PEC EPE Chemical common mistakes
PEC EPE Chemical · Understand the traps and how to avoid them.
Substitutes mass fractions (wt%) directly into Raoult's law, Dalton's law, or an x-y equilibrium plot, and carries kg/h through a recycle or purge calculation that requires kmol/h.
Real stream tables and datasheets are usually in kg/h and wt%, and under time pressure the numbers are copied straight into equations that need mole fractions. The resulting error is small enough to look plausible, so it survives a quick sanity check.
What to do instead
Before any VLE or reactive balance, rebuild the stream table in kmol/h and mole fraction, showing the divide-by-MW conversion as an explicit working line. Write the basis in capitals at the top of the page (MOLES) and treat any wt% appearing mid-solution as a trigger to convert.
Computes a reactor or separator energy balance using only the standard heat of reaction at 25 C, or an adiabatic temperature rise as dT = (-dH_rxn)/Cp without adding the sensible enthalpy between 25 C and the actual feed/product temperatures.
Heats of formation are tabulated at 298 K and the shortcut often matches a numeric distractor. Candidates also drop the minus sign on an exothermic dH, or mix dH (open, steady-flow) with dU (constant volume).
What to do instead
Write the full enthalpy path: reactants at T to reactants at 298 K, reaction at 298 K, then products from 298 K to T, with each mCpdT term shown. Choose dH for steady-flow and dU for constant-volume cases, then check the sign by asking whether the outlet stream should be hotter or colder.
Uses the Darcy friction factor in the Fanning equation (or the reverse), producing a factor-of-four error in pressure drop, and mixes diameter and radius between the Reynolds number and other correlations.
Both quantities are called 'friction factor' in textbooks and on the Moody chart, and the answer options often differ by roughly that factor, so the slip is not obvious. Under time pressure the head-loss form f(L/D)(v^2/2g) is recalled without checking which f it expects.
What to do instead
Write f_D = 4 f_F at the top of the calculation and note which chart or correlation you are using. Use Re = rhovD/mu with D the inside diameter in consistent units, and cross-check the pressure drop with a second form (head loss versus Darcy-Weisbach) before selecting an option.
Applies the plain LMTD to a multipass or cross-flow exchanger without the correction factor F, pairs the terminal temperatures incorrectly (hot-end difference against cold-end difference), and uses an arithmetic mean when the two end differences are far apart.
Counter-current LMTD is the drilled formula and F is an extra step; the wrong pairing still produces a number close to a distractor. Classroom examples usually have similar end differences, so the arithmetic-mean habit carries over unnoticed.
What to do instead
Sketch the four terminal temperatures and label hot-in, hot-out, cold-in, cold-out before subtracting, taking both differences at the same end. Apply F for multipass and cross-flow and reject any design with F below about 0.75. Fix the area basis explicitly: U_i with A_i, or U_o with A_o, never mixed.
Reads the number of equilibrium stages at total reflux (or at minimum reflux) off the McCabe-Thiele construction and reports it as the design stage count, and uses a single constant relative volatility for a system where alpha varies strongly across the column.
The total-reflux staircase is the easiest to draw because the operating line lies on the diagonal, so candidates start and stop there. Minimum reflux is often part of the same question and the two answers get conflated.
What to do instead
Always draw the operating line for the stated reflux ratio R, y = [R/(R+1)]x + x_D/(R+1), and step off between that line and the equilibrium curve. Use the pinch point to obtain R_min, report N at the actual R (typically 1.1 to 1.5 R_min), and cross-check with the Gilliland correlation or a shortcut method.
Sizes a gas-phase reactor with the constant-density liquid-phase design equation, ignoring the expansion factor epsilon, and uses the first-order conversion expression for a second- or zero-order reaction.
The familiar V = F_A0X/(-r_A) form is memorised for constant density, while exam items often add inerts or a mole change so that volumetric flow becomes v0(1 + epsilonX). The reaction order is read from the units of k but not carried into the integrated form.
What to do instead
Read the units of k first to fix the order, then write the rate law plus a stoichiometric table including epsilon = y_A0*delta. Decide constant versus variable density explicitly before integrating; use exit conditions for a CSTR and integrate for a PFR. Sanity-check that V_CSTR exceeds V_PFR for positive order.
Adds depreciation back as a cash inflow while also deducting it as an expense, discounts a uniform annual cost with a single-payment factor instead of the annuity factor, and forgets working-capital recovery at the end of the project life.
Depreciation is non-cash and matters only through the tax shield, so the add-back rule from cash-flow accounting is misapplied. The time-value factors look alike and are chosen by pattern rather than by the shape of the cash flow.
What to do instead
Draw a dated cash-flow timeline first: capital and working capital at t = 0, working capital recovered at t = n, annual revenues and costs as a uniform series, salvage at t = n. Use (P/A,i,n) for series and (P/F,i,n) for single amounts, compute tax on (revenue - cash operating cost - depreciation), and confirm NPV at i = 0 equals the undiscounted sum.
Sizes a pressure-relief device for a fire or blocked-outlet case using gauge set pressure and applies the 10% overpressure to that gauge value, and uses the vapour relief formula for a liquid-full vessel undergoing thermal expansion.
Datasheets and P&IDs quote set pressure in barg, while the relief correlations need absolute inlet pressure. The fire case has several sub-cases, and the liquid thermal-expansion relief rate is small and easy to overlook.
What to do instead
Convert to absolute before any relief calculation: P1 = 1.10 x (set pressure in bara) plus atmospheric, with the overpressure applied consistently. Identify the governing scenario (blocked outlet, external fire, thermal expansion, control-valve failure) and use the matching API 520/521 method; for liquid expansion use the volumetric rate from heat input and cubical expansion, not the vapour equation.
Confuses controller gain Kc with proportional band (treats PB = 2% as Kc = 2), mixes reset units (repeats per minute against minutes per repeat), and specifies a fail-open control valve where the safe state is closed.
Gain and proportional band describe the same proportional setting but are reciprocals, and tuning rules from different vendors use different reset conventions. Valve fail-safe action looks like a control question but is really a hazard-evaluation question.
What to do instead
Convert once and write it down: Kc = 100/PB%. State the reset units next to every number, noting that repeats/min = 1/(min/repeat). For valve action ask what happens on loss of air or signal, then choose air-to-close for fuel, steam and feed lines, and air-to-open only where closure creates a worse hazard.
Specifies 304 or 316 stainless steel for a chloride-bearing or wet CO2/H2S stream, and treats a corrosion allowance as protection against pitting, crevice attack or stress-corrosion cracking.
Stainless is the default 'corrosion-resistant' answer, and uniform-corrosion thinking (add a few millimetres of allowance) is the only model many candidates have. Localised mechanisms are governed by alloy chemistry, temperature and stress, not by wall thickness.
What to do instead
Match mechanism to material before selecting: chloride-bearing service at temperature calls for duplex or 6Mo rather than 300-series austenitics; wet CO2/H2S service requires checking NACE MR0175/ISO 15156 hardness limits. State explicitly that a corrosion allowance addresses only general thinning, and specify coating, cathodic protection or an alloy change for localised attack.