IESL Part II Mechanical common mistakes
IESL Part II Mechanical · Understand the traps and how to avoid them.
Writing the energy or Bernoulli equation without first fixing a datum and drawing a control volume, then losing a static-head or velocity-head term when the flow changes area or elevation.
Candidates carry over the habit of 'plugging into the formula' and treat the equation as a list of numbers rather than a statement about a defined system; under time pressure the datum and the sign of the elevation term get skipped.
What to do instead
Before any calculation, sketch the system, mark the datum at the lowest point, label stations 1 and 2, and write the full equation symbolically with every term including losses. Only then substitute values. Check that head terms carry consistent units (m or kPa) before summing.
Treating a shaft or beam as if it carried only bending or only torsion, and adding stresses arithmetically instead of combining them at the critical point on the cross-section.
Textbook problems usually isolate one loading mode, so candidates never build the habit of locating the point of maximum combined stress and computing principal stresses there.
What to do instead
For every shaft or beam, identify the critical section (highest moment and torque together), compute bending and torsional shear at the same point, then combine via Mohr's circle or the principal-stress formulas. State the chosen point explicitly in the answer.
Sizing a component against yield strength alone without applying the correct factor of safety basis or checking fatigue, even when the problem states cyclic loading, stress concentration, or a surface finish.
Candidates default to the simple 'stress < yield/FoS' check because it is fast, and they miss that endurance-limit modification factors (surface, size, load, notch) must all be applied.
What to do instead
Read the loading description first and classify it as static or fatigue. For fatigue, multiply the endurance limit by every applicable Marin factor and use the modified Goodman or Soderberg criterion. Show the factor chain so partial credit is visible.
Selecting a standard component (bearing, belt, gear, fastener) from a catalogue using nominal load only, ignoring the application service factor, life exponent, or duty cycle given in the problem.
Catalogue selection feels like a lookup task, and candidates skip the adjustment step because the base rating number is the most visible figure in the table.
What to do instead
Always convert the stated load to a design load first (apply the service factor, then the life exponent for bearings), and only then compare against the catalogue rating. Write the adjusted design load next to the rating so the examiner sees the reasoning.
Mixing gauge and absolute pressure, or using the wrong gas constant and temperature scale in thermodynamics problems, producing answers off by roughly one atmosphere or by a factor of the molecular weight.
Problems often state pressures in bar or kPa gauge while property tables are absolute, and candidates substitute Celsius directly into ideal-gas relations.
What to do instead
Convert every pressure to absolute immediately at the start of the solution and every temperature to kelvin. Use the specific gas constant R = Ru/M for the actual gas, not the universal constant, and annotate the conversion in the working.
Computing a heat-exchanger or cycle performance figure with the wrong mean temperature difference or by omitting the correction factor for a non-counterflow arrangement.
Candidates memorize the counterflow LMTD formula and apply it to every configuration, forgetting that multipass and crossflow units need an F correction and that the end temperatures must be checked for a valid approach.
What to do instead
Identify the flow arrangement from the problem statement, compute LMTD from the terminal differences, then multiply by the appropriate correction factor F (or use effectiveness-NTU when outlet temperatures are unknown). Verify both terminal differences are positive before proceeding.
Using a grossly wrong friction factor or assuming fully turbulent flow without checking the Reynolds number against the stated velocity, diameter, and fluid.
Candidates reach for a single friction-factor value or the smooth-pipe curve without computing Re, and the Moody chart region changes the answer by tens of percent.
What to do instead
Compute Re first and state whether the flow is laminar, transitional, or turbulent. Select the friction factor method accordingly (64/Re for laminar, Colebrook or Swamee-Jain for turbulent) and record the relative roughness used.
Presenting a final numerical answer with no units, or with units that are dimensionally inconsistent with the requested quantity (for example a stress reported in newtons).
Candidates focus on getting the number and treat units as decoration, especially when working quickly through multi-step calculations.
What to do instead
Carry units through every line of working and do a dimensional check of the final expression before substituting numbers. State the unit in the final answer box and confirm it matches what the question asked for.
Ignoring the stated manufacturing or assembly constraint (tolerance stack-up, fits and clearances, surface finish, material availability) and proposing a design that cannot actually be built or assembled.
Candidates treat design questions as pure stress calculations and overlook that the exam expects an appreciation of producibility and standard fits.
What to do instead
Before finalizing a design answer, run a short tolerance stack-up on the critical dimension, specify a standard fit (for example H7/g6) where relevant, and note the manufacturing process implied. Add one sentence on why the chosen tolerances are achievable.
Running out of time because the candidate spends too long on a single long calculation and leaves short conceptual or descriptive questions unanswered.
Marks per minute vary widely across question types, and candidates work in the order presented rather than the order of return on effort.
What to do instead
Do a first pass answering all short and conceptual questions, flagging long calculations. Allocate a fixed time budget per section and abandon any single item that exceeds it, returning only after every question has an attempted answer.