ECE Board Exam common mistakes
ECE Board Exam · Understand the traps and how to avoid them.
Entering a trig or exponential expression into the calculator as sin30 instead of sin(30), or leaving the calculator in radian mode (or degree mode) because the previous item was a calculus problem.
The Mathematics portion mixes geometry/trig items (degrees) with calculus and complex-number items (radians). Under time pressure candidates key in expressions left-to-right without bracketing and never look at the DEG/RAD annunciator, so a correct method produces a wrong number.
What to do instead
Before every item, glance at the DEG/RAD indicator and set it deliberately. Write the expression on scratch paper with explicit brackets first (e.g. sin(30), 10^(2.5), e^(-0.5)), then key it in exactly as written. Anchor with a known value such as sin(30) = 0.5 or ln(e) = 1 to confirm the mode.
Attacking age, mixture, work-rate, and motion word problems by hunting for an equation to plug numbers into, rather than defining the unknowns and translating each clause of the sentence.
These items look unfamiliar even though the algebra is elementary, and candidates recognize a 'type' from a review book and force-fit a remembered formula that does not match this problem's phrasing.
What to do instead
Use a fixed four-step template every time: (1) declare each unknown with its unit, (2) write one equation per clause of the problem statement, (3) solve symbolically before substituting numbers, (4) substitute the answer back into the original wording and confirm it satisfies every clause.
Applying active-region transistor equations blindly (Ic = beta*Ib, VCE = VCC - Ic*RC) without first testing which region the device is actually in.
Nearly every textbook bias example is deliberately set in the active region, so candidates stop performing the region check and report a VCE that is physically impossible (below about 0.2 V or above VCC).
What to do instead
Always run the region test as an explicit step: assume active region, compute Ic and VCE, then check. If VCE < ~0.3 V, re-solve using VCE(sat) ~ 0.2 V and Ic(sat) = (VCC - 0.2)/RC. If VBE < 0.7 V, the device is in cutoff and all currents are zero. State the region in your scratch work before choosing an answer.
Solving an op-amp circuit with the ideal virtual-short assumption and reporting an output voltage or power that exceeds the supply rails, without ever checking saturation.
The virtual-short algebra yields a clean gain and output that is independent of the supply, so the rail limit never appears in the equations and is easy to forget under exam pressure.
What to do instead
After computing vo, immediately compare |vo| with the saturation limit (approximately VCC - 1 V for a typical op-amp). If the computed value exceeds it, the correct answer is the clipped rail value, not the linear result. Also verify the input common-mode range and the output current limit before finalizing.
Mixing DC bias quantities with AC small-signal quantities, for example substituting the quiescent collector current directly into a signal-gain expression, or computing Av = -RC/re without first finding IEQ and re = 26 mV/IEQ.
Gain formulas are memorized as single-line results, so the intermediate DC operating point that the AC parameter depends on gets skipped, and capacitor/source substitutions are done inconsistently.
What to do instead
Force a two-column layout on scratch paper. Left column: DC analysis with capacitors open and AC sources zeroed, ending in ICQ and VCEQ. Right column: AC analysis with capacitors shorted, DC supplies grounded, gm or re computed from ICQ, then gain. Never move a number from the left column to the right without an explicit conversion step.
Carrying inconsistent unit systems through a solution - mixing lbf with lbm, using g = 9.81 m/s^2 alongside pounds, or treating kg and kgf as interchangeable.
General Engineering items are drawn from mechanics, fluids, thermodynamics, and chemistry, and the unit is frequently stated only in the problem prose, not in the formula being used.
What to do instead
Convert every given quantity to a single coherent system (SI preferred) as the first line of work, writing the conversion factor as an explicit fraction (e.g. 1 lbf = 4.448 N). Perform a dimensional check on the final expression; if the units do not reduce to the requested quantity, the setup is wrong regardless of the arithmetic.
Applying a memorized formula with the wrong sign convention - using +9.81 m/s^2 for a body rising under gravity, or writing the first law as Q + W when the problem defines work as done by the system.
Different references define work and heat flow with opposite signs, and direction words in the problem statement ('rising', 'work done on', 'heat rejected') are read quickly and discarded instead of being recorded.
What to do instead
At the top of the solution write the convention you are using, e.g. 'up = positive; work done BY the system = positive; dU = Q - W'. Translate every direction word in the stem into a sign at once, then solve. If the item defines the opposite convention, convert once at the end rather than throughout.
Using 20 log for a power ratio or 10 log for a voltage/current ratio, and treating dBm as a plain ratio instead of an absolute power level referred to 1 mW.
The 10-versus-20 distinction is a one-character difference on the page, and dBm/dBW are frequently mixed with dB in the same item, so the mistake survives a plausibility check because the number 'looks right'.
What to do instead
Anchor every decibel computation on P = V^2/R: dB = 10 log(P2/P1) and, for equal resistance, dB = 20 log(V2/V1). Convert absolute levels to linear units first - P(mW) = 10^(dBm/10), and dBm = dBW + 30 - then do all arithmetic in linear units and convert back at the end.
Applying the wrong bandwidth or modulation expression - using B = 2fm for an FM signal, applying Carson's rule to AM, or counting both sidebands for an SSB signal.
The formulas for AM, DSB, SSB, and FM look similar and are recalled as a blur; candidates pick the first bandwidth expression that comes to mind instead of first classifying the modulation scheme named in the stem.
What to do instead
Write the whole family before choosing: AM DSB-FC, B = 2fm; SSB, B = fm; FM, B = 2(delta-f + fm) with delta-f = kf*Vm; PM, B = 2(beta + 1)fm. Explicitly label the modulation type stated in the problem, write the modulation index, then select the matching line.
Confusing the Nyquist bit-rate limit with the Shannon capacity limit - using 2B log2(M) where Shannon's formula applies, or inserting the SNR in dB directly into log2(1 + SNR) without converting to a linear ratio.
Both expressions contain B and log2, so a single glance at the notation is enough to trigger the wrong one, and the dB-to-linear conversion is an easy silent omission.
What to do instead
Decide from the given data which formula is valid: Nyquist requires the number of signal levels M and noiseless assumption; Shannon requires a signal-to-noise ratio and no M. Always convert SNR from dB to a linear power ratio first (SNR = 10^(SNR_dB/10)), then substitute. If both M and SNR are given, state which limit the item is asking about before computing.