Battery Life Budgeting: From Microamps to Months

A battery-powered product spends most of its life asleep, waking briefly to measure, compute or transmit. Its battery life is the cell capacity divided by the average current, and the average is dominated by whichever of those states contributes most charge. Working it out on paper before the hardware exists shows where the energy goes and which design choice will actually extend life. This article walks through the calculation.

At a glance

225 mAh
Typical nominal capacity of a CR2032 lithium coin cell.
18 µA
Average current in the worked example below: mostly active bursts, not sleep.
521 days
Ideal life of that example on a CR2032, before derating.
2 µA
The sleep current assumed in the example. Halving it changes little.

The problem

It is natural to focus on sleep current, because the device sleeps for almost all of the time. But a radio or a sensor drawing milliamps for a few milliseconds can easily consume more charge than hours of sleeping at microamps.

The calculation also flatters. Real cells deliver less than nominal capacity at low temperature and under pulsed loads, they self-discharge, and the product stops working at a cut-off voltage well before the cell is empty.

Find the largest term in the energy budget and work on that. It is rarely the one you expected.

A worked example

Suppose a sensor sleeps at 2 µA and wakes every 10 seconds for 20 ms, drawing 8 mA while it measures and transmits. The active duty cycle is 0.02 / 10 = 0.2%, so the active contribution is 8 mA x 0.002 = 16 µA. Adding the 2 µA of sleep gives an average of 18 µA.

A 225 mAh cell then lasts 225 / 0.018 = 12,500 hours, about 521 days. Note that 16 of the 18 µA come from the active burst. Halving the sleep current gains only about 6% more life; halving the burst length or doubling the interval gains far more.

What good looks like

List every operating state with its current and how long the device spends in it per cycle.

Multiply each current by its share of the time and add the results to get the average current.

Divide usable capacity by the average, then derate for temperature, pulse loading, self-discharge and cut-off voltage.

The example energy budget
StateCurrentShare of timeContribution
Sleep2 µA99.8%2.0 µA
Measure and transmit8 mA0.2%16.0 µA
Average18.0 µA
Ideal life on 225 mAh521 days
Sleep at 2 µA; 8 mA for 20 ms every 10 seconds. Calculated, not measured.
Ideal battery life by wake interval
days
Every 1 s58
Every 5 s276
Every 10 s521
Every 30 s1278
Same example, CR2032 at 225 mAh. Calculated, before derating; self-discharge and shelf life cap the longest figures in practice.

What it changes

Informed Trade-offs

The budget shows which change buys the most life for the least effort.

Right-Sized Cell

Capacity is chosen from a calculation, not from what fits the enclosure.

Realistic Claims

A derated estimate gives a battery life figure the product can meet in the field.

Early Warning

A budget that does not close on paper will not close in hardware either.

Where to go next

Measure the real current profile on the first prototype with a tool that captures short pulses. Compare it with the budget and find out why they differ.

Check the cell’s pulse capability. Coin cells have high internal resistance, and a radio burst can pull the voltage below the brown-out threshold.

Problems this prevents

  • Optimising sleep while bursts dominate
  • Battery life quoted from nominal capacity
  • Brown-out resets during radio transmissions
  • A cell chosen before the budget was done

Discuss your project

Email support@aavico.com or use Get in Touch when the website finishes loading.