Power Supplies and Power Electronics

Introduction and Context

Every mechatronic system requires electrical energy in forms suitable for sensors, controllers, actuators, communication devices, and safety circuits. A source such as the mains supply, a battery, or a photovoltaic panel rarely provides the exact voltage, current, isolation, regulation, and transient performance required by every subsystem. Power supplies and power-electronic converters perform this energy transformation.

Power electronics is the controlled conversion of electrical power using semiconductor switches, diodes, capacitors, inductors, transformers, and control circuits. Unlike small-signal electronics, where voltage gain and signal fidelity are usually dominant, power electronics must simultaneously manage efficiency, heat, electromagnetic interference, reliability, isolation, fault energy, and user safety.

This chapter develops the principles needed to design and evaluate linear regulators, switched-mode power supplies, transistor drive circuits, rectifiers, converters, inverters, and safe motor-actuation interfaces.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Explain the operation and limitations of rectifiers, filters, linear regulators, and switching converters.
2. Calculate ripple voltage, regulator power loss, efficiency, duty ratio, and approximate thermal requirements.
3. Select and drive power transistors for low-side and high-side switching.
4. Distinguish isolated from non-isolated power conversion and identify appropriate protection methods.
5. Analyze buck, boost, buck–boost, and inverter circuits under stated assumptions.
6. Design a basic power supply and motor interface while considering safety, heat dissipation, EMC, and reliability.

1. Essential Power-Supply Terminology

A power supply converts an input source into a specified output voltage and current. Important specifications include:

• Nominal output voltage: the intended operating voltage.
• Line regulation: output change caused by input-voltage variation.
• Load regulation: output change caused by load-current variation.
• Ripple: periodic residual variation superimposed on a DC output.
• Transient response: the response to a rapid change in load or input.
• Efficiency: useful output power divided by input power.
• Isolation: absence of a direct conductive path between input and output.
• Dropout voltage: minimum input-to-output voltage required by a regulator.
• Inrush current: temporary input current when capacitors are initially charged.
• Switching frequency: frequency at which a power semiconductor is intentionally switched.

For a DC supply, output power is

where V_out is output voltage in volts, I_out is output current in amperes, and P_out is output power in watts. Efficiency is

where P_in is input power. This definition applies to any converter when input and output powers are measured under the same operating condition.

2. From AC Mains to Regulated DC

A traditional linear supply commonly contains a transformer, rectifier, smoothing capacitor, regulator, and protection components. The transformer can reduce voltage and provide galvanic isolation. A bridge rectifier converts both half-cycles of AC into pulsating DC. The capacitor then stores charge and supplies the load between rectifier peaks.

For a full-wave rectifier supplied from an AC source of frequency f_ac, the dominant ripple frequency is

This assumes a single-phase full-wave circuit and an approximately steady load.

If a capacitor-input filter supplies an approximately constant load current, the peak-to-peak ripple is approximated by

where I_load is load current in amperes, C is capacitance in farads, and ΔV_r is ripple voltage in volts. The approximation assumes the capacitor discharges nearly linearly between charging peaks and that the ripple is small relative to the DC voltage.

The approximate minimum capacitor voltage is

where V_peak is the rectified source peak and V_drops is the total instantaneous rectifier voltage drop. A regulator must receive sufficient voltage at this minimum point to maintain regulation.

Worked Example 1: Filter Capacitor

A full-wave rectifier operates from 50 Hz mains. The load current is 0.50 A and the permitted ripple is 2 V peak-to-peak. Since f_r = 2 × 50 = 100\ Hz,

Thus, a minimum idealized capacitance of 2,500 μF is indicated. A practical design would select a higher standard value after considering capacitor tolerance, temperature, ripple-current rating, lifetime, and the regulator’s minimum input voltage.

A capacitor-input rectifier draws narrow, high-amplitude current pulses. Consequently, the transformer, diode bridge, fuse, and capacitor must be rated for peak and RMS current, not only average load current.

3. Linear Voltage Regulators

A linear regulator controls a series transistor so that the output remains nearly constant. The transistor operates in its active region and behaves approximately as a voltage-controlled variable resistance. Linear regulators are electrically simple and produce low output ripple and low conducted switching noise.

For an idealized linear regulator, power dissipated in the pass element is

This assumes that regulator quiescent current is negligible. Including quiescent current I_q, a more complete estimate is

The approximate efficiency, neglecting control-current effects, is

A linear regulator is attractive when the voltage difference and output power are small, or when low noise is especially important. Its principal disadvantage is heat. For example, reducing 24 V to 5 V at 1 A dissipates approximately 19 W, which generally requires substantial thermal management.

Regulators may be fixed-output or adjustable. An adjustable regulator often uses a reference voltage and a resistor divider. Practical designs must include input and output bypass capacitors of the manufacturer-specified type and value, observe stability requirements, and protect against reverse current where necessary.

Thermal Design

A semiconductor junction must remain below its maximum permitted temperature. A first-order thermal model is

where T_j is junction temperature in °C, T_a is ambient temperature in °C, and θ_JA is junction-to-ambient thermal resistance in °C/W. For a package mounted on a heatsink,

where the terms represent junction-to-case, case-to-sink, and sink-to-ambient thermal resistance. These equations are steady-state approximations; transient thermal impedance, airflow, enclosure temperature, and mounting conditions can materially change performance.

The design must satisfy

with a suitable engineering margin. Thermal interface materials reduce contact resistance but do not eliminate the need for correct mechanical mounting.

4. Switching Regulators

A switching regulator rapidly alternates a transistor between low-resistance ON and high-resistance OFF states. An inductor, capacitor, and diode or synchronous transistor average the switched waveform. Ideally, the switch dissipates little power in either state, allowing much higher efficiency than a linear regulator.

Switching conversion introduces ripple, electromagnetic interference, control-loop stability issues, and switching losses. The design is therefore a system problem involving power semiconductors, magnetics, layout, sensing, compensation, filtering, and protection.

4.1 Buck Converter

A buck converter reduces DC voltage. In continuous-conduction mode (CCM), with ideal components and steady state,

where D is duty ratio, defined as the fraction of each switching period for which the switch is ON. Thus,

where t_on is ON time, T_s is switching period, and f_s is switching frequency.

The inductor ripple current for an ideal buck converter is approximately

where L is inductance. The equation assumes constant input voltage, ideal switching, and CCM during the interval considered. The minimum inductor current is

CCM requires I_L(min) > 0. If this condition is not met, discontinuous-conduction equations must be used.

4.2 Boost Converter

A boost converter increases DC voltage. Under ideal CCM operation,

As duty ratio approaches unity, current stress, conduction loss, control sensitivity, and component voltage stress increase sharply. The ideal relationship must not be used without checking the operating mode and nonideal losses.

4.3 Buck–Boost Converter

The conventional inverting buck–boost converter can increase or decrease voltage while reversing polarity. In ideal CCM operation,

The negative sign indicates polarity reversal. Non-inverting buck–boost topologies, such as four-switch converters, avoid this polarity reversal but require more complex control.

Worked Example 2: Buck Converter

An ideal buck converter reduces 24 V to 12 V at 100 kHz in CCM. The duty ratio is

For L = 100\ μH, the inductor ripple is

At a 2 A average inductor current, the current varies approximately from 1.70 A to 2.30 A. A real design must additionally evaluate inductor saturation, winding resistance, switch and diode losses, capacitor ripple current, transient response, and minimum-load operation.

5. Power Semiconductor Devices

Diodes

A power diode conducts primarily in one direction. Important ratings include repetitive reverse voltage, average and surge forward current, reverse-recovery time, forward voltage, and thermal resistance. Fast-recovery and Schottky diodes reduce particular switching losses, but Schottky devices may have higher reverse leakage and voltage limitations.

MOSFETs

A power MOSFET is voltage-controlled at its insulated gate. Important parameters include drain-source voltage rating, continuous and pulsed current, R_DS(on), total gate charge, body-diode behavior, avalanche rating, and thermal characteristics. Conduction loss is approximately

where I_RMS is the RMS drain current. The resistance increases with junction temperature, so the estimate should use a temperature-appropriate value.

A rough switching-loss estimate is

where V_DS is switched voltage, I_D is drain current during transition, t_r and t_f are rise and fall times, and f_s is switching frequency. This is an engineering estimate assuming approximately linear voltage and current overlap; turn-on and turn-off delays, gate resistance, reverse recovery, and parasitic inductance can add significant loss.

IGBTs and Other Devices

IGBTs combine MOS gate control with bipolar conduction and are useful at higher voltage and moderate-to-high power, although their switching speed is generally lower than that of MOSFETs. Thyristors are robust high-power devices that latch ON and are widely used in controlled rectifiers and AC controllers. Wide-bandgap devices based on silicon carbide or gallium nitride can reduce switching loss and increase operating frequency, but they demand careful gate driving, layout, insulation, and cost evaluation.

6. Transistor Drive Circuits

A power transistor is not driven safely merely by connecting a logic output to its gate or base. The driver must supply appropriate voltage and current, charge and discharge the device rapidly, prevent undesired turn-on, and withstand switching transients.

Low-Side MOSFET Drive

In a low-side N-channel MOSFET switch, the load is connected to the positive supply and the MOSFET connects the load to ground. Advantages include simple gate referencing and low gate-drive voltage requirements. A gate resistor limits peak driver current and damps ringing; a gate-to-source pull-down ensures OFF state during reset or power-up.

High-Side Drive

A high-side N-channel MOSFET requires a gate voltage above its source by the specified gate-drive value. A bootstrap driver can provide this voltage in applications where the high-side switch is periodically turned OFF. An isolated driver or isolated bias supply may be required for continuous high-side operation or safety isolation.

BJT Drive

A bipolar transistor requires base current. A conservative switching design uses a forced beta, defined as

where I_C is collector current and I_B is base current. Selecting a forced beta lower than the transistor’s nominal small-signal gain improves saturation reliability but increases base-drive demand. A base resistor is selected from the available driver voltage, base-emitter voltage, and desired base current, while ensuring that the driver can safely source or sink the resulting current.

Gate-Drive Best Practices

Use short, low-inductance gate loops; separate power and control grounds appropriately; provide a defined OFF-state path; use dead time in complementary half-bridges; and verify gate-source voltage with an oscilloscope using a suitable probe. Excessive dead time increases diode conduction loss, whereas insufficient dead time can cause shoot-through, in which both transistors conduct simultaneously.

7. Isolation, Protection, and Safety

Isolation is required when a user-accessible circuit, communication interface, or low-voltage control domain must be separated from hazardous voltage. Transformers, optocouplers, digital isolators, isolated gate drivers, and isolated DC–DC converters are common means of achieving it.

Isolation design includes more than a component’s nominal voltage rating. Creepage is the shortest distance along an insulating surface; clearance is the shortest distance through air. Required distances depend on working voltage, pollution degree, material group, overvoltage category, altitude, and applicable safety standards. Designers must consult the relevant standards rather than relying on generic spacing rules.

Protection commonly includes:

• input fuses or circuit breakers for overcurrent and fire protection;
• transient-voltage suppressors, snubbers, or metal-oxide varistors for voltage spikes;
• current limiting or foldback for overloads;
• thermal shutdown and temperature sensing;
• reverse-polarity protection;
• soft start or inrush limiting;
• undervoltage lockout;
• desaturation or overcurrent detection in high-power switches;
• flyback diodes, TVS clamps, or active clamps for inductive loads.

A flyback diode placed across a DC relay coil or small solenoid provides a current path when the transistor turns OFF. Without it, the collapsing magnetic field can create a damaging voltage transient. For fast release, a diode combined with a Zener or TVS clamp may be preferable, provided transistor voltage ratings are respected.

Capacitors connected to hazardous supplies can retain lethal charge after disconnection. Bleeder resistors, discharge-time calculations, shielding, earthing, interlocks, touch-safe enclosures, and documented lockout procedures are essential. Compliance with applicable electrical-safety, EMC, machinery, and product standards is part of engineering responsibility.

8. Inverters and Motor-Actuation Interfaces

An inverter converts DC to controlled AC or synthesized waveforms. A single-phase half-bridge uses two switches and a split DC bus; a full bridge uses four switches and can apply positive, negative, or zero voltage to a load. A three-phase inverter uses three half-bridges and is fundamental to induction-motor, permanent-magnet motor, and brushless DC motor drives.

Pulse-width modulation (PWM) controls the average voltage. In a full bridge, reversing the diagonal pair reverses load voltage. Complementary devices in one leg must never be commanded ON simultaneously. Hardware interlocks and driver-enforced dead time are preferable to software timing alone.

For a DC motor, approximate electromagnetic torque is proportional to armature current:

and back electromotive force is approximately

where T_e is electromagnetic torque in N·m, I_a is armature current in A, E_b is back EMF in V, ω_m is angular speed in rad/s, and k_t, k_e are motor constants in consistent SI units. These relations neglect magnetic saturation, friction, commutation effects, and temperature variation.

A safe motor interface should include current sensing, hardware overcurrent shutdown, an emergency-stop path, controlled startup, regenerative-energy management, and a failure mode that removes torque. Logic-level compatibility does not guarantee adequate gate voltage, current rating, thermal performance, or fault immunity.

9. Efficiency, Heat, and Electromagnetic Compatibility

Converter efficiency is not constant. It changes with input voltage, load, switching frequency, temperature, operating mode, and control strategy. Total loss may be estimated as

where the terms represent conduction, switching, gate-drive, magnetic, capacitor, and control losses. This is a loss budget rather than a universal exact equation; each term must be modeled using the device and operating waveform.

Switching nodes have high dV/dt and dI/dt, which generate capacitive and inductive interference. Good practice includes minimizing high-current loop area, placing ceramic bypass capacitors close to switching devices, controlling gate slew rate, using Kelvin source connections where appropriate, separating noisy power grounds from sensitive analog grounds, and adding common-mode or differential-mode filtering when required. Layout is part of the circuit, not an afterthought.

Measurement also requires care. A long oscilloscope ground lead can create a false ringing waveform or short a hazardous node. Differential probes, isolated instruments, current probes, and appropriately rated test equipment should be used.

10. Design Methodology

A disciplined power-supply design proceeds as follows:

1. Define input range, output voltage, output-current range, transient requirements, isolation, environmental conditions, lifetime, and applicable standards.
2. Select topology: linear, buck, boost, buck–boost, flyback, forward, half-bridge, full-bridge, or inverter.
3. Establish worst-case duty ratio, current, voltage stress, ripple, and operating modes.
4. Select semiconductors with voltage, current, thermal, switching, and avalanche margins.
5. Select inductors, transformers, capacitors, and resistors using RMS, peak, saturation, insulation, ripple-current, and lifetime ratings.
6. Design feedback compensation and protection before optimizing efficiency.
7. Perform thermal, fault, startup, short-circuit, and transient analysis.
8. Lay out the PCB with minimized current loops and controlled return paths.
9. Test progressively using current-limited sources, isolation procedures, thermal monitoring, and appropriate probes.
10. Verify conducted and radiated emissions, safety clearances, efficiency, regulation, and fault behavior.

11. Illustrative Case: 24 V Control Supply and Motor Interface

Background

Consider an illustrative small autonomous conveyor requiring a 24 V DC input, a 5 V microcontroller supply, and bidirectional control of a 24 V brushed DC motor. The motor operates at up to 3 A continuously and 6 A during acceleration.

Problem

A direct linear reduction from 24 V to 5 V for a 0.5 A controller load would dissipate

This is excessive for a compact enclosure. The motor also produces inductive voltage transients and requires controlled reversal.

Analysis and Approach

A synchronous buck converter is selected for the logic supply. Its ideal duty ratio is approximately D = 5/24 = 0.208, with allowance for losses. The design includes input ceramic and bulk capacitors, an inductor selected for acceptable ripple and saturation current, output capacitors with adequate ripple-current rating, undervoltage lockout, and a thermal shutdown.

The motor is connected to a four-switch MOSFET H-bridge. A dedicated high- and low-side driver provides controlled gate current and dead time. Current is measured with a low-side shunt and amplifier, while a hardware comparator disables the bridge if current exceeds the safe threshold. A TVS suppressor and appropriately rated bulk capacitor limit bus transients. The controller permits reversal only after reducing PWM command to zero and applying a short controlled delay.

Outcome

The buck converter reduces logic-supply dissipation substantially compared with the linear alternative. The H-bridge provides bidirectional torque, while current limiting protects the motor, switches, and mechanical transmission. Final validation must include locked-rotor testing, rapid reversal, low-battery operation, thermal soak, EMI testing, and emergency-stop verification.

Learning Points

Topology selection depends on power level and constraints, not voltage conversion alone. Protection must be designed around realistic faults, including stalled motors, wiring shorts, failed transistors, and controller resets. Safe behavior requires hardware support rather than software assumptions only.

12. Contemporary Developments and Professional Practice

Modern power supplies increasingly use digital control, synchronous rectification, wide-bandgap switches, integrated power modules, and condition monitoring. Digital controllers can adapt switching frequency, optimize light-load operation, record fault histories, and support predictive maintenance. However, software does not remove the need for analog undervoltage, overcurrent, gate-interlock, and emergency protection because processor failure is itself a credible fault.

AI-assisted design tools can help estimate magnetic components, identify layout parasitics, and classify thermal or waveform anomalies. Their outputs must be independently verified against datasheets, measurements, safety standards, and worst-case analysis. In safety-critical power systems, traceability and explainability are more important than merely obtaining an apparently optimal design.

Chapter Summary

Power supplies transform electrical energy into regulated, safe, and useful forms. Rectifiers and capacitor filters create DC but produce ripple and pulsed input current. Linear regulators provide low-noise regulation but dissipate approximately the voltage drop multiplied by load current. Switching regulators use controlled semiconductor switching and energy-storage elements to achieve higher efficiency, with buck, boost, and buck–boost converters providing fundamental non-isolated conversion functions.

Power-device selection requires attention to voltage, current, conduction loss, switching loss, gate charge, thermal resistance, reverse recovery, and transient stress. Correct transistor drive requires defined OFF states, adequate gate or base current, high-side-drive methods where necessary, and dead time in bridge circuits. Isolation, creepage, clearance, fusing, current limiting, thermal protection, transient suppression, and safe measurement are essential engineering requirements.

A successful design balances regulation, efficiency, heat, EMC, reliability, manufacturability, and safety. Motor interfaces additionally require inductive-load protection, current control, hardware interlocks, controlled reversal, and a fail-safe torque-removal path.

Key Terms

Power supply: Circuit that converts an electrical source into a specified output.

Power electronics: Controlled conversion and management of electrical power using semiconductor devices.

Rectifier: Circuit that converts AC into unidirectional or DC voltage.

Ripple: Residual periodic variation on a nominally DC voltage or current.

Linear regulator: Regulator using a controlled pass device operating in its active region.

Switching regulator: Regulator that transfers energy through high-frequency switching.

Duty ratio: Fraction of a switching period during which a switch is ON.

Buck converter: Non-isolated step-down DC–DC converter.

Boost converter: Non-isolated step-up DC–DC converter.

Buck–boost converter: Converter capable of stepping voltage up or down, with polarity dependent on topology.

CCM: Continuous-conduction mode, in which inductor current does not reach zero.

MOSFET: Voltage-controlled power transistor with an insulated gate.

Gate driver: Circuit that supplies controlled voltage and current to a transistor gate.

Dead time: Deliberate interval preventing simultaneous conduction in complementary switches.

Isolation: Galvanic separation between electrical domains.

Creepage: Insulation-surface distance between conductors.

Clearance: Shortest distance through air between conductors.

Inrush current: Initial charging current drawn when a supply is energized.

Thermal resistance: Temperature rise per unit dissipated power, expressed in °C/W.

PWM: Pulse-width modulation used to control average voltage or power.

Review Questions

5 Short Answer Questions

1. Why does a full-wave rectifier have twice the AC source frequency as its dominant ripple frequency?
2. State two advantages and two disadvantages of a linear regulator.
3. Define duty ratio and explain its role in a buck converter.
4. Why is dead time required in a half-bridge?
5. Distinguish creepage from clearance.

5 Long Answer Questions

1. Explain the operation of a transformer-rectifier-capacitor linear power supply, including ripple and regulator dropout considerations.
2. Compare linear and switching regulators in terms of efficiency, noise, thermal design, complexity, and application suitability.
3. Derive the ideal CCM voltage relationship for a buck converter using inductor volt-second balance.
4. Explain the major components of a MOSFET gate-drive circuit and the consequences of poor gate-drive design.
5. Discuss the protection and safety requirements of an isolated motor-drive power system.

3 Analytical / Application Questions

1. A 24 V source feeds an ideal buck converter producing 9 V at 2 A. Determine the duty ratio and explain which nonideal effects must be included before selecting components.
2. A linear regulator supplies 12 V from 18 V at 0.8 A. Calculate its approximate pass-device dissipation and discuss whether a heatsink is likely to be required.
3. Design a fault-management strategy for a 24 V H-bridge driving a motor that can draw six times its normal current during a stall. Identify sensing, shutdown, transient, thermal, and restart requirements.

Multiple Choice Questions with Answers

Question 1

A full-wave rectifier connected to 50 Hz AC has a dominant ripple frequency of:

A. 25 Hz

B. 50 Hz

C. 100 Hz

D. 150 Hz

Correct Answer: C. 100 Hz

Question 2

A linear regulator changes 24 V to 5 V at 1 A. Neglecting quiescent current, its approximate pass-device power loss is:

A. 5 W

B. 19 W

C. 24 W

D. 29 W

Correct Answer: B. 19 W

Question 3

For an ideal buck converter operating in CCM with 24 V input and 12 V output, the duty ratio is:

A. 0.25

B. 0.50

C. 1.50

D. 2.00

Correct Answer: B. 0.50

Question 4

Which device parameter most directly represents MOSFET conduction resistance when ON?

A. R_DS(on)

B. V_BR

C. t_r

D. Q_rr

Correct Answer: A. R_DS(on)

Question 5

The main purpose of dead time in a half-bridge is to:

A. Increase transformer voltage

B. Prevent simultaneous conduction of opposing switches

C. Increase output ripple

D. Eliminate all switching loss

Correct Answer: B. Prevent simultaneous conduction of opposing switches

Question 6

A flyback diode across a DC relay coil primarily:

A. Raises coil current during steady state

B. Provides a path for inductive current when the switch turns OFF

C. Increases the switching frequency

D. Provides galvanic isolation

Correct Answer: B. Provides a path for inductive current when the switch turns OFF

Question 7

For an ideal CCM boost converter, increasing duty ratio generally:

A. Decreases output voltage toward zero

B. Leaves output voltage unchanged

C. Increases output voltage according to 1/(1−D)

D. Reverses output polarity in every topology

Correct Answer: C. Increases output voltage according to 1/(1−D)

Question 8

Which quantity is measured in °C/W and used in junction-temperature estimation?

A. Capacitance

B. Thermal resistance

C. Duty ratio

D. Creepage

Correct Answer: B. Thermal resistance

Question 9

Which feature is most directly associated with galvanic isolation?

A. A resistor divider

B. A transformer or digital isolator

C. A heatsink

D. A gate pull-down resistor

Correct Answer: B. A transformer or digital isolator

Question 10

A four-switch H-bridge is particularly useful for a brushed DC motor because it can:

A. Provide bidirectional motor voltage and torque

B. Eliminate the need for current protection

C. Operate without any transistor dead time

D. Convert all DC input power into mechanical power

Correct Answer: A. Provide bidirectional motor voltage and torque