Analog for Digital Engineers · All levels
ADCs (Analog-to-Digital): Tricky Q&A
Senior interview and review questions for ADCs (Analog-to-Digital).
Section Q&A bank
Use these drills after completing all topics in ADCs (Analog-to-Digital). Answer with context, mechanism proof, artifact, owner, and release decision.
A design samples at 80 MS/s and claims 35 MHz signal bandwidth without an anti-alias filter because Nyquist is 40 MHz. Why is this unsafe?
diagram
[INT][ANALOG][DATA-CONVERTERS-ADC]
Q: A design samples at 80 MS/s and claims 35 MHz signal bandwidth without an anti-alias filter because Nyquist is 40 MHz. Why is this unsafe?
A:
Nyquist only defines alias folding boundaries; it does not suppress out-of-band energy. Any interferer above 40 MHz (or image content from analog front-end nonidealities) folds into baseband and can corrupt the 35 MHz band. Real designs need front-end filtering sized to expected blocker levels and sampling-clock phase-noise constraints, not just a nominal fs/2 check.
FOLLOW-UP TRAP: Equating Nyquist compliance with automatic anti-alias protection.Why can a pipeline ADC outperform SAR in high-speed systems even if raw comparator performance is similar?
diagram
[INT][ANALOG][DATA-CONVERTERS-ADC]
Q: Why can a pipeline ADC outperform SAR in high-speed systems even if raw comparator performance is similar?
A:
Pipeline stages process different samples concurrently, so throughput scales with stage cadence instead of full-resolution bit-cycling per sample. SAR must complete sequential decision cycles and DAC settling for each conversion, which becomes a bottleneck at high rates. Pipeline keeps throughput high at the cost of latency, residue-amplifier linearity burden, and calibration infrastructure.
FOLLOW-UP TRAP: Comparing only single-comparator speed and ignoring architectural concurrency.A delta-sigma ADC advertises very high resolution. Why might in-band performance still disappoint in a real product?
diagram
[INT][ANALOG][DATA-CONVERTERS-ADC]
Q: A delta-sigma ADC advertises very high resolution. Why might in-band performance still disappoint in a real product?
A:
The headline resolution assumes intended OSR, stable loop behavior, clean reference/clocking, and a decimation filter configuration aligned with signal bandwidth. Reduced OSR, clock jitter, idle tones, reference coupling, or insufficient decimation rejection can leak shaped noise/spurs into the passband. System integration mistakes can erase theoretical noise-shaping gains quickly.
FOLLOW-UP TRAP: Treating noise-shaping theory as guaranteed field performance.Two ADCs show similar ENOB, but one has much worse SFDR. Which is better for blocker-rich RF sensing and why?
diagram
[INT][ANALOG][DATA-CONVERTERS-ADC]
Q: Two ADCs show similar ENOB, but one has much worse SFDR. Which is better for blocker-rich RF sensing and why?
A:
The higher-SFDR ADC is usually better because blocker-rich environments are limited by deterministic spurs that can mask weak tones or create false detections. ENOB/SNDR summarizes overall noise-plus-distortion power, but SFDR isolates worst-case spur dominance, which directly impacts spectral dynamic range and detection reliability under strong adjacent signals.
FOLLOW-UP TRAP: Assuming ENOB alone predicts spectral usability in spur-sensitive applications.Q&A drill guide
diagram
SYMPTOM -> ANALOG MECHANISM -> EVIDENCE -> OWNER ACTION -> VALIDATIONSketch while answering
diagram
ADC VALIDATION FLOW
front-end assumptions -> sampler behavior -> quantization path -> metric interpretationCommon traps
Verify coherent sampling and FFT setup before root-cause claims.
Classify whether loss is noise, distortion, or folded interference.
Audit architecture-fit assumptions against workload bandwidth.
Key takeaways
Tie every analog claim to one measurable metric and one proving artifact.
Prefer minimal reversible mitigations with explicit owner and rollback criteria.
Common pitfalls
Treating all noise as one scalar instead of path and frequency dependent behavior.
Changing multiple analog knobs at once and losing causality.
Declaring closure from nominal behavior without stress replay evidence.