How Much LiPo Battery Capacity Does Your Device Need?

Last technical review: August 2026

Need an initial estimate first? Use THOR Power’s battery runtime and required-capacity calculators, then return to this guide to turn the estimate into a testable OEM requirement.

The Real Problem: 100 mA for 10 Hours Does Not Automatically Approve 1,000 mAh

A product team measures about 100 mA and needs 10 hours of runtime. Multiplying current by time gives 1,000 mAh, which is useful for a first screen. It is not yet a battery specification.

Real products sleep, wake, transmit, drive displays or motors and pass power through converters. Battery voltage changes during discharge, and operation ends at a defined system boundary. A battery can meet the label-mAh calculation and still miss runtime; a short peak can also be mistaken for continuous demand and force an oversized battery.

1. Define What the Runtime Target Means

Before measuring current, define the conditions under which the runtime claim must be true. Otherwise, two teams can test the same battery and produce different but technically defensible results.

• Starting condition: exact full-charge voltage, charger termination current, allowed post-charge rest time and initial state of charge. Confirm whether the design is a 4.2 V or 4.35 V lithium battery system before setting the runtime baseline.

• Ending condition: device UVLO, firmware shutdown, displayed 0%, loss of a required function or another approved product boundary.

• Use profile: operating states, duration, repetition rate, radios, display brightness, motors, heaters and user interactions.

• Environment: required low and high operating temperature, airflow and enclosure condition.

• Life point: beginning-of-life only, or the required runtime after the defined cycle/calendar-aging condition.

• Production requirement: minimum guaranteed performance, not only a typical prototype result.

A supplier cannot responsibly recommend capacity from runtime alone. The same 10-hour target can require different nominal capacity when the cutoff voltage, temperature, charger, enclosure and end-of-life requirement change.

2. Measure the Real Device Load

Most compact electronics do not draw one constant current. A wireless sensor may sleep for most of its cycle, wake for processing and transmit in short bursts. Texas Instruments notes that pulsed-load analysis must account for the time spent in all operating modes; optimizing only the heavy-load event can miss the energy consumed by long low-power states.

Source: Texas Instruments, Selecting a DC/DC Converter for Maximum Battery Life in Pulsed-Load Applications (SLVA740A)

Measure at the battery terminals when possible

A 100 mA measurement on a regulated 3.3 V rail is not automatically 100 mA from a LiPo cell. Converter efficiency, quiescent current and changing battery voltage alter the battery-side current. Measuring voltage and current at the battery terminals captures the complete power path and avoids guessing converter loss twice.

Use a current profiler or power analyzer for pulsed loads

A handheld multimeter may show a stable average while missing a millisecond radio burst or motor inrush. Record peak amplitude, pulse duration, repetition interval and battery voltage during the event. Sampling bandwidth must be high enough to preserve the event that could cause a reset or protection action.

If no battery is available yet

Use a programmable DC source with sufficient transient capability to sweep the intended LiPo battery-voltage range. Set the protective current limit above the expected normal load so the source does not create an artificial voltage collapse. Measure device input power at representative high, middle and low battery voltages. If battery source impedance is being emulated, define that impedance explicitly and verify that the source, leads and measurement shunt preserve the real transient behavior. After the electrical boundary is defined, use the 3.7V LiPo battery model range to shortlist candidates by capacity and size.

3. Calculate the First Capacity Requirement

Choose the calculation path that matches the data actually available. Do not mix battery-side current with downstream load power in the same equation without accounting for the converter and power path.

Path A: battery-side current is known

The usable fraction represents the part of nominal charge that the final device can remove before its approved cutoff under the intended load. Temperature, aging and cutoff effects are not always independent. Apply separate factors only when the evidence shows they represent distinct losses; otherwise use a measured combined usable-capacity factor. Aging also can increase resistance, so an end-of-life Energy Gate pass does not replace a repeat Power Gate at the required low SOC and temperature. At this stage, custom LiPo battery selection is still a screening step, not a model approval.

Path B: device power is known

Wh is the stronger comparison when battery voltage changes materially or when comparing different voltage platforms. mAh remains useful within the same voltage class, but mAh without voltage is not an energy comparison.

4. Worked Example: From 424 mAh to a Candidate Rating

Consider an illustrative 1S wireless device with battery-terminal current measured in three operating states:

The time-weighted current is 0.85 × 12 + 0.14 × 180 + 0.01 × 700 = 42.4 mA. For a 10-hour target, the ideal charge requirement is 424 mAh.

Now assume—not as a universal rule, but as explicit project inputs—that final-device tests or controlled cell data show:

• The device can use 85% of nominal capacity before the approved cutoff under the intended load and temperature.

• The product must still meet its runtime requirement when available capacity has declined to 80% of beginning-of-life capacity.

This calculation supports a candidate range; it does not approve a specific 624 mAh or 650 mAh cell. The selected model must still fit the finished pack envelope, support the 700 mA pulse and pass runtime validation at the required temperature and life condition. If the 85% usable factor already came from an aged-cell low-temperature test, applying separate temperature and end-of-life factors would double-count the same loss. End-of-life approval must also repeat the Power Gate because resistance growth can reduce loaded-voltage margin even when remaining capacity still passes the energy calculation. For this illustrative 624–650 mAh candidate range, review medium-capacity LiPo batteries only after the electrical and mechanical limits are defined.

5. Check Peak Current and Loaded-Voltage Margin

Capacity mainly supports operating time. The worst required current event determines whether the battery and current path can maintain sufficient loaded voltage. A low-average-current device can still reset during a radio burst, motor start, heater event or flash pulse.

Record peak current and battery/device voltage together at the required low state of charge and temperature. Also define pulse duration and repetition. A current rating without duration can be meaningless: a cell may tolerate a short event that it cannot support continuously or repeatedly without excessive heat or voltage loss.

• Confirm the exact cell’s pulse and continuous-discharge capability under its datasheet conditions.

• Confirm PCM/BMS overcurrent delay and threshold tolerances do not conflict with the normal load pulse.

• Measure voltage on both sides of the connector or protection path when diagnosing unexpected shutdown.

• Repeat the check with production-intent wire length, connector, welds and enclosure temperature.

6. Identify the Boundary That Ends Runtime

Runtime ends when the first required product function becomes unacceptable—not when every nominal milliamp-hour has left the cell. The active boundary may be device UVLO, a firmware or gauge empty point, loaded-voltage loss, or protection action.

Texas Instruments fuel-gauge guidance illustrates the system dependency: Design Capacity is a nominal starting value, while Terminate Voltage should reflect the minimum operating voltage of the actual system. Gauge configuration is device- and gauge-specific, but the broader lesson is that available runtime depends on the product boundary, not the capacity label alone.

Source: Texas Instruments, Quickstart Guide for bq27441-G1 (SLUUAP7)

7. Select a Candidate Cell and the Evidence Behind Its Rating

Ask for more than a model number and headline capacity. A useful supplier response ties the rating to the exact cell or finished pack, its characterization condition and the intended production configuration. Candidate selection may begin from a standard model or a custom configuration, but approval remains tied to the exact finished battery.

Before release, compare the controlled finished-pack drawing with the product cavity and keep-outs; nominal cell dimensions alone are not enough.

IEC 61960-3 provides performance-test and marking criteria within its published scope, but a standard reference does not replace the exact supplier datasheet and final-device test. Use the applicable product, transport and market standards separately from the runtime calculation.

Reference: IEC 61960-3:2017

8. Validate Runtime in the Final Device

Release the production-intent configuration, not a spreadsheet. The final test must connect the load model, usable energy, peak-loaded voltage, cutoff behavior, temperature, life target and mechanical assembly. Plan the sequence using the battery-pack prototype testing and validation guide.

For debugging, log battery current, battery voltage, device rail voltage, shutdown reason and relevant gauge/protection flags on the same time base. A simple elapsed-time result can prove failure but may not reveal the cause.

9. What the Buyer Should Send—and What the Supplier Should Return

Key Takeaways

• Define the approved full-charge condition, ending boundary, temperature and life point before calculating capacity.

• Use the complete LiPo battery-side duty cycle for energy sizing; treat peak current as a separate Power Gate and avoid counting converter loss twice.

• Convert ideal mAh or Wh into a candidate rating only with project-specific usable-capacity evidence, and release against minimum guaranteed capacity rather than a typical label value.

• Repeat the loaded-voltage check at low SOC, required temperature and end-of-life condition using the production-intent PCM, wire, connector and enclosure.

• Release measured usable runtime and a known shutdown boundary in the finished LiPo configuration—not nominal mAh alone.

FAQ: LiPo Battery Capacity for Device Runtime

Conclusion

A defensible LiPo capacity decision begins with the real duty cycle, not the largest or smallest current in a specification. Calculate the initial charge or energy requirement, translate it into a candidate rating with explicit project evidence, keep peak power as a separate check and identify the boundary that actually ends operation.

This workflow prevents an undersized battery that misses runtime and an oversized battery selected because a short pulse, early cutoff or measurement error was mistaken for insufficient capacity.

Technical References

1. Texas Instruments — Selecting a DC/DC Converter for Maximum Battery Life in Pulsed-Load Applications, SLVA740A, revised July 2021.
2. Texas Instruments — Quickstart Guide for bq27441-G1, SLUUAP7.
3. IEC 61960-3:2017 — Secondary lithium cells and batteries for portable applications.

Evidence standard: Published specifications must remain tied to the exact source and test condition. Numerical examples and diagrams in this article are illustrative engineering screens, not universal battery-sizing allowances or THOR Power first-party performance claims. Final selection must use the exact cell/pack documentation and measurements from the intended device, charger, duty cycle, temperature range and production configuration.