What is battery-aware design, and how does Peukert's law change how you evaluate a low-power architecture?
From PDVerse Low-Power Physical Design Mentor Guide · pdVerse Mentor Guide
Definition
Battery-aware design is the practice of optimizing a chip's power profile against the actual, non-linear discharge behavior of its battery rather than against a simplified constant-capacity assumption, because a battery's usable energy depends on the rate at which current is drawn from it.
Mentor Explanation
Peukert's law states that a battery's effective capacity decreases as discharge current increases, expressed roughly as t = C / I^k, where t is runtime, C is the rated capacity, I is discharge current, and k (Peukert's exponent, typically 1.1–1.3 for common battery chemistries) captures how much worse the battery performs under heavy load. This means a chip that draws current in a few high, spiky bursts can drain a battery measurably faster than one that draws the same total charge more smoothly over time — even though a naive average-power calculation would call them equivalent. This is why some low-power architectures deliberately smooth or spread out current draw (e.g., staggering multiple blocks' peak activity rather than letting them all burst simultaneously) purely for battery-lifetime reasons, separate from any thermal or IR-drop concern.
Example
Two IoT sensor designs both average 1 mA over a duty cycle, but Design A draws a smooth 1 mA continuously while Design B bursts to 20 mA for short active windows and sleeps otherwise. Even at identical average current, Peukert's law predicts Design B will measurably underperform Design A's rated battery life, because the battery's effective capacity drops at the higher instantaneous discharge rate during those bursts.
Why It Matters
A design that looks fine on paper — same total energy per operation as a competing design — can deliver meaningfully worse real-world battery life if its current profile is spikier, and Peukert's law is the formula that explains why "average power" alone is an incomplete metric for battery-powered products. This is a frequently-missed nuance: engineers optimize average or peak power without ever modeling how the battery itself responds to the shape of the current draw over time.
Command
# Peukert's law relationship (evaluation formula, not an EDA tool command):
# t_actual = C_rated / I^k (k ~ 1.1-1.3 typical for common chemistries)
# Compare I_burst-shaped profile vs I_avg-smoothed profile at equal total charge
# to estimate real battery-life delta before committing to an activity schedule.Common Beginner Mistake
Sizing a battery-powered design purely from a spreadsheet of average milliwatts, without ever modeling the current waveform's peak-to-average ratio. Two designs with identical average power can have very different real battery life once Peukert's law is applied to their actual discharge profiles — a design review that stops at "average power looks good" can still ship a product with worse-than-expected battery life.
Low-Power & UPF Handbook
Master Low-Power VLSI & Multivoltage Design
Read the complete low-power guide library covering power domains, level shifters, isolation clamps, state retention, and UPF signoff verification.
Continue practising