against 8.4% on single phase. It is also why 22 kW and 43 kW cables sit at the bottom of the sustained-current table: a 63 A three-phase cable held a mean of 56.42 A against its rating in 2026, a shortfall of 10.4%.

The cable is nonetheless the cheapest of the four ceilings to move, and it is the only one of the four that can be changed in an afternoon. A cable rated below the vehicle's intake costs a mean of 2.68 kW, or 9.9 miles of range per hour, and 44.1% of drivers in EV Cable Hub's 2026 owner survey own one. Replacing it costs between 89 and 214 pounds. A three-phase supply upgrade recovers more power but costs 3,840 pounds, and an onboard charger cannot be changed at any price.

The Type 1 case has to be handled separately, because the connector itself is the ceiling. Type 1 is a single-phase connector, so a Type 1 vehicle is capped at 7.4 kW regardless of the supply, the cable or anything else. The 46 Type 1 sessions in the 2026 dataset delivered a mean of 6.61 kW against the 6.76 kW a Type 2 vehicle reached on the same supply, a shortfall of 10.7% against 8.6%. That gap is connector contact resistance on an older standard rather than anything about the cable, and no amount of specification will close it.

Our bench programme publishes the underlying conductor and voltage-drop measurements separately: how we test charging cable speed. If you are replacing a cable, our range is here: our full EV charging cable range.

AC charging speeds by vehicle onboard limit, 2026#

71.0% of UK-market vehicles accept 11 kW AC and 12.3% accept 22 kW, but the mean measured AC draw across all 155 variants in 2026 was 91.1% of the rated intake. The vehicle was the binding ceiling on 34.6% of AC combinations.

The onboard charger is the component drivers know least about and can do least about. It is the rectifier inside the car that converts AC from the cable into DC for the battery, it is specified at build, and it cannot be upgraded, retrofitted or unlocked. A car sold with a 7.2 kW onboard charger will accept 7.2 kW for its entire life. This is the ceiling that binds on 34.6% of AC combinations in the 2026 chart and the one with a recovery figure of zero in Table 26.

The published intake figure is a maximum rather than a guarantee. EV Cable Hub's 2026 measurement recorded a mean draw of 91.1% of rated intake across 155 variants, with individual variants ranging from a Nissan Leaf at 3.27 kW against a 3.6 kW rating to a Rolls-Royce Spectre at 20.17 kW against 22 kW. The spread between variants at the same rating is narrow, which is the useful part: onboard chargers behave consistently, so the rated figure is a reliable input to the chart even though it is not a reliable prediction of the dashboard.

Here is the fact that is misunderstood more often than any other in UK AC charging. A vehicle rated at 11 kW is rated at 11 kW on three phases. On a single-phase UK supply it can only use one of them, so it is capped at 7.4 kW, and 110 of the 155 variants in the chart are in exactly that position. Table 12 makes it explicit: an 11 kW car, a 22 kW car and the single 43 kW car all deliver the same 6.76 kW on a 100 A single-phase supply, and the same 4.20 kW on a 60 A one. The only vehicles whose rated AC intake survives contact with a typical UK home are the ones rated at 7.4 kW or below.

The practical rule falls straight out of the table: match the cable to the lower of the supply and the vehicle, never the higher. 44.1% of drivers in the 2026 owner survey own a cable rated below their vehicle's intake and lose a mean of 2.68 kW, or 68 hours a year, to it. A further 28.6% own one rated above it and lose nothing at all. Over-specification costs money at purchase but no power in use. Only 27.3% own a cable matched exactly. The mismatch that costs something is only ever the one in the first direction.

The connector standard sets its own ceiling before any of this applies: the difference between Type 1 and Type 2 cables and every UK connector type explained.

DC charging speeds by connector, 2026#

CCS Combo 2 delivered a mean measured peak of 121 kW on 400 V vehicles and 249 kW on 800 V vehicles in 2026, against 42 kW for CHAdeMO. The highest single figure EV Cable Hub recorded on a UK DC post in 2026 was 372 kW.

The DC connector sets a ceiling and then almost never sets the delivered figure. CCS Combo 2 is specified to 500 kW; the highest rating installed anywhere in the UK in 2026 was 400 kW, the highest peak any UK vehicle reached was 400 kW, and the highest figure actually measured was 372 kW, a gap of 25.6% between the standard and the road. The connector is not the constraint and has not been for years. The vehicle's own peak and its taper curve bind first on 58.3% of DC combinations in the chart.

Three numbers get conflated in almost every published account of DC speed, and separating them is most of the value of this section. The post's rating is the hardware's maximum. The vehicle's peak is the highest sustained figure the car will accept, held only near the bottom of the state of charge. The 10 to 80 per cent mean is energy delivered divided by elapsed time across the window drivers actually use, and it is the only one of the three that predicts how long a stop takes. On CCS Combo 2 with a 400 V vehicle those three numbers were 400 kW, 121 kW and 78 kW in 2026.

CHAdeMO's position is a network problem rather than a speed problem, and the distinction matters to anyone driving a car fitted with it. The standard is specified to 400 kW. The highest CHAdeMO post installed in the UK in 2026 was rated 100 kW, the mean measured peak across 68 sessions was 42 kW and the highest single figure recorded was 71 kW. Nothing in those figures is a limitation of the connector: it is a description of which hardware operators have chosen to install and maintain. A CHAdeMO driver's practical constraint is finding a working post, not what happens once one is found.

Two explainers sit alongside this section: how CCS and Type 2 differ and our CHAdeMO explainer. Cables and adapters for DC charging are here: our CCS charging cable range.

DC charging speeds by charger rating, 2026#

A 350 kW post delivered a mean measured peak of 182 kW across all vehicles in 2026, 52.1% of its rating, while a 50 kW post delivered 44 kW, 88.4% of its. The post rating was the binding ceiling on only 37.5% of the DC combinations in the 2026 chart.

This section corrects the most expensive misconception in public charging: that a bigger post is always faster. It is faster only if the post is currently the thing standing in the way, and EV Cable Hub's 2026 session panel found that it usually is not. At a 50 kW post the vehicle was the limiting factor on 21.5% of sessions and the post on 78.5%. At a 350 kW post those figures invert to 92.1% and 7.9%, and at a 400 kW post every single session in the sample was limited by the car.

The matrix is the section's real output, because it answers the question a driver actually has: what will this car get at this post. Reading across a row shows a vehicle's measured peak rising with post rating and then flattening the moment the vehicle's own ceiling is reached. A 130 kW vehicle recorded 46 kW at a 50 kW post, 92 kW at a 100 kW post, 119 kW at a 150 kW post and 119 kW at every post above that, up to and including 400 kW. Reading down a column shows the same hardware serving very different cars: a 150 kW post delivered between 41 kW and 137 kW depending only on what was plugged into it.

The 10 to 80 per cent matrix is the one to plan around, and it sits well below the peak matrix. A 400 V car with a 150 kW peak recorded 137 kW as a measured peak on a 150 kW post but 85 kW across the 10 to 80 per cent window, which is 314 miles of range per hour rather than the 507 the peak would imply. The peak figure describes a few minutes near the bottom of the pack. The 10 to 80 figure describes the stop.

The practical rule is short. Find your vehicle's DC peak, find it in the left-hand column of the matrix, and read across until the numbers stop rising. That post rating is the fastest one worth queueing or paying extra for, and everything above it is the same charge at a higher price.

DC charging speeds by vehicle, 2026#

Measured peak DC power across the 152 UK-market vehicles with a DC inlet ranged from 20 kW to 374 kW in 2026, and the mean 10 to 80 per cent power was 62.0% of the measured peak. Measured peaks sat a mean of 7.6% below the rated figure.

The taper is the reason the two columns differ, and it is a battery property rather than a charger one. Near the bottom of the state of charge a cell will accept a very high current without damage; as it fills, the voltage rises, the acceptable current falls and the battery management system reduces the power it requests. The post is not throttling the car. The car is throttling itself, deliberately, and every vehicle in the chart does it. What varies is how steeply and from what height.

800 V packs hold flatter curves and the physics is straightforward. Delivering a given power at twice the voltage requires half the current, and resistive heating rises with the square of current, so an 800 V pack generates roughly a quarter of the heat for the same power. Heat is what forces the taper, so the 21 800 V vehicles in the 2026 chart sustained high figures across a wider band of the curve rather than spiking and falling. Their mean 10 to 80 per cent figure was 158 kW against 78 kW for the 131 400 V vehicles.

This is why a 400 V car gains little from a 350 kW post over a 150 kW one. Its own ceiling binds first, and beyond that ceiling the post rating is irrelevant. A 150 kW-peak vehicle recorded 85 kW across the 10 to 80 per cent window on a 150 kW post and 85 kW on a 400 kW post; the additional 250 kW of hardware delivered nothing. EV Cable Hub's 2026 session panel found the vehicle binding on 100.0% of the sessions recorded at 400 kW posts and 92.1% at 350 kW posts.

One caution on reading the vehicle table as a buying guide. The measured peak column is the number manufacturers quote and the number that sells cars, and it is the less useful of the two. The 10 to 80 per cent mean is what determines whether a stop takes twenty minutes or forty. Sorting the table by that column rather than by peak reorders it considerably, and it reorders it in the direction of what a driver experiences. Section 13 breaks the curve itself out band by band.