low-power-design

name	low-power-design
description	Optimize embedded power consumption — sleep modes, peripheral shutdown, battery life estimation, and power profiling
allowed-tools	["Read","Grep"]
effort	high

When to activate

• Optimizing battery life for embedded devices
• Designing sleep/wake strategies
• Profiling power consumption with current meters
• Estimating battery life from power budgets
• Implementing dynamic voltage and frequency scaling

When NOT to use

• For mains-powered devices
• For high-performance computing (power is secondary)
• For battery charger circuit design

Instructions

1. Power budget. Measure each component: MCU active, MCU sleep, radio TX/RX, sensors, LEDs. Create power budget spreadsheet.
2. Sleep modes. Stop (CPU halt, RAM retained), Standby (most clocks off), Shutdown (only RTC + backup RAM). Wake sources: GPIO, RTC, UART.
3. Duty cycling. Active for X ms, sleep for Y ms. Duty cycle = X/(X+Y). Lower duty cycle = longer battery life.
4. Peripheral management. Disable unused peripherals. Clock gating. Power down sensors between reads. Radio off between transmissions.
5. Software optimization. Minimize active time. Batch sensor reads. Process locally before transmitting. Use DMA for transfers.
6. Battery estimation. Battery capacity (mAh) / Average current (mA) = Hours. Account for self-discharge, temperature, load profile.
7. Profiling. Current meter (Nordic PPK2, Otii). Log at high sample rate. Identify unexpected wake sources and idle current.

Example

Power Budget: Environmental Sensor (CR2032, 225mAh)

Mode          | Duration | Current  | Energy/cycle
--------------|----------|----------|-------------
Sleep         | 295s     | 2.5 µA   | 0.206 µAh
Wake + Sense  | 50ms     | 3.2 mA   | 0.044 µAh
BLE TX        | 5ms      | 12 mA    | 0.017 µAh
Total/cycle   | 300s     |          | 0.267 µAh

Average current: 0.267 µAh / (300/3600)h = 3.2 µA
Battery life: 225 mAh / 0.0032 mA = 70,312 hours ≈ 8.0 years