Power Systems Engineering Interview Questions

What they’re really asking

They want to confirm you have the absolute fundamentals of three-phase relationships and know that nameplate voltage is line-to-line by convention.

Strong answer structure

State that nameplate/rated voltage is line-to-line (V_LL) by convention. In a balanced wye system V_LL = sqrt(3) * V_LN, so for 480 V the phase (line-to-neutral) voltage is 480/sqrt(3) = 277 V. Note the 30-degree phase shift between line and phase voltages, and that in delta there is no neutral so loads see V_LL directly. Mention currents are the inverse: in delta, I_line = sqrt(3) * I_phase.

Likely follow-ups

• Why is 277 V a common lighting voltage in North American commercial buildings?
• Draw the phasor diagram showing the 30-degree shift.
• How do line and phase currents relate in a delta-connected load?

What they’re really asking

Checks whether you understand power formulas conceptually rather than memorizing them, especially the line vs phase quantity bookkeeping.

Strong answer structure

P = sqrt(3) * V_LL * I_L * cos(phi). Explain that total three-phase power is genuinely 3 * V_phase * I_phase * cos(phi); the sqrt(3) appears only because we substitute line-to-line voltage (V_LL = sqrt(3) * V_phase) so the user can plug in the measurable nameplate/line quantities. Clarify cos(phi) is the displacement power factor, phi is the angle between phase voltage and phase current. Mention S = sqrt(3) * V_LL * I_L and Q = sqrt(3) * V_LL * I_L * sin(phi).

Likely follow-ups

• Show the algebra that turns 3*V_phase*I_phase into sqrt(3)*V_LL*I_L.
• How does this change for an unbalanced load?
• What is the relationship between P, Q, and S?

What they’re really asking

They want to see that you understand normalization across voltage levels and the practical benefits beyond just 'making numbers smaller'.

Strong answer structure

Per-unit normalizes quantities to chosen base values: quantity_pu = actual/base. Pick S_base (often system-wide, e.g. 100 MVA) and V_base per voltage zone; then I_base and Z_base follow. Key benefits: transformer turns ratios disappear (a transformer becomes a single series impedance), per-unit impedances of similar equipment fall in predictable ranges so errors are easy to spot, and it eliminates sqrt(3) and 3 factors from three-phase calculations. Mention manufacturers give equipment impedance in percent on the equipment base.

Likely follow-ups

• What happens to per-unit impedance when you change the base?
• Why does per-unit make transformer turns ratios disappear?
• What is a typical per-unit reactance for a large generator?

What they’re really asking

Tests whether you can actually execute a base change, a daily task in fault and load-flow studies.

Strong answer structure

Use Z_new_pu = Z_old_pu * (S_base_new/S_base_old) * (V_base_old/V_base_new)^2. Voltage base is unchanged (138 kV stays the system voltage at that bus), so the voltage ratio term is 1. Z_new = 0.08 * (100/25) = 0.32 pu = 32% on 100 MVA base. Emphasize that the equipment-base impedance scales linearly with the MVA base ratio and with the square of the voltage base ratio.

Likely follow-ups

• When would the voltage-ratio term not equal 1?
• What is the actual ohmic impedance referred to the 138 kV side?
• Why do utilities standardize on a 100 MVA base?

What they’re really asking

They want to know if you understand winding connections and phase displacement, which matters for paralleling and protection.

Strong answer structure

D = HV delta, y = LV wye, n = LV neutral brought out. The 11 is the clock-hour phase displacement: LV lags/leads HV by 11 * 30 = 330 degrees, equivalently LV leads HV by 30 degrees. The clock number matters because transformers paralleled together must have the same vector group (and shift) or large circulating currents flow. The delta winding also provides a path for zero-sequence current and traps triplen harmonics. The grounded wye gives a neutral reference and a ground-fault current path.

Likely follow-ups

• Why can you not parallel a Dyn11 with a Dyn1?
• How does the delta winding help with zero-sequence and harmonics?
• What vector group is typical for a utility distribution substation transformer?

What they’re really asking

Symmetrical components are the backbone of unbalanced fault analysis; they want to see you grasp the transformation and its physical meaning.

Strong answer structure

Any set of three unbalanced phasors can be decomposed into three balanced sets: positive sequence (same rotation as system, abc), negative sequence (opposite rotation, acb), and zero sequence (all three in phase). Use the operator a = 1 angle 120 degrees. Usefulness: decouples a complicated unbalanced network into three independent sequence networks that can be solved separately, then recombined. Balanced loads and faults only excite positive sequence; unbalanced faults couple the networks. Zero sequence depends on grounding/transformer connections. Negative sequence is a key indicator for motor and generator protection.

Likely follow-ups

• What does the presence of negative-sequence current tell you about a motor?
• Why is zero sequence blocked by a delta winding?
• How do the sequence networks connect for a single-line-to-ground fault?

What they’re really asking

Tests deep fault-analysis fluency, the ability to reason about which fault type is worst-case at a given location.

Strong answer structure

For an SLG fault the three sequence networks connect in series: I_a1 = I_a2 = I_a0 = E / (Z1 + Z2 + Z0 + 3*Z_f). Fault phase current I_a = 3 * I_a0. A three-phase fault uses only the positive-sequence network: I_3ph = E / Z1. Whether SLG exceeds 3-phase depends on grounding: near a solidly grounded wye-connected source where Z0 < Z1, SLG can exceed the 3-phase current and become the worst case (which is why utilities sometimes use grounding reactors). With high Z0, SLG is much smaller. Always check both to size equipment and set ground relays.

Likely follow-ups

• When is SLG current larger than three-phase current?
• How does a neutral grounding reactor change the picture?
• How do you connect the networks for a line-to-line fault?

What they’re really asking

They want to know if you understand the transient nature of fault current and why breaker ratings account for asymmetry.

Strong answer structure

The symmetrical component is the steady-state AC RMS fault current. The asymmetrical current includes a decaying DC offset that appears because inductive circuit current cannot change instantaneously; its magnitude depends on the point on the voltage wave at fault inception (worst case when voltage crosses zero) and decays with the X/R time constant L/R. The total asymmetrical current peaks in the first half cycle and drives mechanical/momentary breaker ratings, while the symmetrical interrupting rating is checked at contact-parting time. Mention the asymmetry factor and that high X/R ratios mean slower decay and higher peaks.

Likely follow-ups

• How does X/R ratio affect the DC decay?
• Why does the worst-case offset occur at a voltage zero crossing?
• What is the difference between momentary and interrupting breaker ratings?

What they’re really asking

Protection coordination is a core skill; they want to know you understand selectivity and time-current curves.

Strong answer structure

Coordination ensures the device closest to the fault operates first, isolating the smallest section. Plot time-current curves (TCC) on log-log axes; downstream devices must clear faster than upstream for any common fault current. The coordination time interval (CTI), typically 0.3-0.4 s for electromechanical, less for digital, is the margin between curves accounting for breaker operating time, relay overtravel, and safety margin. Use pickup (51 settings), time dial, and instantaneous (50) elements. Verify selectivity across the full range of fault currents, not just one point.

Likely follow-ups

• What is the difference between a 50 and 51 element in ANSI device numbering?
• How do inverse, very inverse, and extremely inverse curves differ in application?
• What can break coordination when fault current is very high?

What they’re really asking

Differential protection is subtle; they want to see you understand restraint, mismatch compensation, and inrush blocking.

Strong answer structure

Differential protection compares current entering and leaving the protected zone; ideally they sum to zero, so any difference indicates an internal fault. For a transformer you must compensate for: the turns ratio (different current magnitudes on each side), the vector-group phase shift (e.g. delta-wye 30 degrees, traditionally corrected by CT connection or now in the relay software), and CT ratio mismatch. Use a percentage-restraint characteristic so external faults with CT saturation do not cause false trips. Critically, block on magnetizing inrush and overexcitation using second-harmonic and fifth-harmonic restraint, since inrush looks like a large internal current but is rich in second harmonic.

Likely follow-ups

• How does second-harmonic restraint distinguish inrush from a fault?
• Why use percentage restraint instead of a simple fixed pickup?
• How do modern numerical relays handle the phase-shift compensation?

What they’re really asking

Distance protection is the workhorse of transmission protection; they want to confirm you understand zone reach and coordination by impedance rather than current.

Strong answer structure

A distance relay measures V/I to estimate the impedance to the fault; because line impedance is roughly proportional to length, the relay knows how far away the fault is and trips if it falls inside its reach. Typical settings: Zone 1 reaches about 80-85% of the line with instantaneous trip (underreaching to avoid overreaching the remote bus), Zone 2 reaches about 120-150% with a short time delay to cover the rest of the line and back up the remote bus, Zone 3 provides remote backup with longer delay. Plot on an R-X diagram; use mho or quadrilateral characteristics. Pair with a communication-aided (pilot) scheme such as POTT or DCB for instantaneous clearing across 100% of the line.

Likely follow-ups

• Why is Zone 1 set to underreach rather than 100%?
• What is the difference between a mho and a quadrilateral characteristic?
• How does a POTT scheme give you 100% instantaneous coverage?

What they’re really asking

A quick check that you can move between fault MVA, source impedance, and fault current, a frequent back-of-envelope calculation.

Strong answer structure

Short-circuit MVA = S_base / Z_pu (for a bolted three-phase fault, with 1.0 pu voltage). Equivalently SCC = sqrt(3) * V_LL * I_fault. Given a utility short-circuit MVA, the source impedance in pu on your base is Z_source = S_base / SCC_MVA. Then the three-phase fault current at the bus is I_fault = I_base / Z_total_pu, where Z_total is the source plus any series impedance. Stiffer systems (higher SCC) have lower source impedance and higher fault current.

Likely follow-ups

• What does a 'stiff' versus 'weak' system mean for fault current and voltage stability?
• If the utility gives you 500 MVA SCC at 13.8 kV, what is the bolted three-phase fault current?
• How does adding a transformer change the available fault current downstream?

What they’re really asking

Tests understanding of generator behavior during a fault and proper reactance selection for different study purposes.

Strong answer structure

After a fault the generator's apparent reactance increases over time as flux redistributes: X''d (subtransient, smallest, first ~1-2 cycles) reflects damper winding effects; X'd (transient, next several cycles to ~1 s) reflects field winding response; Xd (synchronous, steady state, largest) is the final value. Use X''d for the first-cycle (momentary) duty and breaker close-and-latch ratings, X'd for interrupting duty / contact parting time on slower breakers, and Xd for steady-state stability and continuous overcurrent. The fault current is highest at the instant of the fault and decays toward the steady-state value.

Likely follow-ups

• Why does the apparent reactance increase with time?
• Which reactance sets the interrupting duty for a 5-cycle breaker?
• How does an AVR change the steady-state fault contribution?

What they’re really asking

Power flow is the central steady-state analysis tool; they want to confirm you understand bus types and why the problem is nonlinear.

Strong answer structure

Each bus has four quantities: P, Q, |V|, and angle theta. Bus types fix two of them: slack/swing bus fixes |V| and theta (=0), provides the angle reference and balances losses (P, Q are outputs); PV (generator) buses fix P and |V| (solve for theta and Q); PQ (load) buses fix P and Q (solve for |V| and theta). Newton-Raphson linearizes the nonlinear power mismatch equations using the Jacobian (dP/dtheta, dP/dV, dQ/dtheta, dQ/dV) and iterates [delta_theta; delta_V] = J^-1 [delta_P; delta_Q] until mismatches converge. It converges quadratically, in few iterations, independent of system size, which is why it dominates over Gauss-Seidel for large systems.

Likely follow-ups

• Why does Newton-Raphson converge faster than Gauss-Seidel?
• What is the fast-decoupled load flow and what assumption makes it work?
• What happens to a PV bus if its reactive limit is hit during iteration?

What they’re really asking

They want to confirm you understand the P-theta and Q-V coupling that underlies control and the fast-decoupled load flow.

Strong answer structure

On a transmission line with X >> R, the real power transfer P = (V1*V2/X)*sin(delta) depends strongly on the angle difference delta, while reactive power Q depends strongly on the voltage magnitude difference (V1 - V2). So P is tightly coupled to angle (and therefore frequency, since angle is the integral of frequency deviation) and Q is tightly coupled to voltage magnitude. This is the basis of separate controls: governors balance real power to hold frequency; AVRs, capacitors, reactors, and tap changers manage reactive power to hold voltage. The weak cross-coupling (dP/dV and dQ/dtheta) is what the fast-decoupled load flow neglects.

Likely follow-ups

• Why does this decoupling break down on a distribution feeder with high R/X?
• How do you raise voltage at a bus that is sagging under heavy load?
• What is the relationship between frequency and system real-power balance?

What they’re really asking

Grid frequency control is fundamental to operations; they want to see you understand the hierarchy and timescales.

Strong answer structure

Frequency reflects the balance of generation and load; a deficit pulls frequency down. Primary control (governor droop, seconds) makes each unit change output proportionally to frequency deviation: droop = percent frequency change to move output 0 to 100%, e.g. 5% droop means a 5% frequency drop drives full output. Droop lets units share load without fighting each other but leaves a steady-state frequency error. Secondary control (AGC, tens of seconds to minutes) restores frequency to nominal and returns interchange to schedule by adjusting setpoints. Tertiary control (minutes and up) is economic dispatch / reserve redeployment to optimize cost and restore reserves.

Likely follow-ups

• Why does droop control leave a steady-state frequency offset?
• What is the role of system inertia in the first instants after a generation trip?
• How does AGC know how much to adjust each unit?

What they’re really asking

Transient stability is advanced; they want to see you understand the dynamic that governs whether a machine stays in synchronism.

Strong answer structure

The swing equation describes rotor dynamics: (2H/omega_s) * d2(delta)/dt2 = P_m - P_e, where H is the inertia constant, delta the rotor angle, P_m mechanical input, P_e electrical output. After a disturbance, if P_m != P_e the rotor accelerates or decelerates. Stability depends on whether the machine can reach a new equilibrium without the angle running away. Use the equal-area criterion: the accelerating area during the fault must be balanced by an available decelerating area after clearing; if not, the machine loses synchronism. This sets the critical clearing time for protection.

Likely follow-ups

• Explain the equal-area criterion graphically.
• How does the inertia constant H affect the critical clearing time?
• Why are low-inertia inverter-based systems a stability concern?

What they’re really asking

Tests breadth across stability categories and whether you can connect each to a real phenomenon.

Strong answer structure

Transient (large-disturbance rotor angle) stability: ability to stay in synchronism after a large disturbance like a fault; failure = pole slip / loss of synchronism. Small-signal stability: damping of small oscillations around an operating point; failure = growing inter-area or local oscillations (e.g. poorly damped 0.1-1 Hz modes), addressed with power system stabilizers. Voltage stability: ability to maintain acceptable voltages, governed by reactive power supply and the nose of the P-V curve; failure = voltage collapse when load demand exceeds the maximum transferable power, often seen after losing reactive support. Cite the 1996 WSCC and 2003 Northeast blackouts as voltage/angle examples.

Likely follow-ups

• What does a power system stabilizer do and what input does it use?
• How do you read the nose point of a P-V (nose) curve?
• What operating practices guard against voltage collapse?

What they’re really asking

This is a modern, high-signal question for renewables roles; they want to know you understand the inverter control paradigm shift.

Strong answer structure

A grid-following (current-source-behaving) inverter uses a phase-locked loop to synchronize to an existing grid voltage and injects current at a controlled angle; it needs a stiff voltage reference and cannot run a passive islanded grid alone. A grid-forming inverter behaves as a controllable voltage source, setting its own voltage magnitude and frequency (via droop or virtual synchronous machine control) and can establish/black-start a grid and ride through disturbances. As synchronous generation retires, system inertia and short-circuit strength drop; grid-forming inverters provide synthetic inertia, fast frequency response, and stability in weak grids, which is why operators increasingly require them.

Likely follow-ups

• What is synthetic or virtual inertia and how is it produced?
• Why do grid-following inverters struggle in a low-short-circuit (weak) grid?
• How does a virtual synchronous machine emulate inertia and damping?

What they’re really asking

Practical DER integration knowledge; they want concrete grid impacts and mitigations, not buzzwords.

Strong answer structure

Key issues: reverse power flow and overvoltage at the end of the feeder during midday low-load/high-generation; voltage flicker and rapid swings from passing clouds; interaction with legacy voltage regulators and capacitor banks causing excessive tap operations; reduced effectiveness of conventional protection because fault current direction and magnitude change; and the duck curve / steep evening ramp at the system level. Mitigations: smart inverters with Volt-VAR and Volt-Watt functions, conservation voltage reduction, energy storage to absorb midday excess and shave the evening ramp, hosting-capacity analysis, advanced or directional protection, and updated interconnection standards like IEEE 1547-2018 with ride-through and grid-support functions.

Likely follow-ups

• What is the duck curve and why does it stress dispatchable generation?
• How do smart inverter Volt-VAR functions help with overvoltage?
• Why does distributed generation complicate feeder protection coordination?

What they’re really asking

Confirms you can model lines at appropriate fidelity and understand when shunt capacitance matters.

Strong answer structure

A line is modeled as a two-port: Vs = A*Vr + B*Ir, Is = C*Vr + D*Ir, with AD - BC = 1 for a reciprocal passive line. Short line (< ~80 km): neglect shunt capacitance, just series Z, A=D=1, B=Z, C=0. Medium line (80-250 km): use a nominal pi model with half the shunt admittance at each end; capacitance now matters for voltage and charging current. Long line (> ~250 km): use the distributed-parameter / hyperbolic model with the characteristic impedance and propagation constant to capture distributed L and C accurately. Choosing the wrong model misestimates voltage regulation and the Ferranti effect.

Likely follow-ups

• What is the Ferranti effect and when is it significant?
• What is surge impedance loading and why is it a useful benchmark?
• Why does the long-line model use hyperbolic functions?

What they’re really asking

System grounding drives ground-fault current, protection strategy, and safety; they want the engineering trade-offs.

Strong answer structure

Solidly grounded: neutral directly to ground, high ground-fault current, fast clearing, lower transient overvoltages, but high arc-flash energy and the fault must trip immediately (common at utility distribution and 480 V). Low/high-resistance grounded: a neutral resistor limits ground-fault current (high-resistance to a few amps), reducing arc-flash and equipment damage and allowing the first ground fault to alarm rather than trip, improving continuity for industrial processes; but ground-fault detection sensitivity and transient overvoltage must be managed. Ungrounded: no intentional ground path, first ground fault does not trip (continuity) but creates sustained overvoltage on healthy phases and risks ferroresonance and difficult fault location; second fault becomes a phase-to-phase fault. Choice balances continuity, safety, and overvoltage.

Likely follow-ups

• Why does an ungrounded system raise the voltage on the unfaulted phases?
• When would you choose high-resistance grounding in an industrial plant?
• How do you detect a ground fault on an ungrounded system?

What they’re really asking

Tests practical voltage-regulation knowledge and awareness of modern DER complications.

Strong answer structure

An OLTC changes the transformer turns ratio in small steps (e.g. +/-10% in 16-32 steps) under load to hold the regulated bus voltage within a deadband, typically driven by a line-drop compensation scheme that estimates the load-center voltage. Traditionally it assumes unidirectional power flow from source to load. With DERs injecting power, midday reverse flow can fool line-drop compensation, cause the regulator to operate in the wrong direction, and drive excessive tap operations that wear the mechanism. Mitigations: bidirectional/cogeneration regulator modes, coordinate with smart-inverter Volt-VAR, widen deadbands, and use measurement-based rather than estimation-based control.

Likely follow-ups

• What is line-drop compensation and what can go wrong with it under reverse flow?
• Why is excessive tap-changer operation a maintenance concern?
• How do capacitor banks complement tap changers for voltage support?

What they’re really asking

CT selection and saturation are practical protection-engineering details that separate experienced engineers from textbook-only candidates.

Strong answer structure

A protection CT must reproduce current accurately up to many times rated (e.g. 20x) during faults, so it has a high knee-point voltage and saturates late; its accuracy class (e.g. ANSI C400, or IEC 5P20) specifies the burden voltage at which ratio error stays within limits. A metering CT is optimized for accuracy at normal load currents and deliberately saturates above a few times rated to protect downstream meters. The knee point is where a small increase in flux (voltage) needs a large increase in magnetizing current, i.e. the onset of saturation. Size by checking that the CT can drive the connected burden (relay + leads) at the maximum fault current without saturating, accounting for the DC offset and X/R via the saturation factor.

Likely follow-ups

• Why is CT saturation dangerous for differential protection?
• How does the DC component of fault current affect CT saturation?
• What does the ANSI relay accuracy class C200 tell you?

What they’re really asking

A bread-and-butter calculation; they want to see you reason with the power triangle and the reactive-power delta.

Strong answer structure

Real power P stays constant; the capacitor supplies reactive power so the source sees less Q. Q_initial = P*tan(acos(0.8)), Q_target = P*tan(acos(0.95)). Required capacitor kVAR = Q_initial - Q_target = P*(tan(phi1) - tan(phi2)). For 0.8 PF tan is 0.75; for 0.95 tan is ~0.329; so kVAR = P*(0.75 - 0.329) = 0.421*P. Benefits: lower line current, reduced losses (I^2*R), freed-up transformer/feeder capacity, better voltage, and avoided utility low-PF penalties. Caution against over-correction (leading PF) and resonance with system inductance.

Likely follow-ups

• What happens if you over-correct into a leading power factor?
• How can capacitor banks create harmonic resonance, and how do you avoid it?
• Why does poor power factor cost money even though the meter bills real energy?

What they’re really asking

Power quality knowledge; they want sources, effects, and practical mitigation, plus awareness of standards.

Strong answer structure

Sources: nonlinear loads, primarily power electronics, VFDs, rectifiers, switched-mode supplies, arc furnaces, and saturated transformers. Effects: transformer and conductor overheating from skin effect and eddy currents, neutral overloading from triplen (3rd, 9th) harmonics in 4-wire systems, capacitor failure and resonance, nuisance breaker trips, and motor torque pulsations. Mitigation: passive LC filters tuned to dominant harmonics, active harmonic filters, multi-pulse (12/18-pulse) or PWM front ends, line/DC-bus reactors on drives, K-rated/derated transformers, and zigzag transformers for triplens. Reference IEEE 519 for voltage and current distortion limits at the point of common coupling.

Likely follow-ups

• Why do triplen harmonics add up in the neutral of a 4-wire system?
• What does IEEE 519 limit, and at what point in the system?
• How does an active filter differ from a passive tuned filter?

What they’re really asking

Operational protection knowledge tying fault physics to a protection practice.

Strong answer structure

Most overhead-line faults are temporary (lightning flashover, branch contact, animal); auto-reclosing trips the breaker to de-energize and extinguish the arc, then recloses after a dead time, often restoring service automatically. Multiple reclose attempts with increasing delays are common before lockout for a permanent fault. On underground cables, faults are almost always permanent (insulation breakdown) and reclosing onto a faulted cable causes further damage and safety hazards, so reclosing is typically disabled or limited. Coordinate dead time with downstream fuse-saving/fuse-blowing philosophy and with synchronism checking on tie lines.

Likely follow-ups

• What is the difference between fuse saving and fuse blowing?
• Why must you use synchronism check before reclosing a tie line?
• How is the reclose dead time chosen?

What they’re really asking

Connects machine theory to practical application; they want both the cause and a comparison of starting methods.

Strong answer structure

At start the induction motor slip is 1, the rotor acts like a shorted secondary, so the locked-rotor current is typically 5-7x full-load current, with poor (lagging) starting power factor, causing voltage dips. Reduced-voltage starting methods: star-delta (reduces starting current/torque to ~1/3, simple but torque dip at transition), autotransformer starter (selectable tap, better torque per amp), soft starter (thyristor-controlled ramp, smooth but harmonics and no speed control), and VFD (best control of current and torque, can also run at variable speed, but highest cost and adds harmonics). Trade-off is always starting torque versus inrush: reducing voltage reduces current but reduces torque by voltage squared, so the load must start at reduced torque.

Likely follow-ups

• Why does starting torque fall with the square of applied voltage?
• When would star-delta starting fail to start the load?
• How does a VFD limit inrush while keeping torque available?

What they’re really asking

They want to gauge your engineering integrity, ownership, and how you handle safety-critical mistakes under pressure.

Strong answer structure

Use STAR. Situation: describe the issued setting/study and how you noticed the discrepancy (e.g. a miscoordination, wrong CT ratio, or base-MVA error found during a later review or a misoperation). Task: your responsibility to correct it without hiding it. Action: you immediately flagged it to your lead and the affected stakeholders, quantified the risk (could the device misoperate or fail to trip?), proposed an interim mitigation, re-ran the calculation, and issued a corrected revision with clear change tracking. Result: the corrected setting prevented a potential misoperation, and you added a peer-review/checklist step so the class of error could not recur. Emphasize transparency over self-protection.

Likely follow-ups

• How did you communicate the risk to non-engineers or operations?
• What process change did you put in place to prevent recurrence?
• How do you balance speed of correction against verifying you are right?

What they’re really asking

Real power studies are full of data gaps; they want to see disciplined engineering judgment, conservative assumptions, and follow-through.

Strong answer structure

Use STAR. Situation: a study or design where key data was missing (e.g. no utility short-circuit data, unknown transformer impedance, or undocumented existing equipment). Task: deliver a defensible result on schedule. Action: you used typical/standard values or manufacturer ranges, chose conservative bounding assumptions (e.g. infinite-bus and a realistic stiff source to bracket fault current high and low), documented every assumption explicitly, ran sensitivity cases, and requested the real data in parallel. Result: the design was robust across the assumed range, and you updated the study when real data arrived, confirming the assumptions were safe. Emphasize that you made assumptions transparent and bounded the risk rather than guessing silently.

Likely follow-ups

• How did you decide whether to bound high or low for a given parameter?
• What did you do when the real data eventually contradicted your assumption?
• How do you communicate assumption-driven uncertainty to a client or reviewer?