FE Electrical and Computer: Pick the Right Circuit Model

Treat every FE Electrical and Computer question as a model-selection problem first. Name the governing equation family and its stated assumptions in one written line, then compute. The scenarios, drill, and readiness checks below train exactly that decision.

Phasor analysis is legal only in sinusoidal steady state

A circuit has three solvable regimes: DC steady state, sinusoidal steady state, and transients. Phasors apply only to the second. Classify the regime before computing, because applying phasor impedance to a switching problem produces a confidently wrong number.

Phasor analysis replaces differentiation with multiplication by jω, so an inductor becomes jωL and a capacitor becomes 1/(jωC). That substitution is valid only when sources are sinusoidal at a single frequency and all transient terms have decayed. If a question specifies a switch opening or closing, or asks for a value at a specific time t, the steady-state assumption does not hold and you must write a differential equation or work in the Laplace domain instead.

Worked scenario: a series RL circuit with R = 10 Ω and L = 0.5 H is energized at t = 0 by a 100 V DC source, and the question asks for the current at t = 50 ms. The tempting mistake is to compute Z = R + jωL and divide 100 by it, yielding a complex number that answers a question nobody asked. The better decision is the first-order transient: i(t) = (V/R)(1 − e^(−Rt/L)) = 10(1 − e^(−1)) ≈ 6.32 A. The phasor answer is not merely imprecise; it solves a different physical situation than the one described.

• Classification cue: a stated frequency and no switching language points to phasors; any t = 0 event or explicit time value points to transients or Laplace.
• Classification cue: DC sources at rest mean inductors are shorts and capacitors are open circuits in steady state.

The √3 factor: line versus phase quantities in wye and delta

In a wye connection, line current equals phase current but line voltage is √3 times phase voltage. In delta, line voltage equals phase voltage but line current is √3 times phase current. Confirm the connection before applying any three-phase formula.

The √3 relationship comes from vector geometry, not from a special constant to memorize. In a balanced wye, the line-to-line voltage is the vector difference of two line-to-neutral phasors 120 degrees apart, which has magnitude √3 times either one. In a balanced delta, the same geometry applies to currents. Drawing the phasor diagram once and deriving both relationships yourself makes the direction of the factor (voltage in wye, current in delta) impossible to confuse under pressure.

Worked scenario: a balanced wye load draws from a 480 V line-to-line, three-phase source, with 10 Ω resistive per phase; find total power. The common mistake is computing P = V²/R with 480 V, treating the line voltage as if it appeared across each phase, which overstates phase power by a factor of 3. The better decision: phase voltage = 480/√3 ≈ 277 V, phase current = 27.7 A (equal to line current in wye), so P = 3 × 277 × 27.7 ≈ 23 kW, checkable as √3 × 480 × 27.7. The error is a clean factor of 3, large enough to make the wrong answer a plausible distractor.

• Wye: V_line = √3 × V_phase; I_line = I_phase.
• Delta: V_line = V_phase; I_line = √3 × I_phase.
• Single-phase formulas apply per phase only after converting to phase quantities.

Ideal op-amp gains collapse outside the linear range

The ideal op-amp model assumes infinite open-loop gain, infinite input impedance, and zero output impedance, which yields the virtual short only under negative feedback. Check both conditions, feedback and rail headroom, before quoting a gain formula.

Gain expressions like −Rf/Rin (inverting) or 1 + Rf/Rin (non-inverting) are consequences of the ideal assumptions under negative feedback, not definitions to recite. Deriving them from the virtual short plus zero input current takes under a minute and makes them reconstructible even when a question changes the topology, such as adding a third resistor or moving the source. Treat the gain formulas as outputs of the model, and the model's validity conditions as the input you actually check.

Worked scenario: a non-inverting amplifier with supply rails at ±10 V has Rf = 4 kΩ and Rin = 1 kΩ, driven by a 3 V input. The tempting mistake is to compute v_out = 5 × 3 = 15 V and move on. The better decision is to notice that a linear gain of 5 requires 15 V from an amplifier that cannot exceed its rails, so the output saturates near +10 V (slightly less in a real device). The distinction matters because answer choices for such a setup typically include both the linear result and the clipped result, and only the validity check separates them.

• Virtual short (v+ = v−) requires negative feedback and linear operation.
• Saturation is the correct answer whenever the linear result exceeds the rails.
• No input current flows into ideal inputs regardless of feedback topology.

Per-unit impedance: why transformer problems use bases

Per-unit expresses voltage, current, impedance, and power as ratios to chosen base values, with Z_base = V_base²/S_base. Its payoff is that transformer impedances become identical referred from either side, removing turns-ratio bookkeeping.

Pick a three-phase power base S_base and a line-to-line voltage base V_base, then everything follows: I_base = S_base/(√3 V_base) and Z_base = V_base²/S_base. An impedance stated as 5% on a transformer nameplate is per-unit 0.05 on the transformer's own rating base. If your chosen study base differs from the nameplate base, convert with Z_pu,new = Z_pu,old × (S_base,new/S_base,old) × (V_base,old/V_base,new)²; on a shared voltage base this reduces to a simple power-base ratio.

Worked scenario: estimate fault current on the 480 V side of a 1000 kVA, 13.8 kV/480 V transformer with 5% impedance, assuming an infinite upstream source. The tempting mistake is to treat 5% as an ohmic value and divide 480 by 0.05. The better decision: with the transformer rating as base, I_base = 1000 kVA/(√3 × 480 V) ≈ 1204 A, so I_fault ≈ I_base/Z_pu = 1204/0.05 ≈ 24 kA. The per-unit fault current is the same number on both transformer sides, which is exactly the property that makes multi-voltage network calculations tractable.

• Convert nameplate per-unit values to your study base before combining impedances.
• Per-unit quantities are dimensionless, so unit errors surface as nonsense ratios rather than wrong units.

Poles, not memorized formulas: reading control questions from the s-plane

The Laplace domain converts circuit differential equations into algebra: inductors become sL, capacitors 1/(sC), and the transfer function's poles govern transient behavior and stability. Locate the poles first; most second-order quantities follow from their position.

For a first-order system, the time constant is the reciprocal of the pole magnitude. For a second-order pair of complex poles at −σ ± jωd, the natural frequency is ωn = √(σ² + ωd²) and the damping ratio is ζ = σ/ωn; poles in the right half of the s-plane mean instability. Because the same algebra serves circuit transients, control loops, and filter questions, one s-plane toolkit covers several specification areas rather than three separate formula sets.

Worked scenario: a unity-feedback system has open-loop transfer function G(s) = K/(s(s + 4)) and you need the steady-state error to a unit step. The tempting mistake is to scan for a memorized error formula without identifying the system type. The better decision: the closed loop is Type 1 (one integrator, the pole at the origin), so the steady-state error to a unit step is zero for any positive K, while the velocity error constant Kv = K/4 governs ramp error instead. System type determines which input the error formula even applies to, and mixing them produces a specific wrong number.

• One pole: τ = 1/|pole|; 2% settling ≈ 4τ.
• Complex pair: ωn = √(σ² + ωd²), ζ = σ/ωn, percent overshoot from ζ via the standard second-order relation.
• Right-half-plane poles: unstable, independent of gain sign conventions.

A handbook drill that builds unit and assumption discipline

The NCEES reference handbook is supplied electronically during the exam, so speed comes from knowing its section structure and each equation's stated conditions, not from memorizing content. Drill with an assumption-first protocol rather than raw problem volume.

Run this drill twice a week: select ten problems you have not solved, and before computing anything, write one line per problem naming (a) the governing equation family, (b) its stated validity conditions from the handbook entry, and (c) the units of every term, converting where needed. Solve afterward. The output you are training is the classification line, not the arithmetic, so grade the line even when the calculation is routine.

Self-check rubric: for each of the ten problems, score one point each for naming the equation before solving, confirming the model assumption (steady state, feedback, connection type, system type), and verifying final units against the handbook's variable definitions. Use the running total as a learning milestone only: a consistent 28–30 out of 30 across several sessions suggests the classification habit is reliable in untimed conditions; then repeat the drill in a timed mixed set and expect the score to dip before it recovers. One administrative note: registration, scheduling, fees, and test-center logistics are handled by NCEES, so confirm current details there rather than in study materials.

• Log every drill problem in three columns: classification line, assumption check, unit check.
• Re-attempt any problem where you scored the assumption check wrong, even if the arithmetic was correct.
• Time the handbook lookup itself, not just the solution, so navigation speed becomes measurable.

A classification-first study sequence and readiness checks

Structure preparation in three passes: re-derive core relationships, then drill classification decisions, then run mixed timed sets. Sequence topics so each new domain reuses the same classification habit rather than a fresh memorization load.

First pass: work through DC circuits, first- and second-order transients, and sinusoidal steady state, re-deriving Thevenin equivalents, the √3 relationships, and the op-amp gain formulas from their base assumptions. Second pass: add power (per-unit, transformers, machines), electronics, and then signals, control, communications, and computer topics, ending each topic's problems with the written classification line from the drill. Final pass: timed mixed sets spanning all topic areas so that identifying the model is part of the problem rather than announced by the section heading.

Readiness checks to finish with: you can derive both √3 relationships from a phasor diagram without notes; you can state, for any op-amp gain formula you use, the two validity conditions it depends on; you can convert a nameplate per-unit impedance to a new base in under a minute; given any second-order transfer function, you can place the poles and read off ωn and ζ; and in a timed mixed set you can locate a needed handbook section without searching page by page. Treat these as binary self-observations, and re-run the mixed set a week later to confirm the habits held.

• Keep a single error log sorted by classification mistake versus arithmetic mistake; review the classification column first.
• Schedule the mixed timed sets early enough that a second mixed set fits before your exam date.
• Cap total time on any single practice question during timed sets and mark it for review, mirroring exam pacing.

References and further reading

Use these references to explore the concepts and check the latest information from the relevant organizations.

• ncees.org
• ncees.org

Continue your preparation

• Fundamentals Of Engineering Fe Electrical And Computer
• Study Guides