EV efficiency — kWh/100 mi, mi/kWh, Wh/km, MPGe explained

Internal-combustion cars use fuel economy (**mpg**, **L/100 km**) because the fuel itself is the cost. EVs charge by kWh, so efficiency is expressed three ways depending on region and convention. All three describe the same physical quantity: energy consumed per unit distance.

The exact conversions

**mi/kWh** × 33.705 = **MPGe** > **kWh/100 mi** = 100 / (mi/kWh) > **Wh/mi** = 1000 / (mi/kWh) > **Wh/km** = Wh/mi ÷ 1.609344 > **kWh/100 km** = Wh/km ÷ 10 > **MPGe** = miles per gallon of gasoline energy-equivalent. The EPA's 33.705 kWh/gal comes from gasoline's higher heating value (HHV, 125,000 BTU/gal) — a controversial choice because it credits EVs for combustion energy they never had to reject as heat.

Reference values (EPA and WLTP)

| Vehicle (2024 MY) | EPA Wh/mi | mi/kWh | MPGe | WLTP Wh/km | Battery | |---|---|---|---|---|---| | Tesla Model 3 LR RWD | 232 | 4.31 | 145 | 138 | 79 kWh | | Tesla Model Y LR AWD | 267 | 3.75 | 126 | 165 | 79 kWh | | Hyundai Ioniq 6 RWD LR | 217 | 4.61 | 155 | 137 | 77.4 kWh | | Lucid Air Grand Touring | 258 | 3.88 | 131 | 168 | 118 kWh | | Ford F-150 Lightning ER | 500 | 2.00 | 68 | 320 | 131 kWh | | Rivian R1T Large | 476 | 2.10 | 71 | 300 | 141 kWh | | Porsche Taycan 4S (2024) | 331 | 3.02 | 102 | 200 | 89 kWh | | Nissan Leaf 40 kWh | 340 | 2.94 | 99 | 209 | 40 kWh |

EPA Wh/mi already includes charging losses (SAE J1634 'multi-cycle test' from wall to wheels), which typically add 10–15% over 'battery-out' consumption. WLTP measures battery-out only — that's why WLTP km/kWh looks better than EPA mi/kWh for the same car.

Charge losses — the number nobody mentions

AC Level 2 charging (240 V, 7–11 kW): 88–92% efficient wall-to-battery > DC fast charge (50–350 kW): 92–95% efficient (skips onboard rectifier) > Cold DC charge (< 0 °C): 70–85% (battery heater draws from grid) > Vampire drain (parked, awake modes on): 1–3% per day > Total 'window sticker' includes AC losses; a full 100 → 80% road trip DC session is more efficient per kWh billed.

Why winter kills range — the physics

Three effects compound in cold weather. Range loss vs mild (21 °C) baseline, from AAA and Recurrent 2023 fleet data:

| Cause | Range loss | Why | |---|---|---| | Cabin heating (resistive) | 15–25% | PTC heater draws 3–7 kW. On a 30-min commute at 60 mph that's 1.5–3.5 kWh — 8–18% of a 20 kWh commute budget. | | Heat pump (if fitted) | 8–15% | COP falls from ~3.0 at 10 °C to ~1.5 at −15 °C. | | Battery heater | 5–10% | Li-ion cells lose 20–30% usable capacity below 0 °C; TMS pre-warms to 10–25 °C. | | Higher rolling resistance (winter tires, cold rubber) | 3–7% | Cₙ increases from ~0.008 to ~0.011. | | Denser air | 2–4% | Aerodynamic drag ∝ ρ_air, which rises 12% from 20 °C to −20 °C. | | **Net at −10 °C, highway** | **~35–40%** | AAA measured 279-mile car → 173 miles. |

Heat pumps (standard on Tesla Model Y since 2021, Ioniq 5/6, ID.4) recover most of the cabin-heating penalty above −10 °C. Below that they revert to resistive and the penalty returns.

Speed and drag — the cubic law

Aerodynamic drag power: **P_drag = ½ ρ Cd A v³**. Doubling speed multiplies drag *power* by eight. At steady cruise:

30 mph (48 km/h): drag ≈ 3 kW → 100 Wh/mi > 55 mph (89 km/h): drag ≈ 12 kW → 218 Wh/mi > 75 mph (121 km/h): drag ≈ 27 kW → 360 Wh/mi > 85 mph (137 km/h): drag ≈ 40 kW → 470 Wh/mi

Slowing from 80 mph to 65 mph on a highway trip typically saves 20–25% of consumed energy. Nothing else the driver can do comes close.

Worked example — planning a 300-mile trip

**Vehicle:** Ioniq 5 SE RWD, 77.4 kWh usable, EPA 260 mi (298 Wh/mi rated). **Conditions:** 5 °C morning, 70 mph highway, 200 mi one-way + 100 mi return.

Baseline consumption at 70 mph (physics-scaled from EPA 65 mph rating): 300 Wh/mi × (70/65)³ × 0.85 [drag-share] + 300 × 0.15 [rolling+aux] = 296 + 45 = **341 Wh/mi** > Cold weather adjustment (+15% for 5 °C with heat pump): 341 × 1.15 = **392 Wh/mi** > 200 mi × 392 Wh/mi = 78.4 kWh — more than usable capacity, needs one DC stop. > Plan: leave 100% → arrive at charger with 15% (80% used = 61.9 kWh, ~158 mi). Charge to 80% (26 min at 235 kW peak, tapering). Finish leg on remaining 65% = ~130 usable miles. ✓

**Rule of thumb:** for winter highway trips, size charging stops to arrive with 15% and leave with 80%. That's where DC fast charging is fastest (below 10% the BMS taper limits current; above 80% the cell voltage taper collapses power).

MPGe — why it exaggerates EV efficiency

EPA's MPGe treats 33.705 kWh of grid electricity as equivalent to 1 gallon of gasoline. Problem: the grid loses 40–60% of primary energy at the power plant plus 6–8% in transmission. Well-to-wheel, a 132-MPGe Tesla in a coal-heavy grid delivers roughly the same CO₂/mile as a 45-MPG Prius. In renewable-heavy grids (California, Norway, Quebec) MPGe is closer to a fair comparison. This is why the EU switched to **Wh/km battery-out** — an honest measurement that leaves grid mix to a separate calculation.

Regen and downhill

Modern EVs recover **60–75%** of the kinetic energy dumped into regen (round-trip battery-in / battery-out), limited by: - Motor inverter current cap (typically 200–350 kW peak absorption) - Battery C-rate limit (a 77 kWh pack absorbs 100 kW comfortably, 200 kW briefly) - Cold battery (< 5 °C): regen is disabled entirely until warmed — Tesla's yellow dashed line, Ioniq's reduced-regen indicator. Regen does not violate conservation of energy — it just means brake heat becomes battery heat. Over a mountain descent, indicated range can *rise*.

Battery degradation

Recurrent Auto's 2024 dataset (15,000 EVs) shows median capacity retention: > Year 1: 98% · Year 3: 94% · Year 5: 90% · Year 8: 85% DC-fast-charging exclusively (Tesloop taxi fleet data) accelerates loss to ~15% at 200,000 mi. Home L2 charging with 20–80% daily cycles is the least aggressive strategy. All modern EVs come with an 8-year / 100,000 mi battery warranty that guarantees ≥70% capacity retention (US federal minimum, 40 CFR §86.1815).

The five-second recap

mi/kWh × 33.705 = MPGe. EPA already includes charge losses; WLTP doesn't. Cold weather takes 25–40%. Speed hurts cubically. Regen recovers ~70%. Battery loses ~15% over 8 years.