ToolsBattery life
Free tool · power budget
Battery life calculator
Model your device’s duty cycle, pick a cell, and see how long it lasts, where the energy actually goes, and what to fix first. Live as you type.
What to fix first
Want a second opinion on this power budget? An engineer can tell you where the microamps are hiding.
free review
Want an engineer to check this power budget?
Send the result to yourself, and get a free 15-minute review of your power budget with an engineer. No sales sequence.
How this is calculated
- Duty: each state’s share of the cycle (time ÷ period, or your %). Sleep is whatever is left.
- Load power per state = current × duty × rail voltage (the pack voltage when “Direct”).
- Regulator loss: LDO = average load current × (Vbattery − Vout) + Iq × Vbattery. Switching = load power × (1/η − 1) + Iq × Vbattery.
- Self-discharge as an equivalent current: rated mAh × %/month ÷ 730 h (a linear approximation).
- Battery current = total power ÷ nominal pack voltage. Usable capacity = rated × (1 − temperature) × (1 − cutoff) × (1 − aging); alkaline cells are also derated for your average drain from the Energizer capacity charts [3].
- Life = usable mAh ÷ battery mA. The bar shows each item’s share of battery power.
- Peak is the largest state current seen by the battery; for a switching regulator it is taken at the end-of-discharge voltage, where it is worst.
- Shipping: Wh = nominal V × rated Ah per cell, times cells. Lithium-metal cells are judged by grams of lithium instead [9].
How srvo helps
Get the power budget right before the enclosure is frozen.
We design battery-powered products end to end: current profiling on real hardware, firmware that actually sleeps, regulator and cell selection, and battery packs that ship legally.
Our EV battery work went from pre-owned OEM modules to four sealed, armored, liquid-cooled packs with a BMS and pressure control. On a powered respirator helmet we packed a blower, Peltier cooling and air-quality sensing into something you wear.
FAQ
Questions founders ask
Why does my real battery die sooner than the calculation says?+
Usually one of five things: the sleep current is higher than the datasheet’s “typical” (an ungated sensor, a pull-up, a regulator’s quiescent current); the firmware wakes more often than you think; it is colder than the lab; the cell’s rated capacity was measured at a gentler load than yours; or self-discharge over a long life. A CR2032 is rated at a 0.19 mA load [1], and an AA delivers roughly half its capacity at 500 mA compared with 25 mA [3].
Measure the whole cycle with a power profiler that spans nanoamps to hundreds of milliamps, then put those numbers in here.
LDO or buck converter: which gives longer battery life?+
An LDO passes the battery current straight through and burns the voltage difference as heat, so its efficiency is roughly Vout ÷ Vin: about 89% making 3.3 V from a 3.7 V cell, but only 66% from 5 V. It is quiet, cheap, and some parts draw only tens of nanoamps.
A switching converter runs at 85–95% under load, but its efficiency drops at microamp loads and its quiescent current runs 24/7, so check the efficiency curve at your sleep current, not the headline figure [8]. For coin cells, running the load directly from the cell is often best if every part tolerates the full voltage range.
Can a CR2032 coin cell handle my radio’s peak current?+
Often, yes. The oft-quoted 15 mA “limit” is not a hard wall: TI measured only about 9% less usable capacity at 30 mA peaks than at 15 mA. The real problem is internal resistance, which rises as the cell drains and in the cold, so the voltage sags under each pulse until the device browns out. A capacitor across the cell carries the pulse and improved usable capacity by more than 40% with poor-quality cells [2].
How do I measure my device’s current profile?+
Use a power profiler or source-measure unit that covers nanoamps to hundreds of milliamps with microsecond sampling, and capture several full wake cycles. Measure sleep separately and for longer: it is the number most often wrong, and it frequently dominates the budget. A shunt resistor and an oscilloscope work for the active pulses if that is what you have.
Does battery size change how we can ship the product?+
Yes. Lithium-ion is regulated by watt-hours (nominal volts × amp-hours): in the US, up to 20 Wh per cell and 100 Wh per battery qualifies for the smaller-battery exception, and the rating must be marked on the case. Lithium-metal cells such as coin cells are regulated by grams of lithium: up to 1 g per cell and 2 g per battery [9].
Every lithium cell must be of a type that passed UN 38.3 testing, and the maker must give you a test summary [10]. Air shipments add IATA packing and state-of-charge rules on top; check with your carrier before you finalise the pack.
Sources & assumptions
Defaults are starting points drawn from these sources, not quotes for your product. Every number above is editable; replace them with your own quotes as soon as you have them.
- Energizer CR2032 product datasheet — Li/MnO₂, 3.0 V; 235 mAh to 2.0 V at a 15 kΩ (0.19 mA) continuous load; ~1%/year self-discharge; 0.109 g lithium; −30 to 60 °C.
- Texas Instruments SWRA349 — Coin cells and peak current draw — 30 mA peaks cost only ~9% capacity vs 15 mA across six vendors; a capacitor across the cell improved usable capacity by >40% with poor-quality cells; average current is what matters most.
- Energizer E91 (AA alkaline) product datasheet — capacity to 0.8 V: ~3,000 mAh at 25 mA, ~2,500 at 100 mA, ~2,000 at 250 mA, ~1,500 at 500 mA (read from the chart); 10-year shelf life.
- Energizer E92 (AAA alkaline) product datasheet — ~1,150 mAh at 25 mA falling to ~420 mAh at 500 mA (read from the chart).
- Panasonic NCR18650B datasheet — rated 3,200 mAh (min), 3.6 V nominal, 2.5 V discharge cutoff.
- A123 ANR26650M1-B datasheet (LiFePO₄) — 2.5 Ah nominal, 3.3 V, 50 A max continuous discharge.
- Battery University BU-802b — Elevating self-discharge — Li-ion ~5% in the first 24 h, then 1–2% per month plus ~3% for the protection circuit; alkaline 2–3% per year; rates roughly double per 10 °C.
- Texas Instruments SLVA740A — Selecting a DC/DC converter for maximum battery life in pulsed-load applications — quiescent current definition; why light-load (PFM / power-save) efficiency decides battery life in mostly-asleep devices.
- US 49 CFR §173.185 — Lithium cells and batteries — every lithium cell/battery must be of a type that passed UN Manual of Tests and Criteria 38.3; smaller-battery exception: ≤ 20 Wh per Li-ion cell, ≤ 100 Wh per battery; ≤ 1 g lithium per Li-metal cell, ≤ 2 g per battery; Wh marked on the case.
- PHMSA — Lithium Battery Test Summary (2024) — manufacturers and distributors must make a UN 38.3 test summary available; button cells installed in equipment are excluded from 10 May 2024.