Power Electronics for EVs Interview Questions & Answers

eMobility Academy — Power Electronics for EVs interview questions

1. Q1.What power converters are found in an EV?

   Fresher

   A typical electric car has: the traction inverter (DC to three-phase AC for the motor, bidirectional for regen); the on-board charger (OBC), which converts grid AC to DC to charge the pack, usually 3.3 kW, 7.4 kW or 11 kW; the DC-DC converter that steps HV down to 12 V to run auxiliaries and charge the 12 V battery, replacing the alternator; and power stages for the electric AC compressor and PTC heater. An off-board DC fast charger contains its own AC-DC and isolated DC-DC stages. In two-wheelers the charger is often off-board, a portable brick, and a small DC-DC converter supplies 12 V lights and electronics from the 48 to 72 V pack.

2. Q2.Explain buck, boost and buck-boost converters.

   Fresher

   A buck converter steps voltage down. A switch chops the input, an inductor and capacitor filter it, and in continuous conduction Vout = D × Vin, where D is duty cycle. A boost converter steps voltage up: the inductor stores energy when the switch is on and releases it to the output in series with the input when it is off, giving Vout = Vin / (1 − D). A buck-boost (inverting) gives Vout = −D/(1 − D) × Vin, and non-inverting four-switch versions can do both step-up and step-down with the same polarity. These are non-isolated topologies. In EVs, buck converters are used for LV rails, and boost stages appear in PFC circuits and in some systems that raise battery voltage for the inverter.

3. Q3.What is power factor correction and why does an on-board charger need it?

   Fresher

   A simple diode bridge with a large capacitor draws current from the grid in short, high peaks near the voltage peak, which gives a poor power factor and high harmonic distortion, wastes distribution capacity and violates grid harmonic standards such as IEC 61000-3-2 and 61000-3-12. A PFC stage, usually an active boost converter after the rectifier, shapes the input current to follow the sinusoidal input voltage so power factor approaches 0.99 and THD is low. It also produces a regulated DC bus, typically around 380 to 400 V for single-phase, which feeds the isolated DC-DC stage. Higher-efficiency OBCs use totem-pole bridgeless PFC with GaN or SiC to eliminate the diode bridge losses.

4. Q4.Why does an on-board charger need galvanic isolation?

   Fresher

   The traction battery must be isolated from the grid so a fault cannot put mains-referenced voltage onto the vehicle's HV system, and so the vehicle's insulation monitoring and protection concepts remain valid. Isolation is achieved with a high-frequency transformer in the DC-DC stage after PFC. Operating the transformer at tens to hundreds of kHz makes it far smaller and lighter than a 50 Hz transformer. Common isolated topologies are the LLC resonant converter for unidirectional chargers and the dual active bridge (DAB) or CLLC for bidirectional chargers that support V2L or V2G. Standards such as IEC 61851 and the vehicle safety standards make isolation essential.

5. Q5.What are SiC and GaN and why are they important for EVs?

   Fresher

   Silicon carbide (SiC) and gallium nitride (GaN) are wide-bandgap semiconductors: about 3.2 eV for 4H-SiC and 3.4 eV for GaN versus 1.12 eV for silicon. This gives a higher breakdown field, so devices can be thinner with lower on-resistance at a given voltage, faster switching with lower switching losses, and better high-temperature operation. SiC MOSFETs (650 V to 1200 V and above) suit traction inverters, especially at 800 V, and high-power chargers. GaN HEMTs (mostly up to 650 V) excel at very high switching frequencies, enabling compact on-board chargers and DC-DC converters. The result is higher efficiency, smaller passives and less cooling, at higher device cost and with more demanding gate drive and EMI design.

6. Q6.What are the conduction and switching losses in a MOSFET?

   Fresher

   Conduction loss occurs when the device is on: P = I²rms × Rds(on), and Rds(on) rises with temperature, often 1.5 to 2 times from 25°C to 150°C for silicon. Switching loss occurs during each turn-on and turn-off transition, when voltage and current overlap: approximately P = 0.5 × V × I × (tr + tf) × fsw, plus capacitive loss from the output capacitance Coss and reverse-recovery loss from the body diode or freewheeling diode. Gate drive loss is Qg × Vgs × fsw. Increasing switching frequency shrinks inductors and capacitors but raises switching loss, which is the central trade-off. Soft-switching topologies such as LLC and phase-shifted full bridge reduce switching loss by achieving zero-voltage switching.

7. Q7.What does a DC-DC converter in an EV do and what topology is used?

   Fresher

   The LV DC-DC converter steps the traction battery voltage, say 300 to 450 V, down to about 13.5 to 14.5 V to supply 12 V loads and keep the 12 V battery charged, typically rated 1.5 to 3 kW in cars. It must be isolated, so common topologies are the phase-shifted full bridge with synchronous rectification on the low-voltage side, or an LLC resonant converter. The output side handles very high current, around 150 to 250 A, so synchronous rectification with low-Rds(on) MOSFETs and careful busbar design are critical. In two-wheelers, a smaller non-isolated or isolated converter from the 48 to 72 V pack supplies 12 V for lights, horn and controllers.

8. Q8.What does a gate driver do and why is it important?

   Fresher

   A gate driver converts the microcontroller's low-power PWM signal into a strong, fast gate voltage and current that charges and discharges the power device's gate capacitance quickly. For high-side switches it also provides level shifting or isolation, using digital isolators, transformers or optocouplers, plus an isolated supply. Good gate drivers include undervoltage lockout, desaturation or overcurrent detection with soft turn-off, Miller clamping to prevent false turn-on from dv/dt, and sometimes negative turn-off voltage for SiC. Gate resistor selection sets switching speed: too fast causes ringing and EMI, too slow raises switching loss. Poor gate drive is a common root cause of device failures in inverters.

9. Q9.Compare LLC resonant and dual active bridge converters for a bidirectional OBC.

   Experienced

   An LLC converter uses a resonant tank (series inductor, capacitor and the transformer's magnetising inductance) and frequency control to achieve zero-voltage switching on the primary and zero-current switching on secondary rectifiers. It is very efficient near its resonant point but has a limited voltage gain range, which is a challenge with wide battery voltage ranges, and it is inherently unidirectional unless made symmetric as a CLLC. A DAB uses two active full bridges and phase-shift control to transfer power in either direction, achieves ZVS over a wide range with suitable modulation (extended or triple phase shift), and handles wide voltage ranges more gracefully, but has higher circulating current at light load. For V2G-capable OBCs, CLLC and DAB are the main choices.

10. Q10.How do you approach thermal design for a power converter?

    Experienced

    First build a loss budget: conduction and switching losses for each device across operating points, plus magnetics core and copper loss, capacitor ESR losses and busbar losses. Then build a thermal resistance network from junction to case, case to heatsink through the thermal interface material, and heatsink to coolant or air. Check that the junction temperature stays below the device rating, typically designed to 125 to 150°C maximum for silicon with margin, at worst-case ambient (45 to 50°C for Indian chargers), or coolant inlet temperature. Validate with thermal simulation and thermocouple or NTC measurements on the prototype. Also consider thermal cycling reliability: repeated swings in junction temperature cause solder fatigue and bond-wire lift-off.

11. Q11.What EMC challenges exist in EV power electronics?

    Experienced

    Fast-switching devices create high dv/dt and di/dt, which generate conducted and radiated emissions that can disturb radios, sensors and CAN communication. Key paths are common-mode current through parasitic capacitances to chassis, such as device-to-heatsink and motor winding-to-frame, and differential-mode ripple on the DC bus. Mitigations: tight commutation loop layout to minimise stray inductance, Y-capacitors and common-mode chokes, shielded HV cables bonded 360° at both ends, careful grounding, gate resistor tuning and spread-spectrum switching. Components are typically tested to CISPR 25, and vehicles in India must meet the EMC requirements in AIS-004 (Part 3). Note that Y-capacitors raise touch current and energy stored to chassis, so they are limited for safety.

12. Q12.Why does a DC fast charger need its own isolation and what is a typical architecture?

    Experienced

    A DC fast charger connects grid-referenced AC to the vehicle battery directly, bypassing the OBC, so it must provide galvanic isolation between grid and vehicle, insulation monitoring of the DC output, and protection. A typical architecture is modular: several power modules of 15 to 40 kW each, combined in parallel. Each module contains a three-phase active front end (Vienna rectifier or two-level PFC) that produces a DC bus of around 700 to 800 V, followed by an isolated DC-DC stage (LLC or phase-shifted full bridge) with a wide output range, often 150 to 1000 V, to suit both 400 V and 800 V vehicles. A controller manages the vehicle communication, module sharing and the OCPP link to the back end.

13. Q13.How do you size the DC-link capacitor in a traction inverter?

    Experienced

    The DC-link capacitor must handle the ripple current drawn by the inverter's switching, keep voltage ripple within limits, and survive thermally for the vehicle's lifetime. The RMS ripple current for a three-phase inverter is a significant fraction of phase current, roughly up to about 0.65 times the RMS phase current depending on modulation index and power factor, and that usually sizes the capacitor more than voltage ripple does. Film capacitors are preferred over electrolytics in traction inverters because of their high ripple current capability, low ESR, self-healing and long life. Also consider the resonance between the capacitor, busbar inductance and battery cable inductance, and the pre-charge time constant.

14. Q14.What is zero-voltage switching and how is it achieved?

    Experienced

    ZVS means turning a switch on when the voltage across it is already zero, so the energy in its output capacitance (0.5 × Coss × V²) is not dissipated and turn-on loss is almost eliminated. It is achieved by letting an inductive current discharge the switch's Coss during the dead time before turn-on, so the body diode conducts and the switch then turns on at near zero volts. In an LLC converter, operating above resonance with enough magnetising current provides ZVS. In a phase-shifted full bridge, leakage or an external resonant inductor does it, though the lagging leg loses ZVS at light load. In a DAB, phase shift and modulation determine the ZVS range. ZVS allows high switching frequency and smaller magnetics.

15. Q15.How would you investigate a field failure of power MOSFETs in a two-wheeler motor controller?

    Experienced

    Start with the failure pattern: which leg, high or low side, and under what condition (hill climb, regen, cold start, after water exposure). Retrieve controller logs for current, bus voltage and temperature before the event. Physically inspect for burn marks, cracked solder and moisture ingress, and decap or X-ray failed devices if needed. Common root causes: avalanche from voltage spikes during regen with an open battery or long cables, so check bus voltage overshoot and snubbers; shoot-through from insufficient dead time or Miller turn-on; thermal overstress from inadequate heatsinking or poor current sharing between paralleled MOSFETs; and gate driver supply issues. Reproduce on a dyno with a high-bandwidth scope and differential probes, then fix and verify with a design change.

16. Q16.What is a Vienna rectifier and why is it used in DC chargers?

    Experienced

    The Vienna rectifier is a three-phase, three-level, unidirectional boost-type PFC rectifier. It uses one bidirectional switch per phase (or a pair of switches) with fast diodes and a split DC bus with a midpoint. Advantages: each switch sees only half the DC bus voltage, so lower-voltage, faster devices can be used; the three-level operation reduces input current ripple and inductor size; and it achieves near-unity power factor with low THD and high efficiency, often above 98%. Its limitation is that it is unidirectional, so for V2G-capable chargers a two-level or three-level active front end such as a T-type or NPC converter is used instead. It is common in DC fast charger power modules.

17. Q17.How does a bidirectional OBC enable V2L and V2G?

    Experienced

    A bidirectional OBC has an active front end in place of a diode-bridge PFC (for example a totem-pole or full-bridge converter) and a bidirectional isolated DC-DC stage (DAB or CLLC). In V2L (vehicle-to-load), the converter runs as an inverter, generating a stable 230 V, 50 Hz output from the battery to power appliances; several cars sold in India offer V2L. In V2G (vehicle-to-grid), the converter synchronises with the grid and exports power under external control, which requires grid-code compliance, anti-islanding protection and communication with the charger and utility, typically through ISO 15118-20 for AC bidirectional. The power electronics are the easier part; regulation, metering and battery-warranty implications are the harder ones.

18. Q18.What does double-pulse testing tell you?

    Experienced

    The double-pulse test characterises a power device's switching behaviour under real voltage and current. With an inductive load, the first pulse ramps the inductor current to the target value; at turn-off you measure turn-off energy, voltage overshoot and tail current. After a short gap, the second pulse turns the device on at that current, letting you measure turn-on energy and the reverse recovery of the complementary diode or body diode. Results guide gate resistor selection, dead time, snubber design and loss modelling. Measurement quality matters: use high-bandwidth differential voltage probes, a coaxial shunt or a Rogowski coil for current, and deskew probes; with SiC and GaN, poor probing easily gives misleading ringing and loss figures.

More free EV career resources: emobility.academy/career-resources