NTN REACT common mistakes
NTN REACT · Understand the traps and how to avoid them.
When asked for the reactor period or the rate of power change, the candidate applies the simple prompt-jump approximation (Δρ in dollars divided by β_eff) to a small step and reports the resulting inverse-period or doubling time as if it were the actual stable period, ignoring the delayed-neutron transient that dominates the first several seconds.
The prompt-jump formula only describes the instantaneous power rise before delayed neutrons reassert control; candidates memorize it as a general period formula because it is the easiest equation to recall under time pressure, and they do not connect it to the inhour equation or the six-group kinetics behaviour.
What to do instead
Before calculating, classify the perturbation: if the reactivity insertion is small compared with β_eff, state explicitly that the prompt jump is a transient and compute the stable period from the inhour equation or a point-kinetics solution. Practice one problem of each type (sub-prompt, prompt-critical, and step vs ramp insertion) and write the governing equation next to the number every time.
The candidate confuses the signs and definitions of reactivity coefficients, reporting a positive moderator temperature coefficient as stabilizing, or mixing up the fuel temperature (Doppler) coefficient with the moderator coefficient when predicting the response to a power increase or a cooldown.
Both coefficients are usually negative in a power reactor, so candidates learn them as a single 'negative feedback' fact and lose track of which physical mechanism (resonance absorption broadening in U-238 versus moderator density change) applies to which transient.
What to do instead
Build a two-column table: mechanism, sign, time scale (prompt for Doppler, seconds-to-minutes for moderator), and the accident it governs (reactivity-initiated accident vs overcooling event). For each transient in a practice question, name the coefficient that dominates first and justify its sign from the physics, not from memory.
When asked to compute shutdown margin or to verify adequate shutdown, the candidate subtracts only the highest-worth control rod and ignores the requirement that the reactor remain subcritical with the single most reactive rod fully withdrawn and all others tripped, and forgets to account for the reactivity worth of the rod being stuck out.
Shutdown margin problems look like simple arithmetic, so candidates treat them as a subtraction exercise and skip the procedural definition of the worst-case rod-out condition, confusing it with the total rod worth.
What to do instead
Write the shutdown margin definition verbatim before computing: reactivity with all rods inserted minus the worth of the single highest-worth rod assumed fully withdrawn, expressed in the same units (pcm or dollars) and referenced to the same state (hot, zero power, equilibrium xenon). Then state whether the result meets the technical specification limit and by how much.
The candidate treats xenon and samarium transients as symmetric or reversible, predicting that power will recover automatically after a xenon burnout peak, or misjudging the timing of the post-shutdown xenon peak and the iodine pit when planning a restart.
Xenon and samarium are taught together as 'fission product poisons', so candidates blur the fact that xenon is a decaying, burnable transient poison while samarium is essentially stable and non-burnable, and they do not rehearse the time constants (iodine half-life about 6.6 hours, xenon about 9.2 hours, peak roughly 8 to 11 hours after shutdown).
What to do instead
Sketch the iodine-xenon curve from memory for a step power change and for a shutdown, marking the peak and its approximate time, and state separately that Sm-149 is stable and only builds in. Practice restart-planning questions by naming the xenon concentration trend and the required operator action at each time point.
On radiation protection calculations the candidate mixes dose quantities and units, adding an absorbed dose in gray to a dose equivalent in sievert without applying the radiation weighting factor, or reporting a committed effective dose when only an external dose was asked for.
The numerical values often look plausible, and candidates default to whichever unit they last used, missing that the weighting factor and the intake-to-commitment conversion are the whole point of the question.
What to do instead
For every dose problem, write the quantity, its symbol, and its unit in a header line before doing arithmetic (absorbed dose D in Gy, equivalent dose H = w_R·D in Sv, effective dose E = Σ w_T·H_T, committed dose for intakes). Convert explicitly and check dimensional consistency at each step; if an intake is involved, state the intake route and the dose coefficient used.
The candidate answers a shielding or attenuation question using the half-value layer as if it were a tenth-value layer, or applies the inverse-square law without accounting for the buildup factor in a thick shield, producing an answer that is off by a large factor.
HVL and TVL problems both reduce to exponential attenuation, and under exam pressure candidates grab whichever constant is memorized first; buildup is often omitted because it requires an extra table lookup they have not practiced.
What to do instead
Write the attenuation equation I = I0·B·exp(-μx) and explicitly identify whether the problem gives HVL, TVL, or μ, converting with μ = ln2/HVL = ln10/TVL. For thick shields, state that B > 1 and that ignoring it is non-conservative; practice one problem each with and without buildup so the magnitude of the correction is familiar.
When asked to classify an event or select the correct emergency action level, the candidate picks the classification based on the apparent severity of the plant symptom rather than on whether the specific EAL threshold (a measured parameter, an instrument reading, or a stated condition) has actually been met.
Candidates remember accident names and general narratives (loss of coolant, station blackout) and map them directly to an alert or general emergency, skipping the step of matching the observed value to the published threshold table.
What to do instead
For each practice scenario, extract the measured parameter and its value, then state which EAL category it falls in and why the next-higher category is not yet met. Memorize the structure of the classification scheme (notification, alert, site area emergency, general emergency) and the recognition categories, and always cite the threshold, not the narrative.
The candidate describes the response to a small-break loss-of-coolant accident or a steam generator tube rupture by listing the correct systems but in the wrong sequence, omitting the step that establishes the heat sink or the step that isolates the break, and treating the pressurizer level indication as a reliable inventory indicator during the transient.
System knowledge is learned component by component, so candidates can name every system involved but have not rehearsed the integrated event timeline, and they trust pressurizer level because it is the most familiar indicator.
What to do instead
Rehearse each design-basis event as a timeline: initiating event, automatic actuation and the signal that causes it, operator actions in order, and the success criteria. For inventory, state explicitly why pressurizer level is misleading during a transient and which indication (subcooling margin, core exit temperature, or containment sump level) is used instead.
The candidate confuses redundancy and diversity, or misidentifies which train is safety-related versus non-safety-related, and cannot explain how a single failure in one division could or could not defeat the safety function.
Redundancy, diversity, independence, and single-failure criteria are taught as a vocabulary list, and candidates match words to definitions without applying them to a specific system lineup.
What to do instead
For each major safety system, draw its divisions from memory and label each component as safety-related or not, then run a single-failure thought experiment: fail one component and state whether the safety function is still met and by what path. Practice explaining why diversity (different technology or different physical principle) adds protection that a second identical train does not.
On instrumentation and control questions the candidate confuses the measured variable with the controlled variable, or misreads a setpoint, deadband, or fail-safe direction, for example assuming a valve fails open when the design basis requires fail-closed, or misidentifying the direction of a bistable trip on loss of instrument power.
Loop diagrams and setpoint tables are read quickly under time pressure, and candidates rely on the everyday meaning of 'fail-safe' rather than on the plant-specific failure direction that protects the safety function.
What to do instead
For every instrument question, state the measured variable, the sensor type, the trip setpoint and its units, the deadband, and the failure direction on loss of power or signal. Practice reading a simple loop from sensor to final element and confirm at each block whether the signal is direct- or reverse-acting before predicting the valve response.
The candidate answers a regulatory or documentation question from general industry practice rather than from the specific requirement, for example describing the content of a technical specification limiting condition for operation or the reporting timeline for a license event without distinguishing the required action, the completion time, and the required notification.
Regulatory material is text-heavy and candidates skim it, retaining the intent of a requirement but not its structure, so they blend action statements, surveillance requirements, and reporting obligations into one answer.
What to do instead
Learn the anatomy of a limiting condition for operation: applicability, action, completion time, and surveillance requirement, and practice dissecting one entry into those parts. For reporting, build a short table of event categories with their notification timeframes and the document that must be submitted, and cite the requirement rather than the practice.