BCECE common mistakes
BCECE · Understand the traps and how to avoid them.
Treating the formula v² = u² + 2as as valid along the actual curved path of a projectile, then substituting the path length for s and getting an answer that is dimensionally plausible but numerically wrong.
The kinematics equations with constant acceleration are vector relations applied component-wise along a straight line. Candidates memorise them as 'the motion formulas' and reuse them after resolving motion into horizontal and vertical parts, forgetting that s in the vertical equation must be the vertical displacement, not the total distance flown.
What to do instead
Before substituting, write down the axis you are working along and confirm acceleration is constant on that axis. For projectiles, always split into x (a = 0) and y (a = -g) and apply the equations separately; use vertical displacement for s. Do three projectile problems where the total path length is deliberately given as a distractor and practise rejecting it.
Applying conservation of linear momentum to a collision where an external impulse (gravity, a string tension, friction) acts during the contact interval, or applying it along a direction in which the surface exerts a force.
The momentum-conservation statement is taught as 'momentum is always conserved in collisions', which is only true when net external impulse is negligible. Collisions involving a hanging mass, a bullet embedding in a block on a rough floor, or a ball bouncing off a wall violate this along the constrained direction.
What to do instead
For every collision, draw the system boundary and list external forces. If a string or wall constrains motion, conserve momentum only perpendicular to the constraint (e.g. along the wall for a ball hitting it at an angle). Where an external impulse is not negligible, use impulse-momentum: change in momentum equals the integral of external force over time.
Computing the equivalent resistance of a mixed series-parallel network by adding all resistors and dividing by the number of parallel branches, or by reducing the network in an arbitrary order without checking which nodes are actually at the same potential.
Candidates pattern-match on the visual layout of the circuit diagram rather than on node connectivity. Two resistors drawn far apart may be in parallel if their ends share nodes, and a wire with no component between two points merges those nodes; the drawing hides this.
What to do instead
Redraw the circuit labelling each node with a letter, then list each resistor by the pair of nodes it connects. Resistors sharing the same node pair are in parallel; resistors in series carry the same current and sit on a single branch between nodes of degree two. Practise by redrawing three bridge-style circuits node-by-node before computing anything.
Using the mirror or lens formula with a sign convention applied inconsistently, for example inserting a virtual image distance as positive while treating the same setup's focal length as negative, producing a magnification with the wrong sign.
Different textbooks and coaching notes use Cartesian and real-is-positive conventions, and candidates mix rules learned from different sources. The arithmetic then looks correct while the physical interpretation (real vs virtual, erect vs inverted) is wrong.
What to do instead
Commit to one convention and write it at the top of the rough sheet: distances measured from the pole/optical centre, positive in the direction of incident light. Then check the sign of magnification against the expected nature of the image. Do five problems twice, once with each convention, and confirm identical physical conclusions.
Balancing a redox equation by inspection and adjusting coefficients until atoms match, without equalising the number of electrons transferred, so the final equation conserves atoms but not charge.
Simple combustion-style balancing is over-learned from earlier classes. In ionic redox equations the spectator ions and the charge balance are invisible in an atom count, so a plausible-looking set of coefficients passes the candidate's own check.
What to do instead
Use the half-reaction method every time: split into oxidation and reduction halves, balance O with H2O and H with H+, balance charge with electrons, multiply to equalise electrons, then add and cancel. Finish by verifying both atom count and total charge on each side. Do this for one acidic and one basic medium example.
Predicting the major product of an electrophilic aromatic substitution or addition reaction without checking the directing effect of substituents already on the ring, or ignoring Markovnikov/anti-Markovnikov regiochemistry when a peroxide or a strong acid is present.
Candidates memorise named reactions as 'reactant A plus reagent B gives product C' and skip the mechanistic step where the intermediate's stability decides the outcome. Substituent effects and radical versus ionic pathways are exactly the conditions that change C.
What to do instead
For each reaction, write the intermediate (arenium ion or carbocation/radical) and identify which position gives the most stable intermediate. Explicitly note whether the reagent set generates radicals (peroxide, light) or ions. Build a one-page table of activating/deactivating groups with their o/p or m directing behaviour and redraw mechanisms rather than memorising products.
Calculating pH of a weak acid or a buffer by treating it as a strong acid (pH = -log C) or by using the Henderson-Hasselbalch equation with concentrations that have not been corrected for dissociation or for the common-ion shift.
The strong-acid formula is the first one learned and is fast to apply. Weak-acid problems require the Ka expression and often an assumption (x << C) whose validity candidates never test, so the shortcut feels legitimate.
What to do instead
Always write the equilibrium table for HA dissociation, then check whether the degree of dissociation is under about 5 percent before dropping x. For buffers, confirm the added strong acid or base has been consumed stoichiometrically before plugging into Henderson-Hasselbalch. Solve one problem by the exact quadratic and compare with the approximation.
Evaluating a definite integral of a modulus, greatest-integer, or piecewise function over an interval that crosses the point where the definition changes, without splitting the interval.
The integrand looks like a single expression and the antiderivative is computed mechanically. The change of definition (where the quantity inside the modulus changes sign, or where the floor jumps) is a property of the interval, not of the formula, so it is easy to overlook.
What to do instead
Before integrating, find the zeros or jump points of the integrand inside the limits and mark them on a number line. Split the integral at each such point and use the correct branch on each subinterval. Practise with integrals of |x - a|, [x], and min/max functions where the breakpoint lies strictly inside the limits.
Applying a differentiation or limit rule beyond its hypotheses, for example L'Hopital's rule on a limit that is not of 0/0 or infinity/infinity form, or differentiating a function given only implicitly or parametrically as if it were explicit in x.
The rule is remembered as a procedure ('differentiate top and bottom') rather than as a theorem with preconditions. Under time pressure the check of the indeterminate form, or the conversion from parametric to Cartesian, is skipped.
What to do instead
For every limit, substitute the value first and state the form out loud before choosing a method. Use L'Hopital only after confirming 0/0 or infinity/infinity, and check differentiability near the point. For implicit or parametric functions, write dy/dx = (dy/dt)/(dx/dt) or differentiate both sides with respect to x, then solve for dy/dx.
Solving a probability or permutations/combinations problem by picking a formula (nCr or nPr) based on keywords such as 'select' or 'arrange', without deciding whether order matters and whether repetition is allowed, and without checking for overcounting of identical items.
Keyword-to-formula mapping replaces the modelling step. Cases with identical objects, at-least constraints, or circular arrangements break the simple mapping, and the resulting count is off by a factorial factor that is hard to notice.
What to do instead
Before computing, answer three questions in writing: does order matter, is repetition allowed, are any objects identical. Handle 'at least one' by complementary counting. For circular arrangements fix one object and divide by symmetries; for identical items divide by the factorial of each multiplicity. Verify small cases by listing outcomes manually.