IESL Part II Electrical common mistakes
IESL Part II Electrical · Understand the traps and how to avoid them.
Treating single-phase and three-phase power formulas as interchangeable - using P = VIcosφ for a balanced three-phase load instead of P = √3·V_L·I_L·cosφ (or per-phase V_ph·I_ph·cosφ summed over three phases).
Most candidates practise power-factor and load calculations on domestic single-phase examples, then carry the same formula into three-phase motor and distribution questions without re-deriving the phase relationship between line and phase quantities.
What to do instead
Before substituting numbers, write down which quantities are given (line or phase, star or delta) and convert explicitly. Build a one-page sheet giving P, Q and S in both line and phase forms for star and delta, and check every answer against the identity S = √3·V_L·I_L for balanced loads.
Selecting cable or protective-device ratings from the load current alone, ignoring derating factors for ambient temperature, grouping, thermal insulation, and the difference between Ib, In, Iz and I2.
Candidates learn the tabulated current-carrying capacity and stop there; the coordination rule Ib ≤ In ≤ Iz and the I2 ≤ 1.45·Iz disconnection condition are less practised, and derating tables are easy to overlook under time pressure.
What to do instead
Always run the full design sequence: design current Ib, then device nominal rating In ≥ Ib, then apply Ca·Cg·Ci·Cd to the tabulated Iz and confirm In ≤ Iz and that the device's I2 satisfies the 1.45 factor. Practise with a real cable table so the derating arithmetic becomes routine.
Miscalculating earth-fault loop impedance or failing to check that the calculated Zs is below the maximum permitted for the chosen protective device and disconnection time.
The relationship between fault current, disconnection time and the device's time-current characteristic is abstract, and candidates often compute Zs but never compare it with the tabulated maximum, or forget the supply impedance and circuit resistance contributions separately.
What to do instead
Work the check as a fixed three-step routine: compute Zs = Ze + (R1 + R2) for the circuit, read the maximum Zs for the device rating and curve from the standard table, then state pass or fail with the disconnection time. Do several worked examples for MCB, fuse and RCD-protected circuits.
Confusing real, reactive and apparent power and mishandling power-factor correction - for example adding capacitor kVAR arithmetically to kW, or computing correction kVAR as P(tanφ1 − tanφ2) with the angles swapped.
Phasor addition is counter-intuitive after years of scalar arithmetic, and the tanφ formula is memorised without the phasor diagram that justifies it, so sign and ordering errors survive into the final answer.
What to do instead
Sketch the power triangle every time, label P on the horizontal, Q on the vertical and S on the hypotenuse, and derive the capacitor rating from the diagram. Verify the result by checking that the new power factor lies between the old value and unity.
Applying the wrong transformer equivalent-circuit referral - mixing referred secondary resistance and reactance with actual secondary values, or using the turns ratio instead of its square when referring impedance.
Impedance scales with the square of the turns ratio while voltage scales linearly, and candidates routinely transfer the linear rule to impedances, especially when the ratio is expressed as a voltage ratio rather than N1/N2.
What to do instead
State the referral rule explicitly before calculating: Z referred to primary = a²·Z secondary with a = N1/N2. Draw the equivalent circuit with every referred quantity labelled, then compute voltage regulation from the referred circuit rather than from mixed-side values.
Ignoring the difference between ideal and practical op-amp behaviour - assuming infinite open-loop gain and zero output impedance in circuits where saturation, finite gain or slew rate determines the answer.
Textbook ideal-op-amp analysis is heavily drilled, so candidates apply virtual-short assumptions even when the question supplies a finite gain, a supply rail, or a signal frequency high enough for slew-rate limiting.
What to do instead
Read the given parameters first and decide which regime applies: if a finite A_OL or supply rail is stated, check for saturation and use the closed-loop gain formula that retains A_OL; if a frequency and slew rate are given, compute the maximum undistorted frequency from SR/(2πV_peak).
Misapplying Boolean algebra and Karnaugh-map minimisation - treating XOR as equivalent to OR, or grouping K-map cells diagonally instead of in powers-of-two rectangles that wrap around edges.
Everyday language equates 'or' with 'one or the other', so inclusive versus exclusive OR blurs, and K-map wrap-around grouping is a spatial skill that is rarely practised enough to become automatic.
What to do instead
Write the truth table for any expression containing XOR before simplifying, and convert XOR to A·B̄ + Ā·B when combining with AND/OR terms. For K-maps, practise maps with wrap-around groupings and always verify the minimised expression against the original truth table.
Confusing modulation bandwidth and noise trade-offs - quoting the Shannon capacity or FM Carson's rule with the wrong bandwidth term, or mixing up FM's constant-envelope noise immunity with AM's power-inefficiency argument.
The formulas look similar and the qualitative claims are taught as slogans, so candidates reproduce the slogan without the underlying spectral reasoning and then substitute the wrong bandwidth.
What to do instead
Derive Carson's rule from the sideband structure (B ≈ 2(Δf + f_m)) and Shannon's capacity from C = B·log2(1 + S/N), then compare the two bandwidth definitions. For each modulation scheme, write one sentence on what happens to bandwidth, power and noise as the index changes.
Omitting the antenna and free-space path terms in a link budget - using received power as transmit power minus cable loss only, or confusing EIRP with transmitter output power.
The decibel arithmetic is easy to chain mechanically, and candidates drop the antenna gain and spreading-loss terms because they treat the budget as a cable-loss calculation rather than a propagation calculation.
What to do instead
Write the budget as a strict chain: P_tx − L_tx + G_tx − FSPL + G_rx − L_rx = P_rx, with FSPL = 32.44 + 20log10(d_km) + 20log10(f_MHz). Compute the link margin against receiver sensitivity as the final step and state it explicitly.
Treating symmetrical components and per-unit analysis as optional bookkeeping - solving unbalanced-fault questions with single-phase reasoning, or converting to per-unit on inconsistent MVA and voltage bases.
Symmetrical components require comfort with complex phasor operators and sequence networks, which many candidates avoid; per-unit conversion then compounds the problem because base changes on each side of a transformer are subtle.
What to do instead
For faults, always draw the sequence networks and connect them according to fault type (series for single line-to-ground, parallel for line-to-line) before computing. For per-unit work, fix S_base once and derive V_base and Z_base on each voltage level from the transformer ratios, then re-check that every impedance shares the same base.