Module 19 · UPSC Optional: Electrical Engineering

Electrical Engineering Paper I: Core Topics

Updated 18 Nov 2026 · 7,911 words · illustrated
Updated 17 Nov 2026
Electrical Engineering Paper I: core syllabus areas, key concepts, standard references and answer-writing practice.
Circuit Theory · Electrical Machines · Power Systems
🎯 Take Mock Test
Circuit TheoryElectromagneticFieldsPower SystemsControl SystemsElectrical MachinesfeedsdrivescontrolsArt. 1: Core Syllabus CoverageMechanism: Energy Conversion Flowvlymbooq
Electrical Engineering Core Topics Classification Diagram for GS1 — Shows the main core topics of Electrical Engineering Paper I and their interrelations. · Open the full diagram →
Ecology BasicsPopulation EcologyBiodiversityConservationClimate ChangeEnvironmentalPollutionEnvironmentalPolicies & ActsInternational Orgs(UNEP, WHO)India Initiatives(Swachh, NAPCC)supportsdrivesimpactsexacerbatesregulatesguidesimplementsEcosystem = biotic + abioticKey policy hooks: NGT Act, Paris Accordvlymbooq
Comprehensive Flow of Core Ecology Topics for GS3 Environment — Shows the logical progression of key ecology concepts and their policy links for quick revision. · Open the full diagram →

Learning Objectives

  • Understand the fundamental concepts of Electrical Engineering Paper I: Core Topics
  • Apply key formulas and techniques to solve problems
  • Practice with exam-level questions to build speed and accuracy

Electrical Engineering Paper I — Circuit Theory, Electromagnetic Theory & Electrical Machines

Electrical Circuits

  • Network Theorems: Kirchhoff's laws (KCL: Kirchhoff's current law - algebraic sum of currents at a node = 0; KVL: Kirchhoff's voltage law - algebraic sum of voltages around a closed loop = 0), superposition theorem, Thevenin's theorem (any linear bilateral network can be replaced by an equivalent circuit consisting of a voltage source V_th in series with resistance R_th), Norton's theorem (equivalent current source I_N in parallel with resistance R_n), maximum power transfer theorem (maximum power transferred when load resistance R_L = Thevenin resistance R_th), compensation theorem, substitution theorem, reciprocity theorem, Tellegen's theorem.
  • Network Analysis: Nodal analysis (node voltage method), mesh analysis (mesh current method), source transformation (voltage source to current source conversion and vice versa), star-delta transformation (star-delta conversion, delta-star transformation), star-delta transformation (Y-Δ transformation), star-delta transformation formulas.
  • Transient Analysis: Transient response of RC circuits (RC circuit charging, RC circuit discharging), RC circuit transient response, RC circuit time constant τ = RC, RL circuit transient response, RL circuit time constant τ = L/R, RC circuit time constant τ = RC, second-order circuits (RLC circuits), series RLC circuit, parallel RLC circuit, resonant frequency f₀ = 1/(2π√(LC)), quality factor Q = ω₀L/R = 1/(ω₀CR), bandwidth BW = f₂ - f₁ = f₀/Q, selectivity, bandwidth.
  • Two-Port Networks: Two-port network parameters (Z-parameters, Y-parameters, H-parameters, ABCD parameters, transmission parameters), interconnection of two-port networks (cascade connection, series connection, parallel connection), image impedance, image impedance Z₁ = √(Z₁₁Z₂₂ - Z₁₂Z₂₁), image impedance Z₂ = √(Z₂₂Z₁₁ - Z₁₂Z₂₁), image impedance Z₁ = Z₂ for symmetrical networks, iterative impedance, image impedance, iterative impedance, image impedance.
  • Filters: Low-pass filter, high-pass filter, band-pass filter, band-stop filter, active filters, passive filters, Butterworth filter, Chebyshev filter, Butterworth filter characteristics, Butterworth filter characteristics (maximally flat magnitude response), Butterworth filter order n, cutoff frequency ω_c, roll-off rate 20n dB/decade, Butterworth filter magnitude response |H(jω)| = 1/√(1 + (ω/ω_c)²ⁿ).
  • Two-Port Network Parameters: Z-parameters (impedance parameters), Y-parameters (admittance parameters), H-parameters (hybrid parameters), ABCD parameters (transmission parameters), T-parameters (transmission parameters), inverse hybrid parameters, scattering parameters (S-parameters), scattering parameters (S-parameters), S-parameters (S₁₁, S₁₂, S₂₁, S₂₂), return loss, insertion loss, insertion loss, return loss, voltage standing wave ratio (VSWR), voltage standing wave ratio VSWR = (1 + |Γ|)/(1 - |Γ|) where Γ = reflection coefficient, voltage standing wave ratio VSWR = (1 + |Γ|)/(1 - |Γ|), return loss RL = -20 log₁₀|Γ| dB, insertion loss IL = -20 log₁₀|S₂₁| dB, return loss RL = -20 log₁₀|Γ| dB.
  • Filter Design: Butterworth filter, Chebyshev filter, elliptic filter (Cauer filter), Bessel filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter, Gaussian filter.

Common Traps

  • Kirchhoff's Laws: KCL - algebraic sum of currents entering a node = 0; KVL - algebraic sum of voltages around a closed loop = 0; sign convention for voltage drops and rises; sign convention for current entering/leaving a node.
  • Superposition Theorem: Applicable only to linear circuits; turn off all independent sources except one (voltage sources → short circuit, current sources → open circuit), calculate contribution, sum algebraically; dependent sources remain active.
  • Maximum Power Transfer: Maximum power transferred to load when R_L = R_th; maximum power P_max = V_th²/(4R_th); efficiency at maximum power transfer = 50%.
  • Resonance: Series resonance: impedance minimum, current maximum, power factor = 1; Parallel resonance: impedance maximum, current minimum, power factor = 1; resonance frequency f₀ = 1/(2π√(LC)) for both series and parallel LC circuits.
  • Transient Response: First-order circuits (RC, RL): time constant τ = RC (RC circuit), τ = L/R (RL circuit); second-order circuits (RLC): underdamped, critically damped, overdamped based on damping ratio ζ; critical damping when ζ = 1.
Technique: Nodal Analysis
Question: Find node voltage V₁ at node 1 connected to 10V source via 2Ω resistor, to ground via 4Ω resistor, and to node 2 via 2Ω resistor; node 2 connected to ground via 6Ω resistor and to node 1 via 2Ω resistor.
Answer: KCL at node 1: (V₁ - 12)/2 + (V₁ - V₂)/2 + V₁/4 = 0 → multiply by 4: 2(V₁ - 12) + 2(V₁ - V₂) + V₁ = 0 → 2V₁ - 24 + 2V₁ - 2V₂ + V₁ = 0 → 5V₁ - 2V₂ = 24. KCL at node 2: (V₂ - V₁)/2 + V₂/6 = 0 → multiply by 6: 3(V₂ - V₁) + V₂ = 0 → 3V₂ - 3V₁ + V₂ = 0 → 4V₂ - 3V₁ = 0 → V₂ = 0.75V₁. Substitute: 5V₁ - 2(0.75V₁) = 24 → 5V₁ - 1.5V₁ = 24 → 3.5V₁ = 24 → V₁ = 6.857V, V₂ = 0.75 × 6.857 = 5.143V.

Common Traps

  • KCL/KCL Sign Convention: Currents entering node = currents leaving node; or sum of currents leaving node = 0; be consistent with sign convention.
  • Source Transformation: Voltage source V in series with R ↔ current source I = V/R in parallel with R; remember R_th = R_N = R_th.
  • Superposition Theorem: Turn off all independent sources except one; voltage sources → short circuit, current sources → open circuit; dependent sources remain active; sum contributions algebraically.
  • Maximum Power Transfer: Maximum power transferred when R_L = R_th; maximum power P_max = V_th²/(4R_th); efficiency at maximum power transfer = 50% (not 100%).
  • Supermesh: When current source is on branch shared by two meshes, create supermesh by excluding current source; apply KVL around supermesh; add constraint equation I₁ - I₂ = I_source.
Mains strategy
Electrical engineering answers: draw circuit diagrams clearly, label all nodes/branches/nodes, show step-by-step derivation of circuit equations, state assumptions clearly, quote IS codes (IS 456 for RCC, IS 800 for steel, IS 1893 for seismic design, IS 10262 for mix design), quote standard values (electrical resistivity of copper 1.72×10⁻⁸ Ω·m, copper density 8.96 g/cm³, electrical resistivity of aluminium 2.83×10⁻⁸ Ω·m, aluminium density 2.7 g/cm³).

Electrical Machines: The Physics of Energy Conversion and Losses

The fundamental principle governing all electrical machines is energy conversion, governed by Lenz's Law and the conservation of energy. A motor converts electrical energy into mechanical energy, while a generator performs the reverse. However, this process is never 100% efficient due to inherent losses. Understanding these losses is critical for analyzing machine performance under varying load conditions.

Copper Losses (I²R Losses): These occur in the armature and field windings. Since they vary with the square of the current, they are variable losses. In a DC machine, armature copper loss is I_a2 R_a, while field copper loss is Ish2 Rsh (for shunt) or constant for series/series-shunt fields.

Iron Losses (Core Losses): These occur in the magnetic circuit (laminated steel). They consist of hysteresis loss and eddy current loss. Hysteresis loss is proportional to frequency (f) and the area of the hysteresis loop (Bmax1.6 to Bmax2). Eddy current loss is proportional to the square of frequency (f2) and the square of flux density (Bmax2). Since supply frequency is constant in most grids, iron losses are often treated as constant losses, though they increase with load due to slight saturation effects increasing Bmax.

Mechanical Losses: Friction and windage losses occur in bearings and due to air resistance on the rotating parts. These are roughly constant but depend on speed (N).

Stray Load Losses: These account for additional losses due to armature reaction, harmonic fluxes, and non-uniform flux distribution when the machine is loaded. They are typically estimated as 1% to 2% of the rated output.

In a synchronous motor, there is no "armature reaction" loss in the traditional sense because the rotor field is DC-excited and locked to the stator field spatially, but armature reaction still affects the terminal voltage required. The power factor correction capability arises here: by adjusting the DC excitation current (I_f), a synchronous motor can operate at leading, unity, or lagging power factor without changing the mechanical load torque.

Examiner's trap
Question: "In a DC shunt motor, if the field flux (φ) is reduced by half while maintaining constant terminal voltage and neglecting armature reaction, what happens to speed?"
Answer: Speed N ∝ (V - I_a R_a)/(φ). If φ halves and we assume I_a R_a drop is negligible initially, speed doubles. However, as speed increases, back EMF (E_b) rises. To maintain steady state, armature current drops. The correct answer must account that the new equilibrium point has higher speed but lower torque capability for the same current. If load torque is constant (fan load ∝ N2), speed will settle at a value slightly less than double due to increased I_a R_a drop and brush contact voltage drop considerations.
Exam Strategy: Handling Numerical Data
When given efficiency (eta) and output power (Pout), always calculate input power first (Pin = Pout/eta). Then, separate losses into fixed (iron + mechanical) and variable (copper) components if possible. Do not assume all losses are copper unless specified. Always state the assumption made regarding armature reaction in numerical problems involving DC machines.

Power Systems: Stability, Protection, and Grid Dynamics

Power system stability refers to the ability of the system to regain a state of equilibrium after being subjected to a disturbance. It is categorized into three types: rotor angle stability (small and large signal), voltage stability, and frequency stability.

Small Signal Stability: Analyzed using linearized swing equations around an operating point. Eigenvalue analysis determines if the system modes are damped or unstable. Weakly damped oscillations often occur due to interaction between turbine-generator shafts (inter-area modes) and power electronics controls.

Large Signal Stability: Involves non-linear dynamics where the system moves far from the operating point. The Equal Area Criterion is a graphical method used for single-machine infinite bus (SMIB) systems to determine the critical clearing time of a fault. If the fault is cleared before the accelerating area exceeds the decelerating area, the machine will remain stable.

Voltage Stability: Occurs when the system cannot maintain acceptable voltages at all buses under normal or near-normal conditions. This is often associated with heavy loading where reactive power demand exceeds supply. The "nose curve" of the P-V curve indicates the voltage collapse point. Solutions involve installing static VAR compensators (SVC), FACTS devices, or upgrading transmission capacity.

Protection Coordination: Relays must operate selectively. A fault near the source should be cleared by the relay closest to it. Time-graded and distance protection schemes ensure this. Distance relays measure impedance (Z = V/I); since Z ∝ length of line, they can detect faults based on location. Inverse time characteristics are used for overcurrent relays: closer faults (higher current) cause faster operation.

Harmonics: Non-linear loads (VFDs, rectifiers) generate harmonic currents. These cause neutral overloading in 3-phase 4-wire systems due to triplen harmonics (3rd, 9th, 15th). Filters are designed using tuned LC circuits to shunt specific harmonic frequencies to ground.

Examiner's trap
Question: "Why does a synchronous motor not start by itself when connected to AC supply?"
Answer: At standstill, the stator produces a rotating magnetic field. The rotor (DC excited) is stationary. Due to inertia, the rotor cannot lock onto the rotating field instantly.

Electromagnetic Induction: The Physics of Energy Conversion

The fundamental principle governing all electrical machines is Faraday's Law of Electromagnetic Induction, which states that an electromotive force (EMF) is induced in a conductor whenever the magnetic flux linking it changes. Mathematically, for a coil with N turns and flux φ, the induced EMF e = -N (dφ)/(dt). The negative sign represents Lenz's Law: the direction of the induced EMF opposes the change in flux that produced it. In practical terms, this means that to generate electricity (generator mode), mechanical work must be done to rotate a conductor within a magnetic field, converting kinetic energy into electrical energy. Conversely, to create motion (motor mode), an external electrical current is passed through a conductor in a magnetic field, producing a force F = BIL (Lorentz force), thereby converting electrical energy into mechanical work.

The distinction between the two modes lies in the source of the power flow. In a generator, the induced EMF drives current against an external load resistance. The terminal voltage V_t is less than the generated EMF (E_g) due to internal armature resistance drop: V_t = E_g - I_a R_a. In a motor, the applied terminal voltage V_t must overcome both the back EMF (E_b) and the resistive drop to drive current: V_t = E_b + I_a R_a. This back EMF is crucial; without it, the armature current would be limited only by the small resistance of the winding, leading to a massive short-circuit current that would destroy the machine instantly. The magnitude of this back EMF is directly proportional to the speed (N) and flux (φ): E_b ∝ Nφ. Thus, if the field flux collapses or the load increases causing a speed drop, the back EMF decreases, drawing more current to restore equilibrium.

The construction of machines optimizes this interaction. In DC machines, commutators and brushes mechanically rectify the alternating EMF generated in the armature coils into a unidirectional output voltage. This allows for stable DC power delivery. In AC machines (alternators), slip rings allow the rotating field to be supplied with DC excitation while the stator windings pick up the induced AC voltage directly, eliminating the need for heavy mechanical commutation on the high-voltage side. The choice between salient pole and cylindrical rotor constructions in synchronous machines depends on the speed of operation; hydroelectric generators (low speed) use salient poles to accommodate many slots for winding distribution, while thermal/gas turbine generators (high speed) use non-salient cylindrical rotors to withstand centrifugal forces.

Examiner's trap
Question: "If the field current of a synchronous motor is increased while keeping mechanical load constant, what happens to armature current and power factor?" Answer: Increasing field current increases the internal generated voltage (E_f). Since the terminal voltage V_t is fixed, the phasor difference between V_t and E_f changes. Initially, if the motor was operating at lagging PF (under-excited), increasing excitation reduces the reactive power drawn from the grid. The armature current magnitude may first decrease to a minimum value when the machine operates at unity power factor, and then increase again as it becomes over-excited (leading PF). Crucially, the real power output remains constant because mechanical load is unchanged; only the reactive power component of the armature current changes.
Full Worked Answer:
1. Identify initial state: Motor is under-excited (E_f < V_t sin δ), drawing lagging VARs.
2. Action: Increase DC field current (I_f) → Flux (φ) increases → E_f increases.
3. Phasor Diagram Analysis: The tip of the E_f vector moves further along the q-axis (assuming constant torque angle initially).
4. Result: The angle between V_t and I_a decreases (PF improves towards unity) then reverses to leading.
5. Magnitude: Current drops until unity PF, then rises. Real power (P = VI cos φ) stays constant as mechanical load is fixed.
6. Conclusion: Armature current first decreases then increases; Power factor shifts from lagging to unity to leading.
Exam-Day Strategy: Handling Vector Diagrams
When drawing phasor diagrams for synchronous machines, always draw the terminal voltage V_t as the reference phasor (horizontal). Place the internal EMF (E_f) at an angle δ (load angle) relative to V_t. The armature current I_a is drawn such that its projection on the real axis gives the active power component. Remember: In a generator, E_f leads V_t; in a motor, E_f lags V_t. Never assume δ = 0 unless specified as no-load or synchronous speed condition without load. Always label the quadrants clearly to distinguish between motoring and generating modes.

Power System Stability: Small Signal vs Large Signal Dynamics

Power system stability is defined as the ability of a power system to regain a state of equilibrium after being subjected to a disturbance. The analysis is divided into three primary categories based on the nature of the disturbance and the response time. Rotor angle stability (often synonymous with transient stability) concerns the ability of synchronous machines to remain in synchronism following a large disturbance like a short circuit or loss of a generator. This is governed by the swing equation M (d2δ)/(dt2) = P_m - P_e, where M is inertia, δ is the rotor angle, P_m is mechanical input power, and P_e is electrical output power. The Equal Area Criterion provides a graphical solution for this: the area under the accelerating power curve (P_m - Pe1) must be less than or equal to the area under the decelerating power curve (Pe2 - P_m) within one cycle of oscillation.

Small signal stability, also known as dynamic stability, deals with the system's response to small disturbances like load fluctuations. Here, linearized models around an operating point are used. Eigenvalue analysis of the state matrix determines if the system modes are damped (real part of eigenvalues < 0) or unstable. Weakly damped oscillations often arise from the interaction between turbine-generator shafts (inter-area modes) and power electronics controls like HVDC links or FACTS devices. If these controls introduce delays or phase shifts, they can excite low-frequency oscillations that grow over time, potentially leading to cascading tripping events.

Voltage stability is a distinct phenomenon related to the ability of the system to maintain acceptable voltages at all buses under normal or near-normal conditions. It is heavily dependent on reactive power balance. The P-V curve (power vs voltage) for a bus shows that as load increases, voltage drops. Beyond a certain point (the "nose" of the curve), the system becomes unstable, and any further increase in load causes voltage collapse. This is often associated with heavy loading where reactive power demand exceeds supply capability from generators and capacitors.

Frequency stability relates to the balance between generation and load. If generation drops below load, frequency falls; if generation exceeds load, frequency rises. Modern grids have automatic generation control (AGC) to maintain frequency within narrow bands (e.g., 50 Hz ± 0.2 Hz). However, severe imbalances can lead to under-frequency load shedding schemes to prevent blackouts.

Examiner's trap
Question: "Why does a synchronous motor not start by itself when connected to AC supply?" Answer: At standstill, the stator produces a rotating magnetic field. The rotor (DC excited) is stationary. Due to inertia, the rotor cannot lock onto the rotating field instantly. Furthermore, if the rotor were unexcited (induction mode), it would act like an induction motor but with very low torque because the slip rings are not designed for high starting currents and the cage winding (if present) has high resistance.

Network Theorems in AC Steady‑State Analysis: Superposition, Thevenin, and Norton with Phasors

In AC steady‑state the principle of superposition holds only for linear circuits where all sources operate at the same frequency. When multiple sinusoidal sources of identical frequency are present, each source can be considered independently by replacing the others with their internal impedances (short circuit for ideal voltage sources, open circuit for ideal current sources). The individual phasor responses are then added vectorially to obtain the total response. This is fundamentally different from the DC case because the phase angles of the phasors must be preserved during addition; scalar addition of magnitudes leads to gross errors.

Thevenin’s theorem in the phasor domain states that any linear two‑terminal network can be replaced by an equivalent voltage source Vth in series with an impedance Zth. Vth is the open‑circuit voltage across the terminals measured with all independent sources active. Zth is the impedance seen from the terminals with all independent sources deactivated (voltage sources shorted, current sources opened) while dependent sources remain. The same procedure yields the Norton equivalent: a current source IN = Vth/Zth in parallel with Zth. Because impedances are complex, the division must be performed in polar or rectangular form, preserving both magnitude and angle.

A frequent examiner twist is to embed a dependent source whose controlling variable lies outside the portion of the circuit being Thevenized. In that case the dependent source cannot be deactivated; its effect must be retained while calculating Zth by applying a test voltage (or current) source at the terminals and measuring the resulting current (or voltage). The ratio gives Zth directly. This “test‑source method” is the only reliable way when dependent sources are present.

Maximum power transfer in AC circuits occurs when the load impedance ZL is the complex conjugate of the Thevenin impedance: ZL = Zth* . If the load is restricted to be purely resistive, the optimum resistance equals the magnitude of Zth. The delivered real power is then Pmax = |Vth|² / (4·Re{Zth}) for the conjugate match, and |Vth|²·Re{Zth} / (|Zth| + Re{Zth})² for the resistive‑only case. These formulas are often asked in a “derive” format, so practice the algebraic steps from the power expression P = Re{VL·IL*}.

When solving networks with multiple sources of different frequencies, superposition must be applied frequency by frequency. The total time‑domain response is the sum of the individual sinusoidal steady‑state responses; phasor addition across frequencies is invalid because the phasors rotate at different speeds. This distinction is a classic trap: a problem may give a 50 Hz source and a 150 Hz harmonic source, expecting the candidate to compute separate phasor solutions and then add the resulting time functions, not the phasors.

Transformer Inrush Current: Mechanism, Harmonic Content, and Protection Coordination

When a transformer is energized, the instantaneous flux in the core is the sum of the residual flux (remanence) and the flux produced by the applied voltage. If the switching instant coincides with a voltage zero crossing where the required flux change is maximum, the core can be driven deep into saturation. In saturation the magnetizing inductance collapses to a very low value, and the primary draws a large, non‑sinusoidal current known as inrush current. Peak values can reach 8–12 times the rated full‑load current and may persist for several cycles (typically 0.1–0.5 s) before decaying as the flux settles to its steady‑state sinusoidal trajectory.

The waveform of inrush current is rich in even and odd harmonics, with a dominant second harmonic (100 Hz for a 50 Hz system) often exceeding 30–50 % of the fundamental component. This high second‑harmonic content is the key signature used by modern differential relays to discriminate between inrush and an internal fault. A typical restraint characteristic adds a second‑harmonic blocking element: if the ratio of second‑harmonic current to fundamental current exceeds a set threshold (commonly 15–20 %), the relay inhibits operation for a predefined time window.

Residual flux magnitude and polarity depend on the instant of previous de‑energization. A transformer switched off at a voltage peak leaves maximum residual flux; re‑energizing at the opposite voltage zero crossing produces the worst‑case inrush. Controlled switching (point‑on‑wave closing) can mitigate this by synchronizing the breaker closure to a voltage angle that minimizes the flux excursion. However, point‑on‑wave schemes add complexity and are not universally deployed.

Protection coordination must account for the fact that inrush current flows only in the primary winding (or the winding connected to the source). The secondary sees negligible current because the magnetizing branch is effectively open‑circuited on the secondary side. Consequently, a differential relay that compares primary and secondary currents will see a large apparent differential during inrush. The second‑harmonic restraint prevents false trips, but the relay must still be set to operate for genuine internal faults that also contain harmonics (e.g., faults with DC offset). The coordination time interval (CTI) between the transformer differential and the upstream overcurrent relay is typically 0.3–0.5 s to allow the inrush to decay before the backup relay sees the residual current.

Inrush also stresses mechanical forces on windings and can cause voltage dips in the supply network. For large power transformers, the inrush apparent power may approach the short‑circuit MVA of the source, leading to temporary voltage depressions that affect other loads. System studies often model inrush as a time‑varying nonlinear inductance to evaluate voltage sag magnitude and duration.

Worked Example: Second‑Harmonic Restraint Setting
A 100 MVA, 220/11 kV transformer is protected by a differential relay with a second‑harmonic restraint threshold of 20 %. During energization the measured primary currents are: fundamental = 1.2 kA, second harmonic = 0.3 kA. Will the relay restrain? If the same transformer experiences an internal phase‑to‑ground fault producing fundamental = 5 kA and second harmonic = 0.8 kA, will the relay operate?
Calculate the second‑harmonic ratio for each case. For inrush: ratio = 0.3 kA / 1.2 kA = 0.25 = 25 %. Since 25 % > 20 %, the relay will restrain (block operation). For the fault: ratio = 0.8 kA / 5 kA = 0.16 = 16 %. Because 16 % < 20 %, the restraint condition is not met and the relay will operate (trip). This illustrates why the threshold must be chosen above typical inrush ratios but below fault ratios.
Exam‑Day Strategy for Transformer Inrush Questions
Memorize the typical second‑harmonic percentage range (30–50 % of fundamental) and the common relay threshold (15–20 %). When a numerical problem gives fundamental and harmonic currents, compute the ratio instantly and compare with the threshold. For theory questions, emphasize the role of residual flux, point‑on‑wave switching, and the fact that inrush appears only on the energized winding. Sketch the flux‑versus‑time waveform showing saturation peaks to earn diagram marks.

Synchronous Machine Transient Reactances: Subtransient, Transient, and Steady‑State Models

The dynamic behavior of a synchronous machine during faults is captured by three distinct reactances that represent the progressive decay of the armature reaction flux as the damper windings and field winding respond. The subtransient reactance Xd″ (direct‑axis) and Xq″ (quadrature‑axis) model the immediate response (first few cycles) when both damper windings and field winding carry induced currents. Because the damper bars are low‑resistance copper or aluminum cages, the subtransient time constant Td″ is very short, typically 0.02–0.05 s. The subtransient reactance is the smallest of the three (≈0.15–0.25 p.u.) and determines the initial symmetrical short‑circuit current Isc″ = Ef/Xd″.

After the damper currents decay, the field winding current begins to change, governed by the transient reactance Xd′ (≈0.25–0.35 p.u.) and the transient open‑circuit time constant Td′ (≈1–3 s). The transient reactance is used for stability studies covering the first swing (0.5–2 s). The steady‑state synchronous reactance Xd (≈1.0–1.8 p.u.) represents the final condition when the field current has reached its new equilibrium under the action of the automatic voltage regulator (AVR). The quadrature‑axis counterparts Xq′ and Xq are generally larger than their direct‑axis equivalents because the q‑axis lacks a field winding.

The three‑reactance model is derived from the Park transformation and the equivalent circuits of the d‑ and q‑axes. Each axis is represented by a voltage source behind a series reactance, with parallel branches for the field and damper windings. The time constants are obtained from the eigenvalues of the linearized flux‑linkage equations. In per‑unit system, the reactances are expressed on the machine MVA base and rated voltage. Conversion to a system base requires the square of the voltage ratio times the MVA ratio.

Examiners often ask for the sequence of reactances during a three‑phase fault at the generator terminals. The correct order is: subtransient current (first 2–3 cycles) → transient current (up to ~1 s) → steady‑state current (after AVR action). The DC offset component, governed by the armature time constant Ta = Xd″/ωRa, decays within a few cycles and is superimposed on the AC envelope. The peak asymmetrical current can be 1.8–2.0 times the symmetrical RMS value, which is critical for breaker rating.

In multi‑machine stability simulations, each generator is represented by its transient reactance Xd′ behind an internal voltage Eq′ that is assumed constant during the study period (classical model). The AVR is omitted for first‑swing studies; its inclusion requires the full flux‑decay model with Xd″, Xd′, and Xd plus the exciter dynamics. Understanding which model to apply for a given time frame is a decisive skill.

Power System Protection: Distance Relaying Principles, Zones, and Coordination

Distance relays measure the impedance Z = V/I at the relay location and compare it with a set reach impedance. Because the line impedance is proportional to length, the relay can determine the electrical distance to a fault. The basic characteristic is a circle (or quadrilateral) on the R‑X plane centered at the origin for a simple impedance relay, or offset for a mho relay. The mho characteristic is inherently directional: its circle passes through the origin, so it only operates for faults in the forward direction (positive real impedance). This self‑directional property eliminates the need for a separate directional element.

Protection schemes employ three zones. Zone 1 is set to 80–85 % of the protected line impedance and operates instantaneously (no intentional time delay). The underreach avoids overreach due to errors in CT/VT ratios, line parameter uncertainty, and fault resistance. Zone 2 covers the remaining 15–20 % of the line plus 120–150 % of the next line section, with a typical time delay of 0.3–0.5 s to coordinate with the downstream Zone 1. Zone 3 provides backup for the remote bus and adjacent lines, set to 120–150 % of the total impedance up to the farthest bus, with a longer delay (0.6–1.0 s). The coordination time interval (CTI) between zones must exceed the circuit‑breaker operating time plus relay reset time plus a safety margin (usually 0.2–0.3 s).

Fault resistance (arc resistance) shifts the measured impedance along the real axis, potentially causing underreach for mho relays. Quadrilateral (polygon) characteristics mitigate this by providing a wide resistive reach independent of the reactive reach. Modern numerical relays implement load‑encroachment blinder characteristics to prevent mal‑operation during heavy load conditions where the load impedance falls inside the mho circle. The blinder is a vertical line on the R‑X plane that blocks operation for impedances with a phase angle less than a set value (typically 30°).

Communication‑assisted schemes (permissive overreaching transfer trip, POTT; directional comparison blocking, DCB) use Zone 2 overreach (set to 120–150 % of line) and a communication channel to achieve instantaneous clearing for faults in the last 15–20 % of the line. In POTT, the local relay sends a permissive signal to the remote end when its Zone 2 operates; the remote end trips instantaneously if its Zone 2 also sees the fault. In DCB, the remote end sends a blocking signal if the fault is behind it, preventing the local Zone 2 from tripping. These schemes require reliable communication (power line carrier, fiber, or microwave) and careful coordination of timer settings.

Examiners frequently test the effect of infeed from a remote source on the apparent impedance seen by a distance relay. Infeed increases the fault current measured by the relay without a proportional increase in voltage, causing the apparent impedance to be lower than the actual line impedance to the fault. This leads to overreach. The correction factor is Zapp = Zline·(1 + Iremote/Ilocal). Candidates must be able to derive this expression and apply it to a two‑ended line with known source impedances.

HVDC Transmission: Converter Topologies, Control Modes, and Commutation Failure

High‑voltage direct current (HVDC) links are built around line‑commutated converters (LCC) using thyristors or voltage‑source converters (VSC) using IGBTs. LCC‑HVDC requires a strong AC network at both ends because the thyristors rely on the AC voltage for natural commutation. The converter bridge is a twelve‑pulse arrangement (two six‑pulse bridges with a 30° phase shift) to reduce harmonic generation. The DC voltage is controlled by the firing angle α (rectifier) and extinction angle γ (inverter). The fundamental DC voltage is Vd = (3√2/π)·VLL·cos α – Id·Xc, where VLL is the line‑to‑line AC voltage and Xc is the commutating reactance.

VSC‑HVDC employs pulse‑width modulation (PWM) to synthesize a controllable AC voltage from the DC capacitor. It can independently control active and reactive power, operates with weak or passive AC networks, and provides black‑start capability. The converter station consists of a phase‑reactor, the VSC valve, a DC capacitor, and a DC cable/overhead line. Control is typically implemented in the dq‑reference frame aligned with the AC voltage vector, allowing decoupled control of P (via Id) and Q (via Iq).

Commutation failure is a phenomenon unique to LCC‑HVDC. It occurs when the inverter extinction angle γ falls below the minimum required (typically 15–18°) due to a dip in AC voltage, excessive DC current, or excessive firing angle. The current in the outgoing valve does not reach zero before the incoming valve is fired, causing a short‑circuit across the AC bus through the two valves. This results in a sudden collapse of DC voltage and a large AC current surge. Repeated commutation failures can lead to converter block. Mitigation includes predictive control that advances the firing angle during AC voltage dips, and the use of capacitor‑commutated converters (CCC) that inject a series capacitor to aid commutation.

Harmonic performance differs markedly. LCC generates characteristic harmonics of order h = 12k ± 1 on the AC side (11th, 13th, 23rd, 25th…) and 12k on the DC side. Filters (tuned, high‑pass, C‑type) are sized to meet IEEE 519 limits. VSC with PWM produces high‑frequency switching harmonics around the carrier frequency (typically 1–2 kHz) and sidebands; these are attenuated by the phase reactor and a small high‑pass filter. The absence of low‑order harmonics simplifies AC filter design for VSC.

Control modes for a bipole LCC link include constant current control (rectifier), constant extinction angle control (inverter), and voltage‑dependent current order limiter (VDCOL) to prevent overcurrent during AC faults. For VSC, the outer loop can be constant power, constant DC voltage, or frequency support (droop). The inner current loop bandwidth is typically 10–20 times the fundamental frequency. Coordination between the two converters in a multi‑terminal HVDC grid requires a communication‑based master‑slave or voltage‑droop scheme to share power and maintain DC voltage stability.

  • Confusing subtransient, transient, and synchronous reactance values when calculating fault currents at different time frames.
  • Assuming the mho relay characteristic is non‑directional and forgetting its inherent directional property.
  • Neglecting the effect of fault resistance on distance relay reach, leading to incorrect zone settings.
  • Mixing up LCC and VSC commutation principles; commutation failure applies only to LCC.
  • Using the wrong harmonic order formula for LCC filters (e.g., using 6k±1 instead of 12k±1).
  • Forgetting that inrush current contains a dominant second harmonic, which is the basis for differential relay restraint.
  • Review these traps before the exam to avoid common mistakes.

Power System Load Flow: Newton‑Raphson, Fast Decoupled, and Convergence Issues

Load‑flow (power‑flow) analysis is the backbone of steady‑state planning and operation. The Newton‑Raphson (NR) method solves the nonlinear nodal power equations Pi(V,δ) and Qi(V,δ) by linearising them around the current iterate and updating the voltage magnitude and angle vector x = [δ, V]T through Δx = –J–1·F(x). The Jacobian J is a 2n×2n matrix partitioned into four sub‑matrices H, N, M, L that contain partial derivatives of active and reactive power with respect to angles and magnitudes. Because H and L are dense, NR exhibits quadratic convergence near the solution but requires a full Jacobian rebuild (or at least a factorisation) each iteration, which dominates CPU time for large systems.

The Fast Decoupled Load Flow (FDLF) exploits the weak coupling between P‑δ and Q‑V in high‑voltage transmission networks where R/X ≪ 1. By neglecting the off‑diagonal Jacobian blocks N and M and approximating H ≈ B′, L ≈ B″ (the imaginary parts of the bus admittance matrix with shunt elements removed), the method reduces to two separate linear systems per iteration: B′·Δδ = ΔP/V and B″·ΔV = ΔQ/V. Both B′ and B″ are constant, sparse, and factorised once, giving a per‑iteration cost roughly one‑tenth of NR. Convergence is linear and typically requires 2–3 times more iterations, but overall solution time is lower for systems above a few hundred buses.

Convergence difficulties arise from ill‑conditioned Jacobians, heavily loaded lines, poor initial guesses (flat start), and the presence of transformer tap‑changing or phase‑shifting devices that introduce discrete control variables. Practical remedies include: (i) using a continuation power‑flow (parameter‑embedding) to trace the solution path from a lightly loaded base case; (ii) applying optimal ordering (e.g., minimum degree) before sparse LU factorisation to limit fill‑in; (iii) scaling the mismatch vector to avoid overflow in poorly conditioned buses; and (iv) limiting the maximum voltage‑angle step (e.g., |Δδ| ≤ 0.5 rad) to prevent divergence. Examiners often ask for the expression of the Jacobian element ∂Pi/∂δk and the condition under which the FDLF approximation is justified.

Traveling Wave Phenomena and Surge Protection in Transmission Lines

When a sudden voltage or current change (lightning stroke, switching operation, fault inception) occurs on a transmission line, the disturbance propagates as a traveling wave governed by the telegrapher’s equations. For a lossless line the wave velocity is v = 1/√(LC) and the surge impedance is Zc = √(L/C). A forward‑traveling voltage wave V⁺(x‑vt) and current wave I⁺ = V⁺/Zc meet a discontinuity (line termination, tower, substation) where part is reflected and part transmitted. The reflection coefficient for voltage at a termination of impedance Zt is Γ = (Zt – Zc)/(Zt + Zc). An open‑circuit (Zt → ∞) yields Γ = +1 (voltage doubles), while a short‑circuit (Zt = 0) gives Γ = –1 (voltage cancels).

Lightning‑induced surges are modelled as a double‑exponential current source i(t) = I₀(e–αt – e–βt) injected at the strike point. The resulting voltage across the line insulation is the convolution of the source with the line’s impulse response. Protection coordination uses surge arresters (metal‑oxide varistors) whose V‑I characteristic follows V = k·Iα (α ≈ 0.02–0.05). The arrester is placed at the substation entrance; its protective level must be below the basic insulation level (BIL) of the equipment. The energy absorbed during a surge is W = ∫ v(t)i(t) dt and must not exceed the arrester’s rated energy capability.

Switching surges, especially during line energisation or reclosing, produce lower‑frequency (tens of kHz) overvoltages that can resonate with shunt reactors and transformer magnetising inductance. The Ferranti effect on long, lightly loaded lines raises the receiving‑end voltage to Vr = Vs/cos(βl), where β = ω√(LC). Mitigation includes controlled closing (point‑on‑wave switching), pre‑insertion resistors, and shunt reactors sized to limit the voltage rise to ≤ 1.05 p.u. Traveling‑wave‑based fault location uses the arrival time of the first wavefront at both ends: d = v·(t₁ – t₂)/2. Modern digital relays sample at ≥ 1 MHz to resolve the wavefront within a few metres.

Worked Example: Newton‑Raphson Jacobian Element
For a two‑bus system (bus 1 slack, bus 2 PQ) with line admittance y = g + jb and bus‑2 voltage V₂∠δ₂, derive the Jacobian element ∂P₂/∂δ₂ and evaluate it at the flat‑start (V₂ = 1.0 p.u., δ₂ = 0).
Active power at bus 2: P₂ = V₂²g – V₂V₁(g cos δ₂ + b sin δ₂). With V₁ = 1∠0, ∂P₂/∂δ₂ = V₂V₁(g sin δ₂ – b cos δ₂). At flat‑start V₂ = 1, δ₂ = 0 ⇒ ∂P₂/∂δ₂ = –b. Since b = –Im(y) for a lossless line, the element equals the line susceptance B. This matches the H sub‑matrix entry used in NR and the constant B′ matrix in FDLF.
Exam‑Day Strategy
When a load‑flow question asks for the number of NR iterations, remember that quadratic convergence means the mismatch roughly squares each step once near the solution. Write the iteration formula, state the stopping criterion (e.g., max |ΔP|,|ΔQ| < 10⁻⁴ p.u.), and note that a flat start typically needs 4–6 iterations for a 100‑bus system. For FDLF, emphasise the constant Jacobian factorisation and the R/X ≪ 1 assumption; if the network has significant resistance (distribution), FDLF may diverge.

Power Quality: Harmonic Distortion, Flicker, and Mitigation Techniques

Power quality (PQ) standards (IEEE 519, IEC 61000‑4‑30) quantify voltage and current deviations that affect sensitive equipment. Total Harmonic Distortion (THD) for voltage is defined as THDV = √(∑h=2^∞ Vh²) / V₁, where Vh is the RMS value of the h‑th harmonic. IEEE 519‑2022 recommends THDV ≤ 5 % at the point of common coupling (PCC) for systems ≤ 69 kV, with individual harmonic limits decreasing with harmonic order (e.g., 3 % for 3rd, 1.5 % for 5th). Current distortion limits depend on the short‑circuit ratio (ISC/IL) at the PCC.

Flicker is the perception of light‑intensity fluctuation caused by rapid voltage magnitude changes. The short‑term flicker severity Pst (10‑min observation) and long‑term Plt (2‑h) are computed per IEC 61000‑4‑15 using a statistical model of a 60 W incandescent lamp. A typical planning limit is Pst ≤ 1.0. Major flicker sources are arc furnaces, rolling mills, and large motor starts. Mitigation includes static VAR compensators (SVC) or static synchronous compensators (STATCOM) that inject reactive power to stabilise the PCC voltage, and series dynamic voltage restorers (DVR) for deep sags.

Harmonic mitigation strategies are classified as passive, active, and hybrid. Passive filters (single‑tuned, double‑tuned, high‑pass C‑type) provide low‑impedance paths for targeted harmonics (typically 5th, 7th, 11th, 13th) and also supply reactive power. Their design must avoid parallel resonance with system inductance; the filter tuning frequency is set slightly below the harmonic (e.g., 4.7 × fundamental for the 5th) to account for component tolerances. Active power filters (APF) based on voltage‑source converters inject compensating currents icomp = iload – ifund using instantaneous p‑q theory or synchronous reference frame (dq) control. Hybrid filters combine a small passive filter for the dominant low‑order harmonics with an APF for the remaining spectrum, reducing the APF rating. Examiners frequently test the calculation of filter reactor and capacitor values: L = 1/(ωh²C) and the required kVAR rating Q = V²·ωhC.

Renewable Energy Grid Integration: Fault Ride‑Through, Voltage Support, and Grid Codes

Modern grid codes (e.g., ENTSO‑E NC RfG, IEEE 1547‑2018, Indian CEA Grid Code) mandate that wind turbines and solar PV plants remain connected during voltage dips and provide dynamic voltage support. Fault Ride‑Through (FRT) capability is defined by a voltage‑versus‑time profile: the plant must stay online for voltage dips down to 0 p.u. for up to 150 ms (type 1) or 0.2 p.u. for 620 ms (type 2). During the fault, the converter must inject reactive current proportional to the voltage dip, typically Iq = K·(1 – Vpcc) with K = 2 p.u. (i.e., 2 p.u. reactive current at zero voltage). This “voltage‑dependent current injection” helps the grid voltage recover.

For doubly‑fed induction generators (DFIG), the rotor‑side converter (RSC) and grid‑side converter (GSC) are coordinated. The RSC controls rotor currents to maintain torque, while the GSC regulates DC‑link voltage and supplies reactive power. During a severe dip, the RSC may saturate; a crowbar circuit across the rotor windings protects the converter but temporarily turns the DFIG into an induction machine consuming reactive power. Modern “crowbar‑less” designs use advanced current‑limiting control to avoid this loss of voltage support.

Full‑scale converter (Type 4) wind turbines and PV inverters inherently provide FRT because the machine is decoupled from the grid. Their control hierarchy includes: (i) an outer DC‑link voltage loop (for wind) or MPPT loop (for PV); (ii) an inner current loop in dq frame with bandwidth 10–20 × fundamental; (iii) a voltage‑support overlay that modifies the q‑axis current reference during faults. Grid codes also require frequency support: active power reduction/injection according to a droop characteristic ΔP = –Kf·Δf (typical Kf = 2–5 %/Hz). Compliance testing uses hardware‑in‑the‑loop (HIL) simulators to verify the response time (< 20 ms) and the magnitude of reactive current injection.

High Voltage Engineering: Insulation Coordination, Breakdown Mechanisms, and Testing Standards

Insulation coordination ensures that the withstand voltage of equipment exceeds the expected overvoltages with an adequate safety margin. The statistical approach defines a withstand voltage Vw (e.g., 95 % probability of withstand) and a protective level Vp of the surge arrester such that Vp ≤ 0.8·Vw. The coordination margin accounts for the arrester’s residual voltage, lead‑length inductance, and the steepness of the incoming surge. IEC 60071‑1/2 and IEEE 1313 provide the methodology.

Breakdown mechanisms differ by insulation type. In gases (air, SF₆), the Townsend avalanche criterion α·d = ln(1+1/γ) (α = first ionisation coefficient, d = gap distance, γ = secondary emission coefficient) predicts the breakdown field. For uniform fields, air breaks down at ≈ 30 kV/cm (dry, 1 atm); SF₆ at ≈ 90 kV/cm. In non‑uniform fields (rod‑plane), the inception voltage for corona is given by Peek’s formula Vc = m·g₀·r·ln(4h/r) where m is surface factor, g₀ = 30 kV/cm, r conductor radius, h gap. Solid dielectrics (XLPE, paper‑oil) fail by thermal breakdown (runaway heating) or electrical treeing; the time‑to‑breakdown follows an inverse power law tbd = A·E–n (E = electric field). Liquid insulation breakdown is governed by bubble formation and streamer propagation; the V‑t characteristic is often expressed as V = k·t–1/3.

Standard withstand tests include: (i) Power‑frequency withstand (dry and wet) – 1 min at 1.1 × rated voltage; (ii) Lightning impulse withstand – 1.2/50 µs wave, three positive and three negative shots, 90 % withstand probability; (iii) Switching impulse withstand – 250/2500 µs wave for voltages ≥ 300 kV; (iv) Partial discharge (PD) measurement – IEC 60270, detection sensitivity ≤ 5 pC for GIS. Type tests are performed on prototype units; routine tests on each manufactured piece include PD and power‑frequency withstand. Condition monitoring uses dissolved gas analysis (DGA) for oil‑filled transformers (key gases: H₂, CH₄, C₂H₂, C₂H₄, CO, CO₂) and online PD mapping for cables and GIS. The Duval triangle and Rogers ratios interpret DGA patterns to discriminate thermal faults (≤ 300 °C), high‑temperature thermal faults (> 700 °C), and arcing.

  • Assuming the Fast Decoupled Load Flow Jacobian is always invertible; in heavily resistive distribution networks the B′/B″ approximation fails and the method diverges.
  • Neglecting the effect of line‑charging susceptance on the FDLF B″ matrix, leading to incorrect reactive‑power mismatches.
  • Using the lossless traveling‑wave velocity v = 1/√(LC) for a line with significant resistance, which overestimates wave speed and misplaces fault location.
  • Confusing the reflection coefficient for voltage (Γ) with that for current (–Γ) when analysing surge arrester coordination.
  • Applying the IEEE 519 THD limits for voltage to current distortion without scaling by the short‑circuit ratio.
  • Forgetting that the FRT reactive‑current injection requirement is defined at the PCC, not at the converter terminals, and ignoring the voltage drop across the step‑up transformer.
  • Treating the arrester protective level as a fixed voltage rather than a function of surge current magnitude and rate of rise (dI/dt).

These advanced topics complete the core analytical toolkit required for Paper I and illustrate the depth of understanding examiners expect.

Practice Questions

← Electrical Engineering Optional: Syllabus, Sources & Strategy Electrical Engineering Paper II: Core Topics →