
Electric Vehicles
Push a car up a slope, watch a hybrid's power flow switch modes, build a battery pack cell by cell, weaken a motor's field and fast-charge a pack to 80%
18 chapters, 21 interactives, few words. Every chapter is mapped to the Anna University, VTU, SPPU, Mumbai University and KTU units it covers: EV and hybrid architectures, vehicle dynamics and tractive effort, drive cycles and range, transmissions and the electronic differential, traction motors and field-oriented control, power electronics, lithium-ion batteries and battery modelling, battery management and thermal management, fuel cells and ultracapacitors, charging standards and CC-CV charging, vehicle-to-grid, hybrid energy management and regenerative braking, powertrain sizing, and EV safety and homologation.
- AUCME336Hybrid and Electric Vehicle Technology
- AUEE3025Electric Vehicle Architecture
- VTUBME306AElectric and Hybrid Vehicle Technology
- VTUBEE515CElectric Vehicle Fundamentals
- SPPU402051EElectric and Hybrid Vehicle
- SPPU303151BElectric Mobility
- MUEEC405Electric and Hybrid Electric Vehicles
- KTUMET453Hybrid and Electric Vehicles
By Skill-Lync · Updated 17 Nov 2026 · 18 chapters · 21 interactives · free, no sign-up
Why electric: history, the limits of IC engines, emissions and India's EV scenario
Which syllabus covers thisAU CME336 Unit I; EE3025 Unit IVTU BME306A Module 1; BEE515C Module 1SPPU 402051E Unit 1; 303151B Unit 1MU EEC405 Module 1KTU MET453 Module 1
Are electric vehicles new?
No. Battery carriages ran in the 1830s, and around 1900 electric cars outsold petrol cars in some US cities because they were quiet and needed no hand crank. Cheap oil, Ford's mass-produced Model T (1908) and the electric self-starter (1912) ended that era. EVs came back with emission laws in the 1990s (GM EV1, Toyota Prius hybrid in 1997) and took off once lithium-ion cells made 300+ km of range practical, from the Tesla Roadster (2008) and Nissan Leaf (2010) on. In India the REVA (2001) was the first mass-made electric car.
Why is an IC engine car inefficient?
An engine turns only about 20-30% of the fuel's energy into work at the wheels on a real drive; the rest leaves as heat in the exhaust and coolant. It is worst exactly where city driving happens: at low load and while idling. An electric drive turns about 80-85% of the battery's energy into wheel work, needs no idling and can recover braking energy. The trade is energy storage: petrol holds about 12,000 Wh/kg, a lithium-ion pack about 150-200 Wh/kg.
Is an EV cheaper and cleaner per kilometre?
Cheaper to run almost everywhere in India: about ₹1.1/km on a home tariff against ₹6-7/km for a petrol car. Cleaner depends on the grid. India's grid emits roughly 0.7 kg CO2 per kWh because it is mostly coal, so an EV is around 100 g/km at the power plant against about 150 g/km at a petrol car's tailpipe. The gap widens as solar and wind grow. Move the sliders to try your own fuel price, tariff and grid.
What happens to EV batteries at the end of life?
A traction battery is usually retired at 70-80% of its original capacity. It can then serve 5-10 more years as stationary storage (second life), and afterwards be recycled to recover lithium, nickel, cobalt and copper. Lead-acid and lithium cells are hazardous waste: they must never go to landfill. India's Battery Waste Management Rules, 2022 make producers responsible for collecting and recycling the batteries they sell.
What does India's EV market look like?
India electrified two- and three-wheelers first: e-rickshaws and electric scooters make up most EVs registered, because a 2-4 kWh battery is affordable and daily trips are short. Cars and buses followed with state and central subsidies (the FAME schemes), local cell and battery-pack manufacturing incentives (PLI for advanced chemistry cells) and city bus tenders. The engineering challenges are heat (cells age faster above 35 °C), cost, charging infrastructure and grid capacity.
Petrol at ₹103/L in a 15 km/L car vs an EV using 0.14 kWh/km at ₹8/kWh. Over 12,000 km a year, the EV saves about:
Petrol costs 103/15 = ₹6.87/km, the EV 0.14 × 8 = ₹1.12/km. The difference, ₹5.75/km, over 12,000 km is ₹68,960.
Worked examples
Running cost per km · Petrol ₹103/L, 15 km/L; EV 0.14 kWh/km at the wall, tariff ₹8/kWh; 12,000 km a year
- Petrol: 103 ÷ 15 = ₹6.87/km
- EV: 0.14 × 8 = ₹1.12/km
- Saving: (6.87 − 1.12) × 12,000 = ₹68,960 a year
Answer: ₹68,960 a year, before maintenance (EVs have no oil changes)
CO2 per km on India's grid · Petrol 2.31 kg CO2 per litre burned, 15 km/L; EV 0.14 kWh/km, grid 0.72 kg CO2/kWh
- Petrol: 2.31 ÷ 15 = 154 g/km at the tailpipe
- EV: 0.14 × 0.72 = 0.1008 kg = 101 g/km at the power plant
- Neither number includes refining, nor making the battery (a large one-time CO2 debt the EV pays back over its first few years)
Answer: 154 vs 101 g CO2/km, about 35% lower for the EV
Efficiency chain from the socket to the wheel · Charging 90%, battery discharge 95%, inverter 97%, motor 93%, gear 97%
- Battery to wheel: 0.95 × 0.97 × 0.93 × 0.97 = 0.831
- Wall to wheel: 0.90 × 0.831 = 0.748
- An IC engine car on a city cycle: about 0.20-0.25 tank to wheel
Answer: 75% from the wall, about three times a petrol car
EV and hybrid architectures: BEV, HEV, PHEV, FCEV, series, parallel and power-split
Which syllabus covers thisAU CME336 Unit II; EE3025 Units I, VVTU BME306A Module 1; BEE515C Modules 2, 4SPPU 402051E Unit 2; 303151B Unit 1MU EEC405 Module 2KTU MET453 Module 1
What are BEV, HEV, PHEV, EREV and FCEV?
A battery electric vehicle (BEV) has only a battery and motor and is charged from the grid. A hybrid (HEV) has an engine and a small battery that is charged only by the engine and braking. A plug-in hybrid (PHEV) has a bigger battery you can charge from the grid for 30-80 km of electric driving. A range-extended EV (EREV) is driven only by its motor, with a small engine that just charges the battery. A fuel cell EV (FCEV) makes its electricity on board from hydrogen.
What are micro, mild, full and plug-in hybrids?
They differ in how much electric power they have. A micro hybrid only stops and restarts the engine at lights (12 V starter-generator). A mild hybrid (often 48 V) assists the engine and recovers some braking energy but cannot drive on electricity alone. A full hybrid (200-300 V and up) can drive short distances on the motor alone. A plug-in adds grid charging. The hybridisation factor compares motor power to total power.
What is the difference between a series and a parallel hybrid?
In a series hybrid the engine only drives a generator; the wheels are always turned by the motor (electrical coupling), so the engine can run at its best point but every kWh is converted twice. In a parallel hybrid the engine and motor both connect mechanically to the wheels and their torques add (torque coupling) or their speeds add (speed coupling). It loses less energy at highway speeds but the engine speed is tied to road speed. Pick an architecture and a mode to watch the power flow.
How does a power-split (series-parallel) hybrid work?
A planetary gearset joins the engine (carrier), a generator (sun) and the output with the traction motor (ring). Part of the engine's power goes mechanically to the wheels and part goes electrically through the generator and motor. Because the generator's speed can change freely, the engine speed is decoupled from road speed: the gearset acts as an electronic CVT. The speeds always satisfy the Willis equation below.
How is a plug-in hybrid different from a hybrid?
A PHEV has a battery of roughly 8-20 kWh instead of about 1-2 kWh, an on-board charger and a charge port. It runs in charge-depleting mode (mostly electric) until the battery reaches its lower limit, then switches to charge-sustaining mode and behaves like an ordinary hybrid. Its fuel economy therefore depends heavily on how often it is plugged in.
A power-split gearset has a 30-tooth sun and 78-tooth ring. The engine (carrier) turns at 2,000 rpm and the ring (wheels) at 3,000 rpm. The generator on the sun turns at:
ω_s = ((30 + 78) × 2,000 − 78 × 3,000) / 30 = -600 rpm. At high road speed the generator spins backwards and acts as a motor: some power circulates electrically.
Worked examples
Hybridisation factor · Full hybrid: engine 72 kW, motor 53 kW. Mild hybrid: engine 100 kW, 48 V motor 12 kW
- Full: HF = 53 ÷ (53 + 72) = 0.424
- Mild: HF = 12 ÷ (12 + 100) = 0.107
- Pure EV: HF = 1, conventional car: HF = 0
Answer: 0.42 for the full hybrid, 0.11 for the mild hybrid
Power-split speeds · Sun 30 teeth (generator), ring 78 teeth (motor and wheels), carrier = engine
- Willis: 108 ω_c = 30 ω_s + 78 ω_r
- Cruising: ω_c = 2,000, ω_r = 3,000 → ω_s = (216,000 − 234,000)/30 = -600 rpm
- Stopped and charging: ω_r = 0, ω_c = 1,200 → ω_s = 108 × 1,200/30 = 4320 rpm
Answer: -600 rpm (backwards) while cruising; 4320 rpm when charging at a standstill
Series vs parallel path efficiency · Engine 33%, generator 92%, power electronics 97%, motor 92%, final gear 97%; parallel mechanical path 95%
- Series: 0.33 × 0.92 × 0.97 × 0.92 × 0.97 = 0.263
- Parallel (engine straight to wheels): 0.33 × 0.95 = 0.314
- Series wins in stop-go traffic only because its engine can stay at its best point or switch off
Answer: 26.3% vs 31.4% at steady highway speed
Vehicle mechanics: rolling, aerodynamic, grade and acceleration resistance, and tractive effort
Which syllabus covers thisAU CME336 Unit I; EE3025 Unit IIVTU BME306A Modules 1, 4SPPU 402051E Unit 4; 303151B Unit 1MU EEC405 Modules 1, 5KTU MET453 Module 1
What forces does a vehicle's motor have to overcome?
Four. Rolling resistance from the tyres flexing (about 1% of the weight for a car on asphalt). Aerodynamic drag, which grows with the square of speed. Grade resistance, the component of weight down a slope. And the inertia force to accelerate the mass, increased by about 5-10% for the rotating wheels, gears and motor (the mass factor δ). Their sum is the tractive effort the wheels must deliver.
What is the dynamic equation of vehicle motion?
Newton's second law along the road: the tractive force minus the total resistance accelerates the vehicle. When the tractive force exactly equals the resistance, the vehicle holds a steady speed; that intersection on a force-speed graph gives the top speed on any grade. Power at the wheel is force times speed, which is why the same car needs about four times the power at 120 km/h that it needs at 60 km/h.
How much force can a tyre actually transmit?
At most the friction coefficient times the normal load on the driven wheels: μ is about 0.8-0.9 on dry asphalt, 0.5 on wet and 0.1 on ice. Beyond that the wheel spins. The tyre transmits force through a little slip: tractive force rises almost linearly with slip up to a peak at around 10-20% slip, then falls. Load transfer under acceleration moves weight to the rear axle, which favours rear-wheel drive for traction.
How do you convert tractive force to motor torque and speed?
Multiply the force by the wheel radius to get wheel torque, then divide by the gear ratio and gear efficiency to get motor torque. Motor speed is wheel speed times the gear ratio. A single-speed EV reduction is usually 7:1 to 11:1, so a motor spinning 14,000 rpm turns the wheels about 1,500 rpm.
A 1,500 kg car (CdA 0.66 m², Crr 0.010) cruises at 120 km/h on a flat road. Which resistance is bigger?
Drag ½ × 1.2 × 0.66 × 33.3² = 440 N; rolling 0.010 × 1,500 × 9.81 = 147 N. Drag passes rolling resistance at about 65 km/h for this car.
Worked examples
Tractive effort on a slope · Car 1,500 kg, Crr 0.010, CdA 0.66 m², ρ 1.2 kg/m³; 80 km/h up a 4% grade
- v = 22.22 m/s, θ = atan(0.04) = 2.29°
- Rolling: 0.010 × 1,500 × 9.81 × cos θ = 147.0 N
- Drag: ½ × 1.2 × 0.66 × 22.22² = 195.6 N
- Grade: 1,500 × 9.81 × sin θ = 588.1 N
- Total 930.7 N; power at the wheel 930.7 × 22.22 = 20.68 kW
Answer: 931 N and 20.7 kW at the wheels
Motor torque and speed for that load · Same 930.7 N; wheel radius 0.31 m; single-speed gear 9:1, 95% efficient
- Wheel torque: 930.7 × 0.31 = 288.5 N·m
- Motor torque: 288.5 ÷ (9 × 0.95) = 33.7 N·m
- Motor speed: 22.22 ÷ 0.31 × 9 = 645.2 rad/s = 6161 rpm
Answer: 33.7 N·m at 6161 rpm
Traction limit of a front-wheel-drive car · 60% of 1,500 kg on the front (driven) axle, μ = 0.8, load transfer ignored
- Normal load on the driven axle: 0.6 × 1,500 × 9.81 = 8829 N
- Maximum tractive force: 0.8 × 8829 = 7063 N
- Maximum acceleration: 7063 ÷ (1.05 × 1,500) = 4.48 m/s²
- Under acceleration weight moves rearwards, so the real front-drive limit is lower
Answer: About 7063 N, 4.5 m/s² before the front wheels spin
Performance: motor torque-speed, gear ratio, top speed, gradeability and 0-100 km/h
Which syllabus covers thisAU CME336 Units I, V; EE3025 Unit IIVTU BME306A Module 4SPPU 402051E Unit 4MU EEC405 Module 5KTU MET453 Module 3
Why does an EV need only one gear?
A traction motor gives full torque from standstill up to its base speed and then constant power up to about three to four times that speed. That is close to the ideal traction curve: high force to launch and climb, constant power for top speed. An IC engine makes useful torque only in a narrow band, so it needs a clutch and five or more gears to cover the same range. Most EVs therefore use one fixed reduction.
How do you read top speed and gradeability off a tractive effort graph?
Plot the motor's tractive effort at the wheels against vehicle speed, then plot the road-load curves for 0%, 10% and 20% grade. Top speed is where the effort curve meets the flat-road curve, or the speed at which the motor reaches its rpm limit, whichever comes first. Gradeability is the steepest grade whose road-load curve still sits under the effort curve. The gap between them at any speed is what is left to accelerate.
How is the gear ratio chosen?
Two limits pull in opposite directions. The maximum motor speed must still reach the target top speed, which caps the ratio from above. The motor's peak torque times the ratio must give the required launch force and gradeability, which sets a minimum. A higher ratio gives a faster 0-100 km/h but a lower top speed and more motor rpm on the highway.
How do you calculate 0-100 km/h time?
Integrate the dynamic equation: at each speed, the acceleration is the spare tractive force divided by the effective mass δm. Time is the integral of δm dv divided by that spare force, from 0 to 27.8 m/s. By hand you split it into a constant-torque part (nearly constant acceleration) and a constant-power part; in practice it is a 10-line loop in MATLAB or Python.
What is the extended speed ratio?
The maximum speed divided by the base speed of the motor (x). A larger x means a longer constant-power region, so a single gear can cover both launch and top speed. Induction and interior permanent magnet motors reach x = 3-5 with field weakening; surface-magnet motors have a shorter constant-power range. Larger x also lets a smaller-torque motor do the same job with a higher gear ratio.
Doubling the gear ratio of an EV (same motor) mainly:
Wheel force = T·G·η/r doubles, but the motor reaches its rpm limit at half the road speed. Power, and so the constant-power part of the curve, is unchanged.
Worked examples
Gear ratio from the top speed · Motor maximum 14,000 rpm, target top speed 160 km/h, wheel radius 0.31 m
- ω_max = 14,000 × 2π/60 = 1466.1 rad/s
- v_top = 160/3.6 = 44.44 m/s
- G_max = 1466.1 × 0.31 ÷ 44.44 = 10.23
Answer: A ratio no higher than about 10.2:1
Base speed and speed ratio · Peak torque 250 N·m, peak power 110 kW, maximum 14,000 rpm, gear 9:1, r = 0.31 m
- ω_b = P/T = 110,000/250 = 440 rad/s = 4202 rpm
- Road speed at base: 440 × 0.31/9 = 15.16 m/s = 54.6 km/h
- x = 14,000/4202 = 3.33
Answer: Full torque up to 55 km/h, constant power beyond; x = 3.3
Gradeability at crawling speed · Same drive: 250 N·m × 9 × 0.95 ÷ 0.31 m; 1,500 kg, Crr 0.010, drag ignored
- Tractive effort: 6895 N
- Solve 1,500 × 9.81 × (sin θ + 0.010 cos θ) = 6,895
- θ = 27.4°, grade = tan θ = 52%
Answer: About 52% (a 27° slope), far beyond any road
0-100 km/h by numerical integration · Same car and drive; δ = 1.05; step δm dv/dt = F_t − F_road every 0.1 ms
- Constant-torque phase to 55 km/h: a ≈ (6,895 − 150)/(1.05 × 1,500) ≈ 4.3 m/s²
- Constant-power phase: F_t = 110,000 × 0.95/v, falling with speed
- Integrating to 27.78 m/s gives t = 7.97 s
Answer: 8.0 s
Drive cycles and energy consumption: MIDC, WLTP, Wh/km and range
Which syllabus covers thisAU CME336 Unit II (fuel efficiency analysis); EE3025 skill activity (drive cycles)VTU BME306A Module 2SPPU 402051E Units 2, 3, 5MU EEC405 Module 5KTU MET453 Module 1
What is a drive cycle?
A fixed speed-versus-time trace that every vehicle is driven through on a chassis dynamometer, so fuel economy, emissions and EV range are measured the same way for every model. India's Modified Indian Driving Cycle (MIDC) for cars repeats a 195-second urban segment four times, then adds an extra-urban segment capped at 90 km/h. Europe now uses WLTP, the US the FTP-75 and HWFET; two-wheelers use their own IDC/WMTC cycles.
How do you get Wh/km from a drive cycle?
Step through the cycle second by second. At each step compute the road load and acceleration force, multiply by speed for wheel power, divide positive power by the drivetrain efficiency (energy taken from the battery) and multiply negative power by the regeneration efficiency (energy returned). Sum over the cycle and divide by its distance. Choose a cycle and change the regeneration efficiency to see how much stop-go driving gives back.
Why do EVs do better in the city than on the highway?
In the city, most of the energy goes into accelerating the mass, and the motor recovers a good part of it when braking, while drag is small at low speed. On the highway drag dominates and grows with v², and nothing is recovered. Petrol cars are the reverse: their worst efficiency is in stop-go traffic. That is why an EV's highway range is often 20-30% lower than its certified city-heavy figure.
How is range estimated?
Usable battery energy divided by consumption per km. Usable energy is less than the nameplate: the BMS keeps a buffer at the top and bottom of the state of charge, and capacity fades with age. Consumption rises with speed, air-conditioning (1-3 kW), heavy loads, cold cells and hills. Certified ranges come from the cycle test; real ranges are usually 15-30% lower.
The ECE-15 urban segment lasts 195 s and covers 0.994 km. It spends 60 s standing still. Its average speed while moving is about:
Overall average 0.994 km / 195 s = 18.35 km/h; excluding the 60 s of idling, 0.994 km / 135 s = 26.5 km/h. The peak is only 50 km/h.
Worked examples
ECE-15 urban segment · Accelerations to 15, 32 and 50 km/h with cruises and stops; 195 s in total
- Each straight-line ramp covers its average speed × time, e.g. 0 → 50 km/h in 26 s = 13.89/2 × 26 = 180.6 m
- Sum of all segments: 994.0 m
- Average speed 18.35 km/h; 26.5 km/h while moving
Answer: 0.994 km at 18.4 km/h average
Energy at a steady 50 km/h · Car 1,500 kg, Crr 0.010, CdA 0.66 m²; drivetrain 90% battery to wheel
- Road load: 147.2 + ½ × 1.2 × 0.66 × 13.89² = 223.5 N
- Energy per km at the wheel: 223.5 N × 1,000 m = 223.5 kJ = 62.1 Wh
- From the battery: 62.1 ÷ 0.9 = 69.0 Wh/km
Answer: 69 Wh/km before air-conditioning and auxiliaries
What one stop from 50 km/h costs · Same car; regeneration returns 60% of braking energy to the battery
- Kinetic energy: ½ × 1,500 × 13.89² = 144.7 kJ = 40.2 Wh
- Recovered: 0.6 × 40.2 = 24.1 Wh
- Lost in friction and conversion: 16.1 Wh, as much as 233 m of steady 50 km/h driving
Answer: 40 Wh of kinetic energy, 24 Wh of it recovered
Range · 40 kWh usable, 140 Wh/km average
- 40,000 ÷ 140 = 286 km
- At a highway 180 Wh/km: 40,000 ÷ 180 = 222 km
Answer: About 286 km mixed, 222 km on the highway
Transmission, differential and brakes: single-speed reduction, electronic differential and powertrain layout
Which syllabus covers thisAU CME336 Unit I (transmission characteristics); EE3025 Unit IIIVTU BME306A Module 3 (mechanical connections of motors)SPPU 402051E Unit 4MU EEC405 Module 2KTU MET453 Module 3
What does an EV transmission look like?
Usually a single fixed reduction of about 7:1 to 11:1, built as two gear stages (or a planetary set) in one housing with the differential. No clutch is needed because the motor starts from zero speed under load. A few performance cars and heavy trucks use two-speed gearboxes to get both strong launch and a high top speed or better motor efficiency on the highway. Reverse is simply the motor turning the other way.
Why does a vehicle need a differential, and what is an electronic differential?
In a turn the outer wheels travel a longer arc than the inner ones, so they must spin faster. A mechanical differential lets them while splitting torque equally. With two motors, one per wheel (or in-wheel hub motors), there is no mechanical differential: the controller sets each wheel's speed or torque from the steering angle and vehicle speed. That electronic differential can also do torque vectoring. Set a turn radius and speed to see the wheel-speed difference.
Where do the motors and battery go?
A single motor drives the front or rear axle through the reduction; dual-motor cars put one on each axle for all-wheel drive. Two- and three-wheelers often use a hub motor inside the wheel (simple, but it adds unsprung mass) or a mid-drive motor with a belt or chain. The battery sits low between the axles in cars (a flat 'skateboard'), which lowers the centre of gravity, and under the floor or seat in scooters and rickshaws.
Do EVs still need friction brakes?
Yes. Regenerative braking (chapter 16) handles most everyday slowing, but the motor cannot hold a car stationary, its power is limited, and a full battery cannot accept charge. Hydraulic disc brakes, sized exactly as on a conventional car for a full emergency stop, take over in those cases. Because they are used less, EV brake pads last much longer, and the discs can rust from disuse.
On a 10 m radius turn at 20 km/h with a 1.5 m track, the outer rear wheel turns faster than the inner one by about:
Outer and inner wheels run on radii of 10.75 m and 9.25 m: the ratio 1.162 does not depend on speed, only on R and the track.
Worked examples
Wheel speeds in a turn · Turn radius 10 m (vehicle centre), track 1.5 m, 20 km/h, wheel radius 0.31 m
- Inner wheel: 5.556 × 9.25/10 = 5.139 m/s → 158.3 rpm
- Outer wheel: 5.556 × 10.75/10 = 5.972 m/s → 184.0 rpm
- Ratio 1.162
Answer: 158 and 184 rpm: the differential (or the two motor controllers) must allow a 16% difference
Ackermann steering angles · Wheelbase 2.6 m, same turn (radius measured to the rear-axle centre)
- Inner front wheel: atan(2.6/9.25) = 15.70°
- Outer front wheel: atan(2.6/10.75) = 13.60°
Answer: 15.7° inside, 13.6° outside, so both wheels roll about the same centre
Front brake torque for an emergency stop · 1,500 kg, 0.8 g deceleration, 70% of braking on the front axle, wheel radius 0.31 m
- Total brake force: 1,500 × 0.8 × 9.81 = 11772 N
- Per front wheel: 0.7 × 11772 ÷ 2 = 4120 N
- Torque: 4120 × 0.31 = 1277 N·m, several times what a traction motor can absorb through the reduction
Answer: About 1277 N·m per front disc
Traction motors: DC, induction, PMSM, BLDC and switched reluctance compared
Which syllabus covers thisAU CME336 Unit III; EE3025 Unit IIIVTU BME306A Module 3; BEE515C Module 2SPPU 402051E Unit 3; 303151B Unit 5MU EEC405 Module 3KTU MET453 Module 2
What does an EV need from its motor?
High torque from standstill, a wide constant-power speed range, high efficiency over the whole drive cycle (not just at one point), high power per kg and per litre, good overload capacity for short bursts, low noise, ruggedness and low cost. No single motor wins all of these, which is why different vehicles use different types.
Which motor types are used in EVs?
Brushed DC motors are simple to control but the brushes wear: early e-rickshaws used them. Induction motors are rugged, cheap and magnet-free, with good field weakening, but have rotor copper loss. Permanent magnet synchronous motors (PMSM) are the most efficient and power-dense and drive most cars today; interior-magnet (IPM) rotors add reluctance torque. Switched reluctance motors have no magnets or windings on the rotor but are noisy with torque ripple.
How does a BLDC hub motor in a scooter work?
It is a permanent magnet motor with trapezoidal back-EMF, driven in six steps. Three Hall sensors report rotor position, and the controller energises the two phases whose back-EMF is flat at that moment while the third floats. It is cheap and simple, which suits two- and three-wheelers; its torque ripples slightly at each step. Sinusoidal PMSM drives with field-oriented control are smoother.
What is a motor efficiency map?
A contour plot of efficiency over the whole torque-speed plane, under the peak torque/power envelope. Copper loss grows with torque squared, so efficiency drops at high torque and low speed; iron loss grows with speed, so it drops again at high speed and light load. A PMSM's best island is usually 95-97%; an induction motor peaks a little lower, but at high speed and light load its lower iron loss can beat the magnet motor, one reason some dual-motor cars put an induction motor on the cruising axle. Overlaying a drive cycle's operating points on the map tells you the cycle efficiency.
Which magnets do traction motors use?
Neodymium-iron-boron (NdFeB) for the highest energy density, with dysprosium added so they keep their magnetism at 150 °C+. Samarium-cobalt survives higher temperatures but costs more. Ferrite is cheap and rare-earth-free but about a third as strong, so a ferrite motor is larger. Rare-earth supply risk (most refining is in China) is why magnet-free induction, SRM and wound-rotor motors are being developed.
A 48 V hub motor with 0.05 Ω winding resistance runs with 46 V back-EMF. Its current is:
I = (V − E_b)/R = (48 − 46)/0.05 = 40 A. Only the 2 V difference drives current: the back-EMF does the real work.
Worked examples
DC hub motor operating point · 48 V supply, back-EMF 46 V, winding resistance 0.05 Ω; scooter at 36 km/h, wheel radius 0.25 m (direct drive)
- Current: (48 − 46)/0.05 = 40 A
- Input 1920 W; mechanical E_b·I = 1840 W; efficiency 95.8% (copper loss only)
- Wheel speed: 10 m/s ÷ 0.25 = 40 rad/s; torque 1840/40 = 46 N·m
Answer: 40 A, 1.84 kW, 46 N·m
Raise the supply by 1 V · Same load torque, so the same 40 A; supply 49 V
- E_b = 49 − 40 × 0.05 = 47 V
- Speed is proportional to back-EMF: 36 × 47/46
- = 36.78 km/h
Answer: 36.8 km/h
BLDC commutation rate · Hub motor with 4 pole pairs at 600 rpm
- Electrical frequency: 600/60 × 4 = 40 Hz
- Six Hall states per electrical cycle: 240 commutations per second
Answer: 40 Hz, 240 commutations/s
Induction motor speed on an inverter · 4-pole induction motor, inverter at 200 Hz, slip 3%
- Synchronous speed: 120 × 200/4 = 6000 rpm
- Rotor: 6000 × (1 − 0.03) = 5820 rpm
Answer: 5820 rpm; the inverter changes speed by changing frequency
Motor control: field-oriented control, MTPA, field weakening and sensorless drives
Which syllabus covers thisAU CME336 Unit IIIVTU BME306A Module 3; BEE515C Module 3SPPU 402051E Unit 3; 303151B Unit 5MU EEC405 Module 3KTU MET453 Module 3
What is field-oriented control?
A way to control an AC motor like a separately excited DC motor. The three phase currents are transformed into two DC quantities in a frame that rotates with the rotor: id along the magnet flux and iq at right angles to it. Torque then depends on iq (plus a reluctance term) and flux on id, and two PI controllers regulate them. The inverter turns the requested voltages back into three-phase PWM.
What do the Clarke and Park transforms do?
Clarke turns the three phase currents (which sum to zero) into two orthogonal components α and β on the stator. Park rotates them by the rotor angle θ into d and q, which are constant in steady state. Constant values are easy for a PI controller to track; sinusoids at hundreds of hertz are not. The inverse transforms rebuild the phase voltages.
What are MTPA and field weakening?
Below base speed the controller picks the id, iq pair that gives the requested torque with the least current: maximum torque per ampere (MTPA). For an IPM motor that means some negative id, to use reluctance torque. As speed rises the back-EMF approaches the battery voltage limit; driving id more negative weakens the effective flux so the motor can go faster at constant power. Raise the speed and watch the voltage ellipse shrink onto the operating point.
Why use space vector PWM?
The inverter's eight switch states form six active voltage vectors and two zero vectors. Space vector PWM builds any reference vector inside the hexagon by time-averaging the two nearest active vectors and a zero vector each switching period. It uses the DC bus 15.5% better than sine-triangle PWM (phase peak Vdc/√3 instead of Vdc/2) and gives lower current ripple.
How can a motor run without a position sensor?
By estimating rotor position from what the controller already measures. At medium and high speed the back-EMF carries the angle: flux-linkage integration, back-EMF zero-crossings (for BLDC) or an observer (sliding-mode, Luenberger, Kalman). At standstill there is no back-EMF, so methods inject a high-frequency signal and read the position from the IPM rotor's saliency (inductance varies with angle). Sensorless drives remove the resolver's cost and a failure point.
An IPM motor (ψm 0.08 Wb, Ld 0.15 mH, Lq 0.45 mH, 4 pole pairs) runs at id = −150 A, iq = 250 A. The reluctance torque is about:
1.5 × 4 × (Ld − Lq) × id × iq = 6 × (−0.3 × 10⁻³) × (−150) × 250 = 67.5 N·m, on top of the magnet torque 6 × 0.08 × 250 = 120 N·m.
Worked examples
Clarke and Park transforms · Phase currents ia = 10 A, ib = −5 A, ic = −5 A; rotor angle θ = 30°
- Clarke (amplitude-invariant): iα = ⅔(10 + 2.5 + 2.5) = 10 A, iβ = (−5 + 5)/√3 = 0
- Park: id = 10 cos 30° = 8.660 A
- iq = −10 sin 30° = -5.0 A
Answer: id = 8.66 A, iq = -5.0 A
Torque of a surface-magnet motor · SPM (Ld = Lq), p = 4 pole pairs, ψm = 0.10 Wb, iq = 300 A, id = 0
- Reluctance term is zero because Ld = Lq
- T = 1.5 × 4 × 0.10 × 300 = 180 N·m
Answer: 180 N·m
Magnet and reluctance torque in an IPM · p = 4, ψm = 0.08 Wb, Ld = 0.15 mH, Lq = 0.45 mH, id = −150 A, iq = 250 A
- Magnet: 1.5 × 4 × 0.08 × 250 = 120 N·m
- Reluctance: 1.5 × 4 × (−0.30 × 10⁻³) × (−150) × 250 = 67.5 N·m
- Total 187.5 N·m: negative id adds 56% more torque
Answer: 187.5 N·m
Bus utilisation: sine PWM vs SVPWM · DC bus 350 V
- Sine-triangle PWM, linear limit: phase peak = Vdc/2 = 175 V
- SVPWM: phase peak = Vdc/√3 = 202.1 V
- Ratio 1.1547
Answer: 202 V vs 175 V peak per phase: 15.5% more voltage, so a higher base speed
Power electronics in an EV: traction inverter, IGBT vs SiC, DC-DC converters and the on-board charger
Which syllabus covers thisAU CME336 Unit III (electric components)VTU BEE515C Module 3SPPU 402051E Units 3, 6MU EEC405 Module 5KTU MET453 Module 3
What power electronics does an EV have?
Four main converters. The traction inverter turns battery DC into variable-frequency three-phase AC for the motor and runs in reverse when regenerating. A DC-DC converter steps the high-voltage battery down to 12-14 V for lights, ECUs and pumps. The on-board charger turns grid AC into DC for the battery. Some vehicles also boost the battery voltage before the inverter. All of them are liquid-cooled, and the high-voltage side is isolated from the chassis.
IGBT, MOSFET or SiC: which switch does a traction inverter use?
Silicon IGBTs have a nearly constant on-state voltage (about 1.5-2 V) and suit 400-800 V, high-current inverters, but switch slowly. Silicon MOSFETs behave like a small resistance and switch fast, so they suit 48-96 V two- and three-wheeler controllers. Silicon carbide (SiC) MOSFETs combine high voltage with low resistance and fast switching, cutting inverter losses by roughly half at light load. They are the default in new 800 V cars despite costing more. Compare switching losses below.
Why does the 12 V DC-DC converter need a transformer?
Going from 400 V to 14 V with a plain buck converter would need a 3.5% duty cycle, impractically short on-times at high current, and it would leave the 12 V system galvanically connected to the high-voltage battery, which is unsafe. So EVs use isolated converters (phase-shifted full bridge or LLC) whose high-frequency transformer does the voltage step-down and provides isolation. The buck converter below shows the non-isolated principle.
What is inside an on-board charger?
Two stages. A boost power factor correction (PFC) stage draws a sinusoidal current in phase with the grid voltage (power factor above 0.99) and charges a DC link at about 400 V. An isolated DC-DC stage then controls the battery current and voltage through the charge profile. Without PFC the charger would draw a peaky current with high harmonics and a poor power factor, overloading wiring for the same real power.
How is the high-voltage system kept safe?
Main contactors on both battery poles open whenever the car is off or a fault is seen. A precharge resistor charges the inverter's DC-link capacitors before the main contactor closes, avoiding a huge inrush current. An insulation monitor checks the resistance from the HV bus to the chassis. HV cables are orange, connectors carry an interlock loop (HVIL) that opens the contactors if unplugged, and a manual service disconnect splits the pack for repair.
Why does a precharge circuit charge the inverter's capacitors before the main contactor closes?
A discharged DC-link capacitor looks like a short circuit. Connected straight to a 400 V pack through milliohms, it would draw thousands of amps. A 100 Ω resistor limits that to 4 A.
Worked examples
Conduction loss: MOSFET vs IGBT inverter · 200 A rms per phase; SiC MOSFET 2 mΩ per switch vs IGBT with 1.6 V on-state drop (diode drop taken the same)
- MOSFET: one switch per phase conducts the whole current: 3 × 200² × 0.002 = 240 W
- IGBT: loss ∝ average |i| = 2√2/π × 200 = 180.1 A; 3 × 1.6 × 180.1 = 864 W
Answer: 240 W vs 864 W of conduction loss (switching loss comes on top)
Precharge timing · DC link 500 µF, precharge resistor 100 Ω, pack 400 V
- τ = RC = 100 × 500 µF = 0.05 s
- 95% charged after 3τ = 0.15 s; peak current 400/100 = 4 A
- Energy dissipated in the resistor: ½CV² = 40 J
Answer: 0.15 s to 95%, peak 4 A instead of a dead short
Why chargers need PFC · 3.3 kW on-board charger on 230 V; power factor 0.99 with PFC vs 0.70 without
- With PFC: 3,300/0.99 = 3333 VA → 14.5 A
- Without: 3,300/0.70 = 4714 VA → 20.5 A
Answer: 14.5 A vs 20.5 A from the same socket for the same charging power
12 V converter current and duty · Auxiliary load 2 kW at 14 V from a 400 V battery
- Output current: 2,000/14 = 142.9 A
- Ideal non-isolated buck duty: 14/400 = 3.5%
Answer: 143 A at a 3.5% duty: hence a transformer-isolated converter
Batteries: lithium-ion cells, chemistries, capacity, C-rate, SoC, DoD and pack design
Which syllabus covers thisAU CME336 Unit IV; EE3025 Unit III (EV conversion components)VTU BME306A Modules 2, 4; BEE515C Module 1SPPU 402051E Units 3, 5; 303151B Unit 2MU EEC405 Module 4KTU MET453 Module 4
How does a lithium-ion cell work?
Lithium ions shuttle between two electrodes through a liquid electrolyte, while electrons go round the external circuit. On discharge, ions leave the graphite anode and slot into the metal-oxide or phosphate cathode; charging pushes them back. A porous separator keeps the electrodes apart. Nothing is consumed, so a good cell lasts 1,000-3,000+ full cycles; capacity fades slowly as side reactions trap lithium.
Which battery chemistries are used in EVs?
Lead-acid (30-40 Wh/kg) still powers many e-rickshaws because it is cheap. NiMH (60-120 Wh/kg) served early hybrids. Li-ion dominates: NMC and NCA (200-270 Wh/kg cell) for long-range cars; LFP (150-200 Wh/kg) for cost, safety and cycle life, now common in Indian cars, buses and three-wheelers; LTO for very fast charging and 10,000+ cycles at low energy. Sodium-ion and metal-air cells (aluminium-air is a primary, non-rechargeable cell) are emerging.
What do Ah, C-rate, SoC and DoD mean?
Capacity in ampere-hours (Ah) is how much charge the cell delivers; energy (Wh) is capacity times voltage. C-rate is current relative to capacity: 1C empties the cell in an hour, 2C in 30 minutes. State of charge (SoC) is the charge left as a percentage; depth of discharge (DoD) is how much was used (DoD = 100% − SoC). Specific energy (Wh/kg) sets range per kg; specific power (W/kg) sets acceleration and fast charging.
How is a battery pack built from cells?
Cells in series add voltage; strings in parallel add capacity and current. A 96s46p pack of 3.6 V, 5 Ah cells gives 345.6 V and 79.5 kWh. Cells are grouped into modules with busbars, voltage and temperature sensing, and a cooling plate, and the modules go into a sealed, crash-proof enclosure with the BMS, contactors and fuses. Housing and cooling add roughly 30-50% to the cell mass. Build a pack below.
What is a Ragone plot?
A log-log chart of specific power (W/kg) against specific energy (Wh/kg) for storage technologies. Supercapacitors sit at high power and low energy, fuel cells and petrol at high energy, batteries in between, with energy cells and power cells of the same chemistry at different points. It shows why vehicles hybridise storage: no single device gives both.
Cylindrical, prismatic or pouch?
Cylindrical cells (18650, 21700, 4680) are cheap, robust and well-cooled from the sides, but a pack needs thousands. Prismatic cells are large metal cans that pack tightly with fewer connections, common for LFP. Pouch cells are light and space-efficient but need external support against swelling. Cell-to-pack designs skip the module level for more energy per litre.
A 60 Ah cell is discharged at 120 A. Its C-rate and run time are:
120/60 = 2C, so the full capacity lasts 1/2 h = 30 minutes (a little less in practice, because capacity falls at high current).
Worked examples
Pack from 21700 cells · NMC 21700 cells: 3.6 V, 5 Ah, 2C continuous; 96 in series, 46 in parallel
- Voltage: 96 × 3.6 = 345.6 V
- Energy: 96 × 46 × 3.6 × 5 = 79,488 Wh = 79.49 kWh
- Continuous power at 2C: 159 kW
- Current at 150 kW: 150,000/345.6 = 434 A, about 9.4 A per cell
Answer: 346 V, 79.5 kWh, 4,416 cells
Usable energy late in life · Same 79.49 kWh pack; BMS window 90% of capacity; capacity at end of life 80%
- 79.49 × 0.9 × 0.8 = 57.2 kWh
Answer: 57.2 kWh usable at end of life, which is what range should be designed on
Mass for 40 kWh: lithium vs lead · Li-ion pack 160 Wh/kg, lead-acid 35 Wh/kg
- Li-ion: 40,000/160 = 250 kg
- Lead-acid: 40,000/35 = 1143 kg
Answer: 250 kg vs 1,143 kg: why lead-acid stays in low-speed rickshaws
Battery modelling: open-circuit voltage, internal resistance, Thevenin model, Peukert and sizing for a cycle
Which syllabus covers thisAU CME336 Unit IV (analysis)VTU BME306A Module 4 (design parameters of batteries)SPPU 402051E Units 3, 5MU EEC405 Module 4KTU MET453 Module 4
What is the open-circuit voltage curve?
The voltage of a rested cell at each state of charge. For NMC it slopes from about 4.2 V at full to 3.0 V empty, so voltage tells you roughly how full the cell is. LFP is flat at about 3.3 V over most of its range, which makes SoC hard to read from voltage. The OCV curve is the backbone of every battery model and of the BMS's SoC correction.
What is the Thevenin (RC) equivalent circuit of a cell?
An OCV source that depends on SoC, a series resistance R0 for the instant voltage drop, and one or two parallel RC pairs for the slower polarisation that builds up over tens of seconds and relaxes after the load stops. It captures how terminal voltage sags under load, heats the cell (I²R) and recovers afterwards, accurately enough for BMS estimators and drive-cycle simulation. Try a pulse load and a cold cell.
What is Peukert's law?
An empirical rule that a battery delivers less capacity at higher current. With the capacity C rated over H hours and a Peukert exponent k (about 1.1-1.3 for lead-acid, close to 1.05 for lithium-ion), the run time at current I is given below. It explains why a lead-acid e-rickshaw battery loses range quickly when driven hard.
What are the constant-current and fractional depletion models?
Two textbook ways to predict range from a drive cycle. The constant-current approach treats each step as a short constant-current discharge and adds up the capacity it removes, using Peukert's law for the effective capacity at that current. The fractional depletion model tracks the fraction of capacity used step by step (depth of discharge) until it reaches the allowed limit; the distance covered by then is the range.
Why do batteries age, and what is state of health?
Calendar ageing happens even when parked, faster at high temperature and high SoC. Cycle ageing grows with depth of discharge, high C-rates, fast charging and cold charging (lithium plating). State of health (SoH) compares the present capacity, or resistance, with the new value; 70-80% capacity is the usual end of life in a vehicle. Keeping cells between 20% and 80% and below 35 °C makes them last much longer.
A lead-acid battery rated 100 Ah at the 20-hour rate (Peukert k = 1.2) is discharged at 25 A. It lasts about:
t = 20 × (100/(25 × 20))^1.2 = 20 × 0.2^1.2 = 2.90 h, so it delivers only 72.5 Ah, not 100.
Worked examples
IR drop and heat in one cell and in a pack · Cell R0 = 25 mΩ; pack 96s46p of these cells
- Cell at 10 A: ΔV = 10 × 0.025 = 0.25 V, heat 10² × 0.025 = 2.5 W
- Pack resistance: 96 × 0.025/46 = 52.2 mΩ
- Pack at 400 A: ΔV = 20.9 V, heat 8.35 kW
Answer: 21 V of sag and 8.3 kW of heat at full power: why packs need cooling
Peukert capacity · 100 Ah at the 20 h rate, k = 1.2, discharged at 25 A
- t = 20 × (100/500)^1.2 = 2.90 h
- Delivered: 2.90 × 25 = 72.5 Ah
Answer: 72 Ah, 27% less than rated
RC relaxation after a pulse · R1 = 15 mΩ, C1 = 2,000 F, a long 5 A pulse that then stops
- τ = R1 C1 = 30 s
- Polarisation voltage at the end of the pulse: 5 × 0.015 = 75 mV
- It decays to 37% (28 mV) after 30 s and to 5% after 90 s
Answer: The terminal voltage creeps back up for about 1.5 minutes: rest the cell before reading its OCV
Battery size for a range target · 250 km at 0.15 kWh/km; usable window 90%; design for 80% capacity at end of life
- Energy needed: 0.15 × 250 = 37.5 kWh
- Nominal: 37.5 ÷ 0.9 ÷ 0.8 = 52.1 kWh
Answer: About 52 kWh nominal
Battery management systems: protection, SoC estimation, cell balancing and thermal management
Which syllabus covers thisAU CME336 Unit IVVTU BME306A Module 2SPPU 402051E Units 5, 6; 303151B Unit 3MU EEC405 Module 4KTU MET453 Module 4
What does a battery management system do?
It measures every cell's voltage, the pack current and temperatures. It protects the pack by opening contactors on over- or under-voltage, over-current, over-temperature or an insulation fault. It estimates state of charge, state of health and the power available right now, balances the cells, controls the cooling and heating, and talks to the vehicle and charger over CAN. A pack without a BMS is not safe to use.
Centralised or distributed BMS?
A centralised BMS wires every cell sense lead back to one board: cheap for small packs (scooters) but a wiring tangle for big ones. A modular or distributed (master-slave) BMS puts a slave board with an analogue front-end chip on each module to measure its cells, and a master that runs the estimators and decisions, linked by an isolated daisy-chain or CAN. Cars and buses use the distributed layout; wireless BMS designs remove even that harness.
How does a BMS estimate state of charge?
Coulomb counting integrates current over time: accurate in the short term but it drifts with sensor offset. OCV lookup is accurate but only after the cell has rested. Real BMSs combine them, usually with an extended Kalman filter running a Thevenin cell model: it predicts voltage from the counted charge, compares with the measured voltage and corrects the SoC continuously. LFP's flat OCV curve makes this harder.
Why do cells need balancing?
No two cells are identical: capacity, self-discharge and temperature differ slightly. In a series string, charging must stop when the fullest cell reaches its limit and discharge when the emptiest cell does, so a mismatch shrinks the usable capacity a little more every cycle. Passive balancing bleeds the high cells through resistors (cheap, wastes heat); active balancing moves charge from high cells to low ones with converters (efficient, costly). Compare the two.
Why does a battery need thermal management?
Lithium-ion cells are happiest at about 15-35 °C. Hotter, they age fast; above about 60 °C they risk thermal runaway, a self-heating chain reaction that can spread cell to cell. Colder, their resistance rises and charging can plate lithium. Small packs use air cooling; cars and buses use liquid cold plates with a chiller, plus heaters for winter, and keep the cell-to-cell spread under about 5 °C. Packs in India are designed for 45 °C ambient.
A 50 Ah pack at 80% SoC supplies 20 A for 30 minutes, then regenerates 10 A for 6 minutes. Its SoC by coulomb counting is:
Out 20 × 0.5 = 10 Ah, back in 10 × 0.1 = 1 Ah, net 9 Ah = 18% of 50 Ah: 80 − 18 = 62%.
Worked examples
Coulomb counting · 50 Ah pack at 80%; 20 A out for 30 min, then 10 A back in for 6 min (coulombic efficiency 1)
- Removed 20 × 0.5 = 10 Ah; returned 10 × 0.1 = 1 Ah
- Net 9 Ah = 9/50 = 18%
- SoC = 80 − 18 = 62%
Answer: 62%
Passive balancing time and heat · One 5 Ah cell sits 2% above the rest; bleed current 100 mA at about 4.1 V
- Excess charge: 0.02 × 5 = 0.1 Ah
- Time: 0.1/0.1 = 1 h
- Heat: 0.1 Ah × 4.1 V = 0.41 Wh
Answer: 1 hour and 0.41 Wh burnt per cell, usually done near full charge
Pack heating at 200 A without cooling · Pack resistance 52.2 mΩ; 4,416 cells × 69 g, specific heat about 1,000 J/kg·K
- Heat: 200² × 0.0522 = 2087 W
- Cell mass: 4,416 × 0.069 = 305 kg
- Rise: 2087 × 60/(305 × 1,000) = 0.41 K per minute
Answer: About 0.4 K/min: a 20-minute climb adds about 8 K
Coolant flow to remove 2 kW · Water-glycol, cp 3,500 J/kg·K, density 1,040 kg/m³, coolant rise 5 K
- ṁ = 2,000/(3,500 × 5) = 0.114 kg/s
- Volume flow: 0.114/1,040 m³/s = 6.6 L/min
Answer: About 6.6 L/min
Fuel cells, ultracapacitors, flywheels and hybrid energy storage
Which syllabus covers thisAU CME336 Unit IVVTU BME306A Module 2; BEE515C Module 5SPPU 402051E Unit 3; 303151B Unit 2MU EEC405 Module 4KTU MET453: Not in the syllabus
How does a hydrogen fuel cell make electricity?
Hydrogen is split on a platinum catalyst at the anode into protons and electrons. The protons cross a polymer membrane (PEM); the electrons cannot, so they travel through the external circuit and do work. At the cathode the protons, electrons and oxygen from air combine into water. A single cell gives under 1 V, so a car stack has 300-400 cells in series. The only exhaust is water vapour.
What types of fuel cells are there?
Named by electrolyte: PEM (about 80 °C, quick start, used in cars and buses), alkaline (AFC, used in spacecraft, needs CO2-free air), phosphoric acid (PAFC, about 200 °C, stationary power), molten carbonate (MCFC, about 650 °C) and solid oxide (SOFC, 600-1,000 °C, can run on natural gas, stationary). Direct methanol cells (DMFC) run on liquid methanol at low power.
What is a fuel cell polarisation curve?
Cell voltage against current density. It starts below the ideal 1.23 V, drops steeply at low current (activation loss to drive the reactions), then slopes gently (ohmic loss in the membrane) and finally falls off a cliff (concentration loss when reactants cannot reach the catalyst fast enough). Efficiency is roughly cell voltage divided by 1.25 V, so a fuel cell is most efficient at part load, unlike an engine.
How is hydrogen stored on a vehicle?
Mostly as compressed gas at 700 bar (cars) or 350 bar (buses and trucks) in carbon-fibre tanks: about 5-6 kg in a car gives 500-650 km. Liquid hydrogen at −253 °C stores more per litre but boils off and costs energy to liquefy. Metal hydrides absorb hydrogen into a solid at low pressure, safe and compact, but heavy. Hydrogen holds 120 MJ/kg (lower heating value), nearly three times petrol per kg, but very little per litre.
What are ultracapacitors and flywheels good for?
Short, high-power bursts. An ultracapacitor stores charge electrostatically, so it charges and discharges in seconds for a million cycles, but holds only 5-10 Wh/kg. A flywheel stores kinetic energy in a spinning rotor (½Iω²) at similar power. Paired with a battery, they take the acceleration and regenerative-braking peaks, so the battery sees a smoother, smaller current and lasts longer: hybrid energy storage.
A PEM fuel cell running at 0.70 V per cell has an efficiency (on hydrogen's lower heating value) of about:
Efficiency = V/1.253 V = 0.70/1.253 = 55.9%. At light load the voltage rises towards 0.8-0.9 V and efficiency towards 65-70%.
Worked examples
Ideal cell voltage · ΔG = −237.1 kJ/mol, ΔH (LHV) = −241.8 kJ/mol, ΔH (HHV) = −285.8 kJ/mol at 25 °C; 2 electrons per H2
- E0 = 237,100/(2 × 96,485) = 1.229 V
- LHV equivalent: 241,800/(2F) = 1.253 V; HHV: 1.481 V
- At 0.70 V: efficiency 0.70/1.253 = 55.9% (LHV)
Answer: 1.23 V ideal; 56% efficient at 0.7 V
Hydrogen consumption · Stack output 80 kW at 50% efficiency; hydrogen LHV 120 MJ/kg
- Hydrogen power: 80/0.5 = 160 kW
- Flow: 160,000/120 × 10⁶ = 1.333 g/s
- = 4.8 kg/h at full power
Answer: 4.8 kg/h
Stack voltage and current density · 370 cells of 300 cm² at 0.70 V, 80 kW
- Stack voltage: 370 × 0.70 = 259 V
- Current: 80,000/259 = 309 A
- Current density: 309/300 = 1.03 A/cm²
Answer: 259 V, 309 A, 1.03 A/cm²
Ultracapacitor energy · 3,000 F, 2.7 V cells; 100 in series (30 F, 270 V); usable down to half voltage
- One cell: ½ × 3,000 × 2.7² = 10,935 J = 3.04 Wh
- Module: ½ × 30 × 270² = 1.093 MJ = 304 Wh
- Usable from V to V/2: 75% = 228 Wh
Answer: 228 Wh usable: enough for one hard acceleration, not for range
Charging: AC and DC levels, CCS2, CHAdeMO and Bharat standards, CC-CV, swapping and solar
Which syllabus covers thisAU EE3025 Unit V (charging mechanisms)VTU BME306A Module 5; BEE515C Module 1SPPU 402051E Unit 6; 303151B Units 3, 5MU EEC405 Modules 4, 6KTU MET453 Module 4
What is the difference between AC and DC charging?
With AC charging the car's own on-board charger converts grid AC to DC, so power is limited by that charger: typically 3.3-11 kW (22 kW three-phase on some cars). With DC fast charging the conversion happens in the station, which feeds DC straight to the battery through the BMS-controlled contactors at 25-350 kW. AC suits overnight home and office charging; DC suits highways and fleets.
What are charging levels 1, 2 and 3?
Level 1 is a household socket (in India 230 V, 15 A, about 3 kW; in the US 120 V, about 1.9 kW). Level 2 is a dedicated AC wallbox at 7-22 kW. Level 3 means DC fast charging. Two-wheelers and three-wheelers mostly charge from a 15 A socket with a portable charger, or swap batteries.
Which connectors and standards does India use?
Cars in India have converged on CCS2 (the European Combined Charging System, Type 2 AC plus two DC pins), with some CHAdeMO (Japanese DC) and Type 2 AC wallboxes; China uses GB/T and North America J1772 and NACS. India's early public chargers followed the government's low-voltage Bharat AC-001 and Bharat DC-001 specifications for small cars and three-wheelers, and light EVs now have their own LEVAC/LECCS connector standards. IEC 60309 is the industrial socket used by AC-001.
Why does fast charging slow down after 80%?
Lithium-ion cells are charged constant-current, constant-voltage (CC-CV). The charger holds a fixed current until the cell voltage reaches its limit (4.2 V for NMC), then holds that voltage while the current tapers off. Forcing current in faster at high SoC would overshoot the voltage limit and plate lithium. So the first 70-80% goes in quickly and the last 20% takes about as long again. Change the charger and the pack size below.
What about battery swapping, wireless and solar charging?
Swapping replaces an empty pack with a full one in about two minutes; it works well for e-rickshaws and delivery scooters with standard packs, and India drafted a battery-swapping policy for them. Wireless (inductive) charging transfers power across an air gap between a ground pad and a car coil at about 90% efficiency. Solar carports or rooftops can charge EVs directly; an MPPT controller keeps the panels at their maximum power point.
A 60 kWh car charges from 10% to 80% on a 50 kW DC charger that holds full power the whole time. It takes about:
70% of 60 kWh = 42 kWh; 42/50 h = 50.4 min. Real sessions take a little longer because power tapers near 80% and the battery may be cold.
Worked examples
Home AC charging time · 40 kWh battery from 20% to 80% on a 7.4 kW wallbox, 90% charging efficiency
- Energy into the battery: 0.6 × 40 = 24 kWh
- Time: 24/(7.4 × 0.9) = 3.60 h
Answer: About 3.6 hours, easily overnight
DC fast-charge current and time · 50 kW charger, pack at 360 V; 60 kWh pack from 10% to 80%, full power held
- Current: 50,000/360 = 139 A
- Energy: 0.7 × 60 = 42 kWh → 50.4 min
Answer: 139 A, about 50 min
Solar charging a car · 3 kWp rooftop, 4.5 peak-sun hours a day, 80% system performance ratio, car uses 0.14 kWh/km
- Energy: 3 × 4.5 × 0.8 = 10.8 kWh/day
- Distance: 10.8/0.14 = 77 km/day
Answer: About 77 km a day, more than most daily commutes
Smart charging and the grid: G2V, V2G, V2H, V2V, aggregators and charging-station capacity
Which syllabus covers thisAU EE3025 Unit VVTU BME306A Module 5SPPU 402051E Unit 6; 303151B Unit 6MU EEC405 Module 6KTU MET453 Module 4
What are G2V, V2G, V2H and V2V?
Grid-to-vehicle (G2V) is ordinary charging. Vehicle-to-grid (V2G) sends energy from the car's battery back into the grid at peak times or to steady its frequency. Vehicle-to-home (V2H) powers a house during an outage or the evening peak. Vehicle-to-vehicle (V2V) charges one EV from another. All except G2V need a bidirectional charger and communication between car, charger and grid operator.
What is smart charging?
Deciding when and how fast each car charges instead of starting at full power the moment it is plugged in. If most cars plug in after work, unmanaged charging lands right on the evening peak and can overload the local transformer. Smart charging shifts it into the night valley within each driver's departure deadline, guided by time-of-day tariffs or a central signal. V2G goes further and shaves the peak. Compare the three on a feeder.
What does an aggregator do in V2G?
One car is too small for the grid operator to deal with. An aggregator bundles thousands of plugged-in EVs into one virtual power plant, bids its capacity into markets for frequency regulation, spinning reserve or peak shaving, sends charge and discharge commands to individual chargers (through protocols like OCPP and ISO 15118) and pays the owners. It must respect each car's minimum SoC at departure and battery-wear limits.
How is a charging station's capacity estimated?
Start from the expected sessions per day, the energy per session and the opening hours to get the average power, then divide by charger rating times a realistic utilisation (rarely above 50-60%) to get the number of chargers. Size the transformer for all chargers at full power with a diversity factor, and check the grid connection, harmonics (filters or active front ends), power factor, earthing and space for queues. Most Indian DC stations need a dedicated distribution transformer.
What is smart metering and why does it matter for EVs?
A smart meter records energy every 15-30 minutes and talks to the utility, so it can bill time-of-day tariffs and carry demand-response signals. That is what makes cheaper night charging, smart charging and paid V2G services possible. India's smart meter rollout under the Revamped Distribution Sector Scheme is therefore an EV enabler as much as a billing upgrade.
Forty homes on one feeder each plug in a 7.4 kW EV at 7 pm. Unmanaged, the extra load at 7 pm is about:
40 × 7.4 = 296 kW at once, on top of the evening peak, enough to overload a typical feeder transformer. Spread through the night, the same energy needs far less peak capacity.
Worked examples
Chargers for a fleet depot · 20 cars a day × 25 kWh each, 10 operating hours, 50 kW DC chargers at 60% utilisation
- Average power: 20 × 25/10 = 50 kW
- Chargers: 50/(50 × 0.6) = 1.67
Answer: 2 chargers (round up), with a transformer for 100 kW plus margin
Vehicle-to-home through an evening · 60 kWh EV at 80% (48 kWh) keeps 50% (30 kWh) for tomorrow; 90% inverter; house uses 2 kW
- Available from the battery: 48 − 30 = 18 kWh
- Delivered to the house: 18 × 0.9 = 16.2 kWh
- At 2 kW: 8.1 hours
Answer: About 8 hours of household power
Hybrid control and energy management: rule-based and optimal strategies, regenerative braking
Which syllabus covers thisAU CME336 Unit V; EE3025 Unit IVVTU BME306A Module 2; BEE515C Module 4SPPU 402051E Unit 2; 303151B Unit 4MU EEC405 Module 6KTU MET453 Module 1
What does a hybrid's supervisory controller decide?
Every few milliseconds it reads the driver's pedal, speed and battery SoC, and decides the mode (electric only, engine only, both, charging, regenerating), how to split the requested power between engine and motor, and whether to start or stop the engine. Lower-level controllers then make the engine, motor and brakes deliver what it asked for. The split decides fuel economy, emissions and battery wear.
What are rule-based energy management strategies?
Simple if-then rules, tuned by engineers. The thermostat (on/off) strategy runs the engine at its single best-efficiency point whenever SoC falls below a low limit and switches it off above a high limit; the battery absorbs the difference. The power follower runs the engine to follow the driver's demand when it is high enough, with the battery filling gaps and keeping SoC in a band. They are robust and run on any ECU. Compare them over a city cycle.
What are optimisation-based strategies?
Dynamic programming finds the truly optimal split for a known drive cycle by searching backwards through every SoC: too slow for a car but the benchmark for others. The equivalent consumption minimisation strategy (ECMS) works in real time: it converts battery power into equivalent fuel with a factor s and picks the split that minimises fuel plus equivalent fuel at each instant. Model predictive control optimises over a short predicted horizon (from navigation data).
How much braking energy can be recovered?
Only what reaches the axle with the motor, within the motor's torque and power limits and the battery's charge-acceptance limit. Under braking, weight shifts forward, so the front axle does 60-70% of the braking: a front motor can recover more than a rear one. The friction brakes take whatever the motor cannot, blended seamlessly by the brake controller. Around 60-70% of braking energy can be returned in city driving, much less in emergency stops. Try a hard stop.
What else does the controller manage?
Drivability and noise: smooth engine starts (the motor spins the engine up), no torque bumps during mode changes, damping of driveline oscillation, and engine speed kept low when cabin noise matters. Emissions: keeping the catalyst warm in a hybrid whose engine stops often. And battery life: limiting SoC swings and cell temperature.
A 1,500 kg car brakes at 0.5 g (CG 0.55 m high, 1.43 m ahead of the rear axle, wheelbase 2.6 m). The share of its weight on the front wheels during the stop is about:
F_zf = m(g b + a h)/L = 1,500 × (9.81 × 1.43 + 4.905 × 0.55)/2.6 = 9650 N, 66% of the weight (55% when standing).
Worked examples
Braking energy per stop · 1,500 kg car stopping from 60 km/h; 60% returned to the battery
- Kinetic energy ½ × 1,500 × 16.67² = 208.3 kJ = 57.9 Wh
- Recovered: 0.6 × 57.9 = 34.7 Wh
Answer: 35 Wh per stop, about a quarter of a kilometre's driving
Load transfer under braking · a = 0.5 g = 4.905 m/s², h = 0.55 m, b = 1.43 m (CG to rear axle), L = 2.6 m
- F_zf = 1,500 × (9.81 × 1.43 + 4.905 × 0.55)/2.6 = 9650 N
- Share 65.6% of the 14,715 N weight
Answer: 66% of the braking should happen on the front axle
ECMS equivalent fuel · Battery supplying 10 kW, equivalence factor s = 2.5, petrol LHV 43 MJ/kg
- ṁ_eq = 2.5 × 10,000/43 × 10⁶ = 0.581 g/s
Answer: 0.58 g/s of 'virtual' fuel, compared against real fuel to choose the split
When regeneration hits its limit · Stop from 100 km/h at 3 m/s²; battery accepts at most 80 kW
- Braking power at the start: 1,500 × 3 × 27.78 = 125 kW
- Motor takes 80 kW (less after losses); friction brakes take the remaining 45 kW
Answer: Regeneration covers about two-thirds at the start and all of it below about 64 km/h
Sizing an EV powertrain end to end: motor power, gear ratio, battery and power electronics
Which syllabus covers thisAU CME336 Unit V; EE3025 Unit IIIVTU BME306A Module 4; BEE515C Module 2 (power rating design)SPPU 402051E Units 3, 4MU EEC405 Module 5KTU MET453 Module 3
In what order is an EV powertrain sized?
Start from the vehicle targets: mass, top speed, 0-100 km/h time, gradeability, range and payload. Top speed sets the continuous power; acceleration sets the peak power; gradeability and launch set peak wheel torque; the motor's maximum speed and top speed set the gear ratio; range on the target drive cycle sets battery energy; peak power and voltage set battery current, cell count and the inverter rating. Then iterate, because the battery's mass changes everything.
How do you estimate the power needed for acceleration?
A standard closed-form estimate (from Ehsani's textbook) for accelerating from rest to v_f in t_a with a motor whose base speed is v_b adds the inertia term to the average rolling and drag power over the run. It shows why a motor with a long constant-power range (small v_b) needs less peak power for the same 0-100 km/h time.
How do you choose battery voltage and configuration?
Higher voltage means less current for the same power, so thinner cables, smaller contactors and lower I²R loss: two-wheelers use 48-72 V, cars 350-400 V and newer cars 800 V for faster DC charging. The series count sets the voltage; the parallel count then comes from the energy needed, and must also keep the peak current per cell within its C-rate limit at the motor's peak power and at the fastest charge rate.
How is a hybrid's engine and motor matched?
In a series hybrid the engine-generator is sized for the average power of the hardest sustained duty (highway or a long grade) and the motor for peak power. In a parallel hybrid the engine typically covers cruising power with the motor adding acceleration power. The battery in a hybrid is sized by power, not energy: it must absorb and deliver the peaks, with only a narrow SoC window used.
Which target usually sets an EV's peak motor power?
Cruising at 150 km/h needs about 37 kW here, but reaching 100 km/h in 9 s needs about 85 kW: acceleration dominates. Range sets the battery, not the motor.
Worked examples
Continuous power for 150 km/h · 1,500 kg, Crr 0.010, CdA 0.66 m², driveline 95%
- Road load: 147.2 + ½ × 1.2 × 0.66 × 41.67² = 834.6 N
- Motor power: 834.6 × 41.67/0.95 = 36.6 kW
Answer: About 37 kW continuous
Peak power for 0-100 km/h in 9 s · δ = 1.05, motor base speed one-third of 100 km/h (x = 3)
- Inertia term: 1.05 × 1,500/18 × (27.78² + 9.26²) = 75.0 kW
- Rolling + drag terms: 6.1 kW
- At the wheel 81.1 kW; motor 85.4 kW
Answer: A motor of about 90 kW peak
Torque for a 30% grade and the gear ratio · Gear 9:1 at 95%, r = 0.31 m; motor maximum 14,000 rpm for 150 km/h
- Grade force: 14,715 × (sin 16.7° + 0.010 cos 16.7°) = 4369 N
- Motor torque at 9:1: 4369 × 0.31/(9 × 0.95) = 158 N·m
- Highest ratio for 150 km/h: 10.91, so 9:1 fits
Answer: At least 158 N·m; ratio 9:1
Battery for 300 km · 150 Wh/km on the cycle incl. auxiliaries, 90% usable window; NMC cells 3.6 V, 5 Ah, 2C; 96 in series
- Nominal energy: 300 × 0.150/0.9 = 50 kWh
- Parallel count: 50,000/(96 × 3.6 × 5) = 28.9 → 29; pack 50.1 kWh at 345.6 V
- Peak current: 85.4 kW/0.97 inverter/345.6 V = 255 A = 8.8 A per cell = 1.8C, inside 2C
Answer: 96s29p, 50 kWh
Vehicle design, high-voltage safety, CAN networks, testing and homologation (AIS-038, AIS-156)
Which syllabus covers thisAU EE3025 Unit I (details and specifications)VTU BEE515C Module 3SPPU 402051E Units 1, 5MU EEC405 self-study (ARAI standards)KTU MET453 Module 5
How does an EV chassis differ from a petrol car's?
The battery is the heaviest single part and has to be protected in a crash, so it sits low between the axles inside a strong frame (side sills, cross members) away from crush zones. That lowers the centre of gravity and improves handling and rollover resistance. Without an engine the front can become a crumple zone and storage. Retrofitting an IC car to electric must recheck weight distribution, brakes, suspension and crash safety.
What are the high-voltage safety rules in an EV?
Anything above 60 V DC is treated as hazardous. HV parts are enclosed (IP67 for the pack), cables are orange, and the HV bus is isolated from the chassis. An insulation monitor must see at least 100 Ω per volt on DC circuits (500 Ω/V on unprotected AC), so 40 kΩ on a 400 V system. Workers use insulated tools and gloves and must disconnect, lock out and verify zero voltage before touching anything.
How do the controllers in an EV talk to each other?
Over CAN (Controller Area Network): a two-wire differential bus at 500 kbit/s (CAN-FD up to several Mbit/s in the data phase). The vehicle control unit, BMS, motor controller, charger and dashboard broadcast messages with IDs; when two transmit at once, the lower ID wins arbitration bit by bit without losing time. Engineers test these networks with Vector CANoe/CANalyzer and simulate missing ECUs during integration.
What tests must an EV pass before sale in India?
Type approval under the Central Motor Vehicle Rules by an agency such as ARAI (Pune) or ICAT (Manesar), to Automotive Industry Standards (AIS). AIS-038 Rev 2 covers electric powertrain safety for cars, buses and trucks; AIS-156 covers the same for two- and three-wheelers, with stricter battery tests added in 2022-23 after scooter fires (thermal propagation, ingress, cell-level data). Energy consumption and range are measured to AIS-039 and AIS-040, plus crash, braking and EMC tests.
Where do sensors and autonomy come in?
EVs are natural platforms for driver assistance and self-driving because every actuator (steering, brakes, motor) is already electrically controlled. Cameras, radar, ultrasonic sensors and lidar feed perception software; the risks are sensor failure and cyber attacks on connected vehicles, which is why functional safety (ISO 26262) and cybersecurity (ISO/SAE 21434, UN R155) processes are now required by carmakers.
On a 500 kbit/s CAN bus, a standard frame with 8 data bytes (no stuff bits) takes about:
108 bits for the frame plus 3 bits of interframe space = 111 bits × 2 µs = 222 µs, so the bus carries at most about 4,500 such frames a second.
Worked examples
CAN frame time and bus capacity · CAN 2.0A frame: SOF 1, ID 11, RTR 1, IDE 1, r0 1, DLC 4, data 64, CRC 15, delimiter 1, ACK 2, EOF 7, interframe 3; 500 kbit/s
- Bits: 111
- Time: 111 × 2 µs = 222 µs
- Maximum: 4,505 frames/s; real buses are kept under about 50% load
Answer: 222 µs per frame (bit stuffing adds up to about 20%)
Insulation resistance limit · 400 V DC bus, 100 Ω/V requirement
- 400 × 100 = 40,000 Ω
- Leakage at that limit: 400/40,000 = 10 mA, the alarm point
Answer: At least 40 kΩ between the HV bus and the chassis
How a floor battery lowers the centre of gravity · Body and rest 1,100 kg at 0.65 m; battery 400 kg at 0.30 m
- h = (400 × 0.30 + 1,100 × 0.65)/1,500 = 0.557 m
Answer: 0.56 m, about 9 cm lower than the body alone
Keep going
- All Skill-Lync electric vehicle courses compared
- EV Powertrain: Component Sizing and Specification course
- Introduction to Battery Technology for Electric Vehicles course
- Pre-Graduate Program in Hybrid Electric Vehicles
- Free lessons: EV fundamentals
- EV engineer salaries from Skill-Lync placements
- Companies hiring EV engineering freshers
- Power Electronics interactive book
- Electrical Machines interactive book
- Automobile Engineering interactive book
- All Skill-Lync interactive books
Recommended reading
- Modern Electric, Hybrid Electric and Fuel Cell Vehicles, M. Ehsani, Y. Gao, S. Longo and K. Ebrahimi
- Electric and Hybrid Vehicles: Design Fundamentals, I. Husain
- Electric Vehicle Technology Explained, J. Larminie and J. Lowry
- Electric and Hybrid Vehicles, T. Denton
- Energy Systems for Electric and Hybrid Vehicles, K. T. Chau (ed.)
- Batteries for Electric Vehicles, D. A. J. Rand, R. Woods and R. M. Dell
- Hybrid Electric Vehicles: Principles and Applications with Practical Perspectives, C. Mi, M. A. Masrur and D. W. Gao
- Energy Management Strategies for Electric and Plug-in Hybrid Electric Vehicles, S. S. Williamson
How this book was made
Chapters follow the official syllabi listed above. Every interactive's formulas and every worked-example number were checked by script. Explanations, animations and examples are original to Skill-Lync. Topics informed by MIT OpenCourseWare 2.627 (Fundamentals of Photovoltaics), CC BY-NC-SA. All explanations, animations and examples are original to Skill-Lync; MIT does not endorse this page.