Real-world e-scooter range: an energy-budget model (P_drag + P_roll + P_grade + P_accel), derating from payload / wind / temperature / altitude / tire pressure / speed, and how to convert Wh into kilometres

Scootify 14 min read

A manufacturer’s stated e-scooter range is a number obtained under laboratory conditions that almost never reproduce in real life: a 65-kg dummy rider, smooth pavement free of potholes and tracks, calm weather with no wind, +20 °C ambient temperature, tires at nominal-maximum pressure, eco-mode at 15–18 km/h, no stops and no accelerations, a fully charged brand-new battery. On a real commute you have an 80–90-kg body plus a backpack, variegated road surface (asphalt + cobblestones + expansion joints), 3–8 m/s wind, 0–25 °C temperature, pressure 20 % below nominal after a week without a pump, sport-mode at 25–30 km/h, 8–15 stops on a 5-km route, and a battery 200–400 cycles old at SoH 85–90 %. The combined derating from ideal to real is 20–60 % of range, and that gap is the origin of «range anxiety» — the fear of not making it back home that forces riders to carry chargers or to artificially cap routes at 30 % of nameplate.

This article provides a formal energy-budget model that lets you estimate real-world range from explicit parameters rather than hope. We unify components that already exist as standalone articles — aerodynamic drag and wind effects, grade-power for hills, payload and Wh/km dependence, tire pressure and rolling resistance, temperature effect on the battery, regenerative braking and its efficiency, winter operation and hot-weather operation — into a single quantitative model with a worked example. This is not an academic exercise: the numbers are consistent with empirical data from manufacturers and the research literature, and they let you answer «how far will I actually go on this route in this weather» with ±10–15 % accuracy.

1. The power equation: foundation of the model

The energy a scooter rider draws from the battery over every metre of travel goes into four distinct physical processes, each governed by its own law. The canonical total-power formulation from Wilson «Bicycling Science» 4th ed. MIT Press and Martin et al. 1998 Journal of Applied Biomechanics 14(3):276–291:

P_total = P_drag + P_roll + P_grade + P_accel

where:

  • P_drag = ½ × ρ × v_air³ × C_d × A (W) — cubic in air-speed relative to the rider
  • P_roll = C_rr × m × g × v_ground (W) — linear in mass and ground speed
  • P_grade = m × g × sin(θ) × v_ground (W) — linear in mass and grade, sinusoidal in angle
  • P_accel = m × a × v_ground (W) — instantaneous acceleration power; integrated this is the kinetic energy of start-stop cycles

Parameters:

  • ρ — air density (~1.225 kg/m³ at ISA sea level; falls with altitude, rises with cold)
  • v_air = v_ground − v_wind — vector difference between scooter ground speed and wind speed along the direction of travel (sign «−» for headwind, «+» for tailwind)
  • C_d × A — drag area (m²); for an upright scooter rider ~0.55–0.70 m² per Wilson MIT Press
  • C_rr — rolling-resistance coefficient (dimensionless); 0.008–0.015 for pneumatic e-scooter tires, 0.020–0.035 for solid — detailed below
  • m — total mass (rider + scooter + cargo), kg
  • g — gravitational acceleration (9.81 m/s²)
  • θ — grade angle (5 % gradient → sin(θ) ≈ 0.050; 10 % → ≈ 0.100)
  • a — instantaneous acceleration, m/s²

Energy per unit distance (Wh/km) is the time-integrated power divided by distance:

E_per_km = (P_total × t) / d = P_total / v_ground × (1 h / 3600 s) × (1000 m/km) = P_total / (3.6 × v_ground)

with v_ground in km/h and the result in Wh/km. From the other side — the battery Wh balance:

Range_km = (E_battery_usable_Wh × η_drivetrain) / E_per_km

where E_battery_usable_Wh is usable capacity (not nameplate — the BMS reserves 5–15 % SoC) and η_drivetrain is the combined drivetrain efficiency (motor × controller × battery internal-resistance loss), typically 0.55–0.75 across the full chain.

2. P_drag — aerodynamic resistance (short summary)

Drag is the largest and fastest-growing component at speeds above 20 km/h. It scales cubically with v_air, which means doubling speed → 8× drag power. The full treatment of CdA, density effects from altitude and temperature, headwind/tailwind asymmetry is in Riding in windy weather. Here are the key facts for the energy budget.

Drag share of total power in a typical commuter scenario:

Ground speedDrag share of P_total (calm air, flat)
10 km/h~10–15 %
15 km/h~25–35 %
20 km/h~40–50 %
25 km/h~50–60 %
30 km/h~60–70 %
40 km/h~75–85 %

This is the fundamental non-linearity through which a 20-km route at 30 km/h burns more Wh than the same route at 20 km/h, even though higher speed shortens trip time. Energetically, riding slower is always cheaper. Riding fast is a paid luxury on a cubically growing Wh balance.

Headwind-grade equivalence: 5 m/s headwind at 25 km/h ground speed adds about the same power as a ~2 % gradient in calm air. A tailwind, conversely, saves ~10–25 % Wh/km, but regenerative braking does not recover it: tailwind simply removes drag-resistance, it does not pump energy into the pack (see Regenerative braking for the asymmetry).

3. P_roll — rolling resistance (full treatment)

Rolling resistance is linear in mass and speed, which means at low speeds (up to ~15 km/h) it is the dominant component of total power on flat ground. Unlike drag, P_roll does not scale cubically — slowing down saves little rolling, but radically saves drag, so net energy gains grow as you slow to 8–15 km/h, beyond which rolling starts to dominate and further deceleration is unprofitable.

The Crr coefficient for e-scooter tires varies widely with construction (Cambridge University Press / Design Society 2024 «Comparison of e-scooter tyre performance using rolling resistance trailer»; Bicycle Rolling Resistance database as cross-reference; Wilson «Bicycling Science» MIT Press for inflated-bike baseline values):

E-scooter tire typeC_rr (typical range)Comment
Pneumatic road tire (50–65 PSI, thin tread)0.008–0.012Best grip/Crr balance; standard on performance models (Apollo Phantom, Dualtron Spider)
Pneumatic urban tire (40–50 PSI, mid-tread)0.011–0.015Baseline for commuter models (Xiaomi 4 Pro, NAVEE)
Pneumatic knobby/off-road0.015–0.025For bumpy/gravel — worse on asphalt
Foam-filled (puncture-proof tubeless foam)0.020–0.028+50–80 % Crr vs pneumatic; mass higher by 0.5–1.2 kg
Solid honeycomb (Tannus, Whatcha, Segway-Max solid retrofits)0.022–0.035Worst Crr; deformation dissipates as hysteresis losses
Solid full-rubber (legacy Ninebot ES2, budget folding)0.030–0.050Worst option; +200–400 % Wh/km vs pneumatic on flat

For comparison, road bike — 0.003–0.006; commuter bike — 0.005–0.010; wide MTB — 0.010–0.020. E-scooter tires are always worse than bicycle tires because of the small diameter (8–11″ vs 26–28″ on a bike): the smaller the diameter, the larger the share of sidewall deformation relative to contact patch, and the larger the hysteresis losses (Wikipedia — Rolling resistance section «Wheel size»; Bicycle Rolling Resistance — Tire Width Aspect Ratio).

Tire pressure is the largest practical Crr lever short of swapping the tire itself. A pneumatic tire at 80 % nominal pressure has +20–30 % Crr; at 60 % — +40–60 %; at 40 % — +80–120 % and risk of pinch flat (Hiboy, How Does Tire Pressure Affect Your Electric Scooter Range). That means a week without a pressure check costs 5–15 % of range simply from natural leakage (~2–5 PSI/week is a normal rate for a tubed pneumatic, not a defect).

Surface is the second major factor. Crr on standard asphalt 0.012 → on wet asphalt 0.014 → on cobblestone 0.025–0.040 → on gravel 0.040–0.070 → on wet leaves or painted lane markings 0.020–0.030 plus a slip risk. Surface as a contact-physics axis is treated separately in Riding on difficult road surfaces.

Mass is linear: doubling m doubles P_roll. This explains why adding a 10-kg backpack raises Wh/km by 5–8 % — primarily the rolling component (drag does not change with mass; grade scales with mass like rolling).

4. P_grade — gravitational resistance (short summary)

A climb is the most predictable line item. On flat ground θ = 0, sin(θ) = 0, P_grade = 0. On a 5 % gradient (sin(θ) ≈ 0.050) at m = 95 kg and v_ground = 6.94 m/s (25 km/h):

P_grade = 95 × 9.81 × 0.050 × 6.94 = 323 W

For a commuter whose grade-free total power is ~250 W, a 5 % climb doubles total power. A 10 % climb triples it. This is why climbing and gradeability is a separate engineering axis with its own constraints (motor thermal limit, controller current limit, battery sag).

A descent is a negative contribution: at −5 % grade the same formula gives P_grade = −323 W, i.e. gravity returns energy. Part of this comes back via regenerative braking, but typically only 5–15 % through η_regen ≈ 60–80 % × η_battery_charge_acceptance ≈ 80–90 % × η_controller ≈ 90–95 %, plus the scooter often dissipates deceleration in drag/friction rather than regen. A round trip (up and back on the same route) loses ~85–95 % of grade energy to heat, which is why hilly terrain is always more expensive on a loop than flat terrain.

Practical rule of thumb: each +1 % grade at 25 km/h costs ~+5–7 % Wh/km. A 5-km route with 50 m of climb (1 % average gradient) is ~+5 % energy vs flat; with 250 m of climb (5 % average) — +25–35 % energy.

5. P_accel — inertial resistance and the start-stop penalty

The kinetic energy of acceleration E_kin = ½ × m × v² is spent on every acceleration and lost on every braking event (minus regen). For m = 95 kg, v = 25 km/h (6.94 m/s):

E_kin = 0.5 × 95 × 6.94² = 2289 J = 0.636 Wh

That sounds small, but a typical urban route has 8–15 full start-stop cycles per 5 km (traffic lights, pedestrians, turns, obstacles). Over 10 cycles that is 6.36 Wh dumped into heat at brake pads + controller. For a 500-Wh battery that is 1.3 % of capacity. Over a 20-km route with 40 cycles — 5.2 % of capacity, or 1 km less range on a 500-Wh pack at a typical 15 Wh/km.

With 10 % regen efficiency it becomes 0.9 % instead of 1.3 % — a noticeable but not dramatic saving. Aggressive throttle behaviour (sport-mode hard acceleration) raises this to 8–12 % loss in dense urban use, because fast accelerations work outside the motor’s efficiency band (peak torque corresponds to low η_motor).

Practical rule: an urban route with 8+ stops/5 km consumes 10–15 % more Wh/km than the same route at coasting cruise speed without stops. This is why manufacturers test on steady-state cycles (NEDC, WLTC, EN 17128 cycle) and the real user gets worse numbers.

6. Drivetrain efficiency: why 500 Wh in the battery becomes only 350 Wh at the wheel

The power equation describes mechanical power at the wheel — what is needed to overcome drag + roll + grade + accel. Between battery and wheel sit three conversion stages, each with its own efficiency:

  1. η_battery — discharge efficiency (including I²R losses on internal resistance). At low-current discharge — 95–98 %; at high current (peak acceleration) — 85–92 % from I²R. A cold battery (−10 °C) — 60–75 %; hot (40 °C+) — 90–95 %.
  2. η_controller — MOSFET half-bridge switching efficiency plus MCU overhead. A typical brushless DC controller for an e-scooter — 90–95 % on cruise, 85–92 % on peak.
  3. η_motor — electromagnetic conversion plus bearing/seal friction. A BLDC hub motor for e-scooters — 75–88 % peak efficiency at nominal RPM; falls to 50–65 % at low RPM (launch) and 60–75 % beyond design RPM.

Combined η_drivetrain = η_battery × η_controller × η_motor ≈ 0.55–0.75 for typical use. That means with a 500-Wh nameplate the wheel receives 275–375 Wh — the rest is dissipated as heat in battery internal resistance, controller MOSFETs and motor windings/bearings. This is why a scooter is warm after riding — close to 30–45 % of all consumed energy becomes heat.

Separately — BMS reserve: the nameplate Wh is nominal capacity (cell × series × Ah × V), and usable Wh is the capacity between BMS cutoff voltages. Most BMS reserve 5–15 % SoC at the low end (defensive low-voltage cutoff) and 0–5 % at the top (do not allow charging to 4.2 V per cell, limit to 4.15 V for cycle life). So usable Wh ≈ 0.80–0.95 × nameplate Wh in a new scooter.

Plus battery aging: after 200–400 charge/discharge cycles SoH (State of Health) is typically 85–90 %; after 500 — 75–85 % for quality Li-ion (details in battery-engineering-lithium-ion-bms-thermal-runaway and battery-lifecycle-recycling-engineering). After 800–1000 cycles — 60–75 %, which is end-of-life for many commuter models on the range criterion.

General usable-Wh formula for a specific scooter:

E_usable = E_nameplate × (1 − BMS_reserve) × SoH × η_drivetrain

Example: Xiaomi 4 Pro nameplate 446 Wh, BMS reserve 10 %, SoH ≈ 88 % after 300 cycles, η_drivetrain ≈ 0.65 in commuter mode:

E_usable_at_wheel = 446 × 0.90 × 0.88 × 0.65 = 230 Wh at wheel

That is 51 % of nameplate — and it is normal, not a defect. This is the fundamental physics that explains why a 446-Wh battery + 15 Wh/km mechanical losses → ~30 km real-world range instead of an advertised «up to 45 km».

7. Derating from payload (rider + cargo mass)

Payload affects P_roll and P_grade linearly, does not affect P_drag (provided no extra frontal area outside the body, such as backpack side-bulges), and affects P_accel linearly through m × v²/2 kinetic energy.

Approximate Wh/km sensitivity from carrying-cargo-and-payload and Ride1Up data:

Δ payloadΔ Wh/km (typical commuter)
+5 kg+2–4 %
+10 kg+5–8 %
+20 kg+10–16 %
+30 kg+15–24 %
+50 kg (max payload)+25–40 %

Why not proportional: drag is 40–60 % of total power → payload affects only the 60 % remainder (rolling + grade + accel). So +50 % m (from 80 to 120 kg) → +30 % rolling/grade/accel → +18 % total Wh/km.

Backpack aerodynamic penalty: a large backpack adds 0.05–0.15 m² to effective Cd·A → +10–20 % drag. That is a separate line item beyond the mass-derate.

8. Derating from temperature

The axis most felt by users, because range drops dramatically in winter. The hit comes from two mechanisms:

(a) Cold battery capacity loss — Li-ion (NCM, LFP, NCA chemistries) loses anywhere from 10 to 50 % of usable capacity at low temperatures because electrolyte viscosity rises and lithium-ion intercalation into the electrode lattice slows down (Battery University — BU-502 Discharging at High and Low Temperatures, Lithium-Ion Batteries under Low-Temperature Environment: Challenges and Prospects, NCBI PMC9698970, 2022):

Temperature (cell)Usable capacity (Li-ion NCM, baseline 20 °C)
+40 °C~100 % (slight high-T bump, but cycle life suffers)
+20 °C100 % (reference)
+10 °C~95 %
0 °C~75–85 % (−15–25 %)
−10 °C~65–75 % (−25–35 %)
−20 °C~50–60 % (−40–50 %)
−30 °C<40 % (−60 % and worse; BMS may shut down)

LFP (LiFePO₄) is more tolerant of high temperatures but more sensitive to cold; cold derating ~5–10 % worse. NCM/NCA — the standard chemistry for most e-scooters — the numbers above apply.

(b) Air-density change — at +20 °C ρ_air = 1.204 kg/m³; at 0 °C — 1.293 kg/m³ (+7.4 %); at −20 °C — 1.395 kg/m³ (+15.9 %). Drag scales with ρ, so winter adds +7–16 % drag power purely from denser cold air. This is separate from the battery loss and stacks on top of it.

(c) Lubricant viscosity and grease drag in bearings/motor rise at cold temperatures, but this contribution is typically small (1–3 %).

Net winter range vs summer baseline:

  • Summer (25 °C, calm) — 100 % (reference)
  • Cool spring/autumn (10 °C) — 90–95 %
  • Cold autumn (0 °C) — 70–80 %
  • Snowless winter (−10 °C) — 55–65 %
  • Hard frost (−20 °C) — 35–45 %

This is why users report «in summer I get 30 km, in winter I barely manage 15» — two derate axes stack additively, not multiplicatively. More detail in winter-operation and hot-weather-operation.

Heat is a separate effect: high T (+35 °C and above) lowers battery internal resistance and gives a temporary capacity boost, but cycle life crashes, motor cooling worsens (warm heatsink vs warm air — small ΔT), and a long sustained run can hit thermal throttling.

9. Derating from altitude (height above sea level)

Altitude affects ρ_air through the barometric formula:

ρ(h) = ρ_0 × exp(−h / H)

where H = R × T / (g × M) ≈ 8400 m is the atmospheric scale height (International Standard Atmosphere ISO 2533:1975, NASA NTRS standard atmosphere reference). At 1000 m altitude ρ ≈ 1.225 × exp(−1000/8400) ≈ 1.089 kg/m³ — −11 % vs sea level, hence −11 % drag power at the same ground speed.

On mountain routes (Carpathians — Dragobrat 1300 m, Ai-Petri 1234 m, lower Hoverla station 1500 m) the drag component drops 12–16 %, which offsets part of the extra grade power. This is a non-trivial point: in the mountains drag savings + grade cost partially cancel, and the real altitude effect on range depends on the route profile.

The battery is a separate question. A sealed Li-ion pack is isolated from atmospheric pressure, so the direct effect is zero. But:

  • Convective cooling at altitude is poorer (rarer air → lower heat-transfer coefficient), so under sustained high power the motor/controller may thermal-throttle 5–10 % sooner
  • Atmospheric pressure is lower → tires at fixed gauge pressure carry higher absolute pressure → a fractional Crr improvement of ~0.5–1.5 % (not very material)

Net effect: altitude gives +5–10 % range at altitude vs sea level at the same temperature, mostly through drag savings.

10. Derating from tire pressure

The cheapest and most neglected axis. A pneumatic tire loses 2–5 PSI/week through rubber permeation + valve leakage (this is normal, not a defect). Over 4 weeks without a check, nominal 50 PSI → 35–42 PSI = 70–85 % of nominal.

The Crr dependence on pressure is non-linear: at constant tire load and falling pressure, casing deformation rises exponentially (tire footprint area = F_normal / P_internal). At 80 % nominal — typically +20–30 % Crr; at 60 % — +40–60 %; at 40 % — +80–120 % and a real risk of pinch flat (Schwalbe technical notes; SILCA tire pressure calculator, Bicycle Rolling Resistance — Pressure Effect).

Practical impact: one week without a pump at 50 PSI → 45 PSI (90 % nominal) → +10 % Crr → +4–5 % Wh/km (since rolling is only part of total). Two weeks → 40 PSI (80 %) → +25 % Crr → +10–12 % Wh/km.

Check pressure weekly. The check itself costs 30 seconds with a hand-held gauge (more accurate than a pump-mounted one — testing shows mass-market pump gauges read ±5–10 PSI off; a standalone digital gauge reads ±0.5–1 PSI — a material difference at nominal 50 PSI).

Intuition from tire engineering: the 15 % tire-drop rule of Frank Berto (Rene Herse Cycles) is the optimal compromise between Crr and grip. For an e-scooter this is typically 45–55 PSI front + 50–60 PSI rear for an 80-kg rider. Exact value — in the manufacturer’s manual.

11. Derating from speed (drag’s cubic non-linearity)

As shown in §2, the drag share grows cubically. This means cutting speed from 30 to 20 km/h saves ~40 % Wh/km, and from 30 to 15 — ~60 %. This is the cheapest and fastest lever for extending range and requires no capital investment.

v_groundP_total typical (95 kg, flat, calm)Wh/km
10 km/h~80 W~22
15 km/h~140 W~26
20 km/h~220 W~30
25 km/h~330 W~36
30 km/h~480 W~46
35 km/h~680 W~58
40 km/h~940 W~71

Notice: at 10 km/h Wh/km is higher than at 15 — because the motor runs in a poor-efficiency band and η_drivetrain drops. The optimal eco speed for most commuter scooters is 18–22 km/h: a trade-off between motor efficiency band (peak η near 20–25 km/h on a typical hub motor design) and drag cost.

12. Derating from start-stop (city vs cruise penalty)

As shown in §5, each acceleration-stop cycle for 95 kg + 25 km/h costs 0.636 Wh, of which ~10 % is regenerable, i.e. 0.57 Wh is lost. On a 5-km urban route with 10 stops that is 5.7 Wh, or 1.1 % capacity of a 500-Wh battery, or ~0.3 km of range.

Approximate city-vs-cruise ratios:

  • Steady cruise (subway-like route, mostly without stops) — baseline Wh/km
  • Light traffic (2–4 stops/km) — +5–10 %
  • Dense urban (4–8 stops/km, frequent slowdowns) — +15–25 %
  • Stop-and-go gridlock (>8 stops/km) — +30–50 %

Adjacent to aggressive throttle behaviour: fast accelerations work in the motor’s low-efficiency band (peak torque coincides with peak current and peak I²R losses). Smooth throttle modulation saves 5–10 % Wh on top of the start-stop derate itself.

13. Worked example: full real-range calculation

Scenario: rider 80 kg + scooter 18 kg + backpack 8 kg = m = 106 kg. A 7-km city-centre route in January: average temperature −5 °C, headwind 4 m/s (Bft 3), 6 stops on route, 50 m total climb (+0.7 % average gradient). Scooter: Xiaomi 4 Pro, nameplate 446 Wh, 250 cycles, SoH 89 %. Mode: sport (25 km/h average).

Battery usable Wh:

E_usable_at_battery = 446 × (1 − 0.10) × 0.89 × (1 − 0.30 cold derate) = 250 Wh

(−30 % cold derate at −5 °C is in the upper part of the 0 °C row of the table in §8)

Wh/km from each axis:

Baseline summer (20 °C, calm, payload 80 kg, flat, smooth):

P_drag (25 km/h) = 0.5 × 1.204 × 6.94³ × 0.60 = 121 W
P_roll (Crr = 0.012, m = 98 kg) = 0.012 × 98 × 9.81 × 6.94 = 80 W
P_grade (flat) = 0
P_accel (steady) = ~10 W avg
P_total = 211 W
Wh/km = 211 / (3.6 × 25) = 12.2 at-wheel
Wh/km from battery = 12.2 / η_drivetrain (0.68) = 17.9 Wh/km
Range = 446 / 17.9 = 24.9 km (baseline summer at nameplate)

What changes in our real scenario:

  1. Cold air (+8 % drag): P_drag → 131 W
  2. Headwind 4 m/s at 25 km/h (v_air = 10.94 m/s instead of 6.94; P_drag = 0.5 × 1.293 × 10.94³ × 0.60 = 506 W vs calm 131 W = +375 W)
  3. Payload +26 kg (m = 106 + 18 = 124 kg scooter+rider vs baseline 98): P_roll +27 %, P_grade proportional
  4. Grade +0.7 %: P_grade = 124 × 9.81 × 0.007 × 6.94 = 59 W (averaged over the route)
  5. Start-stop ~0.6 Wh × 6 = 3.6 Wh for the route (beyond P_accel-steady)
  6. Tire pressure assumed 90 % nominal (a week without a pump): Crr → 0.014 (+17 %)
P_drag = 506 W (headwind dominant)
P_roll = 0.014 × 124 × 9.81 × 6.94 = 118 W
P_grade = 59 W
P_accel_steady = ~10 W
P_total_mechanical = 693 W
P_battery = 693 / 0.68 = 1019 W
Wh/km from battery = 1019 / (3.6 × 25) = 11.3 Wh/km

Trip time 7 km / 25 km/h = 0.28 h = 1008 s. Energy for the route:

E_mechanical = 693 W × 1008 s = 698 544 J = 194 Wh
E_battery = 194 / 0.68 = 285 Wh
+ start-stop = 285 + 3.6 = 289 Wh

But E_usable_at_battery = 250 Wh < E_required = 289 Wh. The scooter will not make it. That means for this route the rider must either:

  • (a) drop speed to 18–20 km/h (drag falls by (25/18)³ = 2.68× → ~190 W instead of 506 at the same effective headwind; P_total → ~377 W; E_battery → 234 Wh — comfortably makes it)
  • (b) insulate the battery with a thermal sleeve and keep it indoors until departure (capacity derate from −30 % → −15 %; E_usable → 304 Wh — makes it with margin)
  • (c) shorten the route by 1.5 km via an alternative path or partial public transport
  • (d) pump tires to nominal, audit the backpack for drag penalty (cinch tight), accept that the headwind is orthogonal to the optimal route — reroute through interior streets

This is the realistic model — it shows margin or deficit, instead of trusting an «up to 45 km» nameplate.

14. Manufacturers vs reality: testing standards

Manufacturers test under standard conditions governed by the following standards:

  • EN 17128:2020 Personal Light Electric Vehicles (CEN/TC 354, AFNOR secretariat; published 21 October 2020, effective 30 April 2021) — the primary European standard for e-scooters not subject to vehicle type-approval. Battery voltage up to 100 VDC; integrated chargers up to 240 VAC input. Covers electrical/mechanical/quality/environmental safety. Range methodology is specified in the standard, but conditions assume a 65–80-kg dummy rider, ambient 20 ± 5 °C, steady-state cycle on a flat track.
  • UNECE R136 uniform technical prescriptions for L-category vehicles with electric powertrain — applies to e-scooters classified as L1e-A or L1e-B (e-bikes / mopeds). Type-approval conditions for type II inspection.
  • SAE J1634 Multi-Cycle Test (MCT) — North-American EV range standard; dynamometer testing with UDDS city cycle + highway cycle + steady-state cycles down to battery cutoff (SAE J1634:2017 latest revision).
  • WMTC (Worldwide harmonized Motorcycle Test Cycle) — UN GTR No.2 — used for type-approval of motorcycles and some high-performance e-scooters; includes accel/decel transients closer to reality than steady-state cycles.
  • WLTC / WLTP (Worldwide Light Vehicles Test Procedure) — for cars, not e-scooters, but referenced by parts of the industry as a «realistic» benchmark.

Why manufacturer range is always above real-world:

  1. Test on 65–80-kg rider, not 100+
  2. Test at 18–20 km/h eco cycle, not 25–30 sport
  3. Test at 20 °C, not winter
  4. Test in calm air, not windy
  5. Test on fully inflated tires
  6. Test on a new battery (SoH 100 %)
  7. Test on a flat track without stops

Each of the 7 items is a 5–15 % derate that stacks. So realistic real-world range for commuter models is 50–65 % of nameplate; for performance models in aggressive use — 40–55 %.

15. Route planning: practical workflow

How to turn this model into action before a ride:

  1. Find total elevation gain on a map (Strava, Komoot, openrouteservice.org return elevation profiles)
  2. Check weather — temperature, wind direction relative to the route, gusts
  3. Weigh payload — backpack included
  4. Check tire pressure (manual gauge, not the pump’s built-in one)
  5. Check SoH on the display or in the app (Xiaomi/Segway/Apollo/NAVEE apps surface battery-health % after ~100 cycles)
  6. Estimate Wh/km from the model for your scenario
  7. Multiply by planned distance and add a 20 % safety margin
  8. If total > 0.85 × E_usable — drop speed, shorten the route, or carry a charger

Simplified pre-trip checklist:

  • Winter? Cut expected range to 50–60 % of nameplate
  • Strong headwind? Drop speed by 20–30 % (non-linear drag savings)
  • Big gradient? Re-plan to flatter alternative (overall cheaper even at +10–15 % distance)
  • Payload > 80 kg? +10–20 % Wh/km baseline

16. Telematics monitoring: what the display and app show

Modern e-scooters expose real-time metrics useful for in-ride model validation:

  • Wh/% remaining — some apps (Apollo Hub, Dualtron Mini, Segway-Ninebot Connect) show estimated range from recent Wh/km
  • Wh/km running average — Mini Motors, Veteran display, Begode-style EUC apps have this counter; scooters less commonly, but via an aftermarket BMS (ASI BAC500, Sabvoton, VESC-derived controllers) it is accessible
  • Battery voltage sag — if voltage under load drops more than expected (Li-ion typical 3.6–3.7 V at 50 % SoC; cold or aged → 3.3–3.5 V), that is a signal of SoH degradation or cold derate
  • Battery temperature — top-tier models (Apollo Phantom V3, NAMI Burn-E2, Dualtron Storm) display cell temperature; reading < 15 °C means cold-derated capacity, > 45 °C means thermal-throttle or BMS shutdown risk

Pre-trip habit: log Wh/% (e.g. 80 % start, 14 km ridden, arrived at 30 %) → compute Wh/km and use as baseline for similar conditions next time. After 5–10 trips you will have a personal Wh/km dataset more accurate than any formula.

17. Recap — 8 key points

  1. Power equation P = P_drag + P_roll + P_grade + P_accel — universal for any wheeled, single-rider transport, validated by Wilson MIT Press + Martin 1998. Directly applicable to e-scooters with adapted Cd·A 0.55–0.70 m² and Crr 0.008–0.035.
  2. Drag dominates above 20 km/h and scales cubically. The single largest lever for range extension is dropping speed by 5–8 km/h.
  3. Rolling resistance depends on (a) tire type — pneumatic 0.008–0.015, solid 0.020–0.035; (b) pressure — each −10 % nominal → +15–25 % Crr; (c) surface — cobblestone/gravel 2–4× asphalt.
  4. Grade — each +1 % gradient at 25 km/h costs ~+5–7 % Wh/km; full descent regen returns no more than 10–15 % of grade energy.
  5. Start-stop — each full cycle at 25 km/h = 0.64 Wh kinetic, ~90 % of which is dumped as heat. 8+ stops/km → +15–25 % Wh/km vs steady cruise.
  6. Drivetrain η_total ≈ 0.55–0.75, so a 500-Wh nameplate delivers 275–375 Wh at the wheel.
  7. Cold derating is the single largest penalty axis: 0 °C → −20–30 % usable Wh; −10 °C → −30–40 %; −20 °C → −50 %. Stacks with the air-density +8–16 % drag penalty.
  8. Manufacturers test under ideal conditions (EN 17128, SAE J1634, WMTC); real-world range = 40–65 % of nameplate. A personal Wh/km dataset from 5–10 trips beats any formula.

Sources (ENG-first, 0 RU)