Physical Design · All levels
Power Signoff Tricky Q&A
20+ senior Power Signoff interview questions.
Q&A bank
Answer with mechanism, pitfall, regression check, and release judgment. At 10+ years, a correct definition is not enough.
Senior answer rubric
Start with the failing metric and analysis context.
Explain the physical or constraint mechanism.
Name the cheapest evidence-gathering experiment.
Choose a bounded fix and state what it can regress.
Close with signoff, waiver, or escalation criteria.
What is the practical difference between static and dynamic IR?
[INT][PD]
Q: What is the practical difference between static and dynamic IR?
A:
Static captures DC drop under average current; dynamic captures transient droop during switching peaks.
FOLLOW-UP TRAP: Assuming static pass implies dynamic pass.Why can IR closure reopen timing?
[INT][PD]
Q: Why can IR closure reopen timing?
A:
Droop lowers effective drive strength and increases delay on critical paths.
FOLLOW-UP TRAP: Treating power and timing signoff as independent.What is the most common root cause of IR hotspots?
[INT][PD]
Q: What is the most common root cause of IR hotspots?
A:
Insufficient local strap/via capacity relative to clustered switching demand.
FOLLOW-UP TRAP: Blaming tool noise without layout correlation.How do EM and IR fixes overlap but differ?
[INT][PD]
Q: How do EM and IR fixes overlap but differ?
A:
Both may widen routes or add vias, but EM is current-density lifetime risk while IR is voltage-drop margin.
FOLLOW-UP TRAP: Using one metric for both.Why are vias often EM-critical?
[INT][PD]
Q: Why are vias often EM-critical?
A:
Vertical current bottlenecks can exceed via current density even if horizontal wire looks safe.
FOLLOW-UP TRAP: Checking wire width only.How do you prioritize EM violations?
[INT][PD]
Q: How do you prioritize EM violations?
A:
Rank by EM ratio severity, net criticality, and temperature/stress profile relevance.
FOLLOW-UP TRAP: Fixing easiest nets first.What proves UPF is physically implemented correctly?
[INT][PD]
Q: What proves UPF is physically implemented correctly?
A:
Boundary cell placement/connectivity checks, always-on rail verification, and mode-wise behavior validation.
FOLLOW-UP TRAP: Only passing RTL low-power simulation.Why can isolation correctness be mode-dependent?
[INT][PD]
Q: Why can isolation correctness be mode-dependent?
A:
Clamp behavior and enable timing vary between functional, test, and low-power sequences.
FOLLOW-UP TRAP: Validating one mode only.How do leakage and dynamic optimization goals conflict?
[INT][PD]
Q: How do leakage and dynamic optimization goals conflict?
A:
VT and gating choices that reduce one component can hurt timing, wake latency, or the other power component.
FOLLOW-UP TRAP: Optimizing a single metric globally.When is vectorless analysis insufficient?
[INT][PD]
Q: When is vectorless analysis insufficient?
A:
When hotspot behavior depends on realistic switching correlations absent from generic toggles.
FOLLOW-UP TRAP: Using vectorless as sole signoff evidence.How do you assess vector quality for power signoff?
[INT][PD]
Q: How do you assess vector quality for power signoff?
A:
Measure activity coverage, scenario representativeness, and hotspot repeatability against expected workloads.
FOLLOW-UP TRAP: Trusting any simulation dump.What is rush current in one line?
[INT][PD]
Q: What is rush current in one line?
A:
Short-duration peak current during domain wake or mass clock enable events.
FOLLOW-UP TRAP: Calling it average dynamic power.Name two rush-current mitigation techniques.
[INT][PD]
Q: Name two rush-current mitigation techniques.
A:
Staged domain wake-up and controlled enable sequencing with decap/grid support.
FOLLOW-UP TRAP: Only increasing global ring width.Why should wake-up analysis include EM?
[INT][PD]
Q: Why should wake-up analysis include EM?
A:
Transient current spikes can violate EM limits even if steady-state EM is clean.
FOLLOW-UP TRAP: Checking only IR during wake.What should a power waiver include?
[INT][PD]
Q: What should a power waiver include?
A:
Hotspot scope, measured risk, mitigation timeline, owner, and expiry condition.
FOLLOW-UP TRAP: Permanent waiver with no action plan.What KPI set belongs on a power dashboard?
[INT][PD]
Q: What KPI set belongs on a power dashboard?
A:
Worst IR, EM ratio hotspots, recurrence trend, unresolved waivers, owner ETA, and regression freshness.
FOLLOW-UP TRAP: Single aggregate power number.How do you communicate power risk to leadership?
[INT][PD]
Q: How do you communicate power risk to leadership?
A:
Translate hotspots into tapeout impact, confidence level, and concrete mitigation status by owner.
FOLLOW-UP TRAP: Sending raw tool screenshots.Which cross-link is most important for power signoff?
[INT][PD]
Q: Which cross-link is most important for power signoff?
A:
Timing closure linkage, because droop-induced delay can invalidate near-zero slack margins.
FOLLOW-UP TRAP: Treating power as isolated reliability task.What is a common late-stage power anti-pattern?
[INT][PD]
Q: What is a common late-stage power anti-pattern?
A:
Fixing hotspots with broad changes that create congestion and new violations elsewhere.
FOLLOW-UP TRAP: Assuming any local green metric is safe.What is the interview-strong power debug narrative?
[INT][PD]
Q: What is the interview-strong power debug narrative?
A:
Metric -> hotspot localization -> mechanism hypothesis -> focused fix -> IR/EM/timing regression proof.
FOLLOW-UP TRAP: Command list without mechanism.How to drill this Q&A bank
Use each question as a two-minute mock. The target answer is not an essay; it is a structured closure response with a metric, mechanism, risk, and follow-up check.
Answer template
MECHANISM: what physical effect or tool step is involved
WHEN: where it appears in the flow
PITFALL: one wrong junior answer
CHECK: report/map/checklist item that proves the answerScoring
5/5: metric, mechanism, experiment, regression, and release decision named without prompting.
4/5: technically correct and names regression, but misses ownership or escalation criteria.
3/5: correct concept but no decision framework.
1/5: tool command or buzzword with no mechanism.
Principal-level review bar
Q&A bank pages in this course should be read like real closure review material. For a 10+ year PD engineer, the bar is not remembering terminology; it is making a release-quality decision under ambiguity.
What excellent looks like
Names the failing metric, corner/mode, database tag, and analysis switches before proposing a fix.
Separates data, constraint, physical, tool, and methodology root causes instead of treating all failures as optimization problems.
Chooses experiments by information gain and reversibility, not by habit.
States regression blast radius across timing, route, power, PV, DFT, package, and tapeout manifest.
Turns recurring failures into methodology guardrails, dashboards, or checklist items.
Closure note template
STAFF / PRINCIPAL CLOSURE NOTE
Context:
stage: <pre-CTS | post-CTS | post-route | post-fill | signoff>
tag: <database / netlist / SDC / library stack>
failing metric: <exact report line>
affected scope: <block / hierarchy / path group / power domain / region>
Hypotheses:
H1: <most likely physical or constraint mechanism>
H2: <competing explanation>
H3: <methodology or input-data issue>
Decision:
next experiment: <cheap check that can falsify H1>
fix candidate: <minimal reversible change>
rollback trigger: <metric that says the fix is wrong>
regression set: <timing / route / power / PV / DFT / package>
escalation owner: <team or reviewer>Tradeoffs a senior engineer must discuss
Technical tradeoff
Power signoff validates whether the physical grid supports real activity. Explain not only the preferred fix, but what margin or schedule you are spending to get it.
Cross-team tradeoff
What must RTL, synthesis, CAD, STA, DFT, package, IP, or foundry agree to before this decision is final?
Which artifact becomes the source of truth after the decision: report, waiver, manifest, ECO script, or methodology deck?
What is the cost of being wrong: one rerun, ECO churn, mask risk, performance loss, or silicon escape?
Leadership communication
"The current blocker is <metric> in <corner/mode/stage>. The leading cause is <mechanism>. I recommend <fix> because it is bounded and reversible. The regression surface is <domains>. If it fails, we escalate to <owner> with <evidence>."Key takeaways
Always connect the concept back to a measurable signoff artifact.
A fix is not complete until you can name the regression checks.
Common pitfalls
Optimizing by habit instead of reading the current report.
Forgetting that a local fix can regress timing, routing, power, or PV elsewhere.