A battery is the part of a small energy system that lets generation and demand happen at different times: it stores energy when there is a surplus and returns it when there is a shortfall. Choosing one well means answering several questions at once: how much energy, how much power, which chemistry, how it is protected, and how long it must last. Those answers interact. This guide builds the mental model behind all of them, and points to the supporting articles that work each number in detail.

It assumes you can already tell energy from power and read a capacity rating.

What a battery stores

A battery stores energy in chemical form and releases it as direct current (DC) electricity. In a lithium-ion cell, energy is held in the difference between two electrodes: during discharge, lithium ions move from the anode through an electrolyte to the cathode while electrons flow through the external circuit to do useful work; charging drives the ions back. The reaction is reversible, which is what makes the cell rechargeable, and it delivers DC — one reason inverters exist, to turn that DC into the AC many appliances expect.

Two consequences follow immediately and shape everything else:

  • Because storage is chemical, a battery's behaviour depends on temperature, on how hard and how often it is worked, and on age — not just on its nameplate.
  • Because it delivers DC at a voltage set by its chemistry and cell count, the same stored energy can be packaged as different voltages (12 V, 24 V, 48 V), a choice that ripples through the rest of the system.

Energy is not power

The most useful distinction is between the two things a battery is rated for:

  • Energyhow much it holds, in watt-hours (Wh) or kilowatt-hours (kWh). This sets how long it can run your loads.
  • Powerhow fast it can deliver that energy, set by its maximum current (often expressed as a C-rate). This sets what it can run at once.

These are independent ratings. The example battery used throughout this guide — a 12.8 V, 100 Ah LiFePO₄ pack — stores about 1 280 Wh of energy and is separately rated for 100 A of standard discharge current (150 A peak). A "1C" rate for a 100 Ah pack is 100 A, which would empty it in about an hour; a gentle 0.1C (10 A) would take about ten hours. A battery can hold plenty of energy yet be unable to supply a high-power load, or supply large currents briefly yet hold little energy. Size both, separately — the reasoning is developed in watt-hours vs. amp-hours.

The specifications, and how they interact

A data sheet is a set of interacting limits, not a list of independent features. The map below shows what each spec answers and what it pulls on:

Specification Answers Interacts with
Energy capacity (Wh/Ah) How much energy is stored Voltage, usable fraction, age
Power capability (A / C-rate) How fast energy can flow Deliverable capacity, heat
Nominal & window voltage System voltage, charge/cut-off Chemistry, series count
Usable depth (DoD) How much you plan to use Cycle life, reserve
Temperature limits Where it can charge/discharge Available capacity, safety
Cycle & calendar life How long it lasts DoD, temperature, state of charge
BMS limits The safe operating window Everything above

Pull on one limit and others move. Discharge more deeply and you get more energy now but fewer cycles later. Pull higher current and the capacity you actually get out drops while heat rises. Store the battery hot and full and it ages faster even sitting idle. The rest of this article walks the threads that matter most.

Usable capacity: the nameplate is not the plan

You rarely design around the full nameplate. Most batteries keep a reserve, and how much depends on chemistry. Flooded lead-acid should be discharged to about 50% or less — its maker states plainly that "50% (or less) discharges are recommended" and that fully discharging "will damage (or kill) the battery." LiFePO₄ can be cycled far deeper, but still lasts longer if you leave some headroom: the example cell is rated up to 3 000 cycles at 100% depth of discharge but up to 5 500 at 60%.

So "usable capacity" is a deliberate choice, not a fixed number, and it shrinks again for temperature, discharge rate, and age. Turning a nameplate into a realistic planning figure is worth doing carefully — see usable vs. nominal battery capacity.

Temperature: performance and a hard limit

Temperature affects a battery two ways, and one of them is a safety limit.

  • Performance. Cold slows the chemistry, so a battery delivers less of its capacity on a cold morning than on a mild afternoon. Use the manufacturer's temperature-derating figures rather than guessing.
  • A hard charging limit. Charging most lithium-ion and LiFePO₄ batteries below about 0 °C (32 °F) drives metallic lithium plating on the anode. Peer-reviewed work describes this as causing "a drastic capacity loss, in addition to safety hazards", and the damage is permanent. That is why data sheets set a minimum charge temperature (0 °C for the example battery) that is stricter than the discharge range. Respect it, and rely on the BMS's low-temperature cutoff; never force charge into a cold battery.

Ageing: cycle life and calendar life

Batteries wear out two ways at once, and both are worth planning for:

  • Cycle ageing comes from use. Each charge/discharge cycle fades capacity a little; deeper cycles, higher currents, and temperature extremes fade it faster. Cycle-life ratings (like "3 000 cycles") run until capacity reaches an end-of-life threshold the maker defines.
  • Calendar ageing comes from time, even in storage. Research on long-term storage finds capacity loss is much greater at high temperature than moderate, and greater at a high state of charge than a low one — driven by growth of the solid-electrolyte interphase (SEI), a film that irreversibly consumes lithium. The practical lesson: for long storage, keep a lithium battery cool and only partially charged, following the manufacturer's storage guidance.

Design around the capacity you will have late in the battery's life, not its day-one figure.

Chemistry sets the trade-offs

There is no universally best chemistry — only fits. The table gives typical characteristics to reason with; always confirm real numbers against the specific product's data sheet.

Trait Flooded lead-acid LiFePO₄ (lithium iron phosphate) Other lithium-ion (e.g. NMC)
Nominal cell voltage ~2 V ~3.2 V ~3.6–3.7 V
Usable depth ~50% Deep (most of the pack) Deep
Cycle life Lower High (thousands) Moderate–high
Energy per mass Low Higher (~116 Wh/kg in the example) Highest
Cold-charge limit More tolerant Restricted below ~0 °C Restricted below ~0 °C
Notable hazards Vents hydrogen when charging Robust; still needs protection Higher energy density raises thermal-runaway stakes
Typical strengths Low cost, rugged, forgiving Long life, weight, deep use Maximum energy in least space

Match the chemistry to what actually constrains you: budget and robustness, weight and space, how deeply and often you cycle, and — importantly — the temperatures the battery will live in. A cheap, heavy, forgiving chemistry can be the right answer for a fixed installation; a light, long-lived one earns its price in a vehicle.

Configuration: voltage and capacity

Batteries are combined two ways, and they do different jobs:

  • Series connections add voltage at the same capacity. Four 3.2 V LiFePO₄ cells in series make a 12.8 V battery; higher system voltages (24 V, 48 V) carry the same power at lower current, which reduces wiring losses.
  • Parallel connections add capacity at the same voltage. Two 100 Ah batteries in parallel give 200 Ah at the original voltage.

Real banks add requirements beyond the arithmetic: units should be well matched and healthy, and the combination must be protected and managed appropriately. Cabling, fusing, and balancing for multi-battery banks carry real hazards and are governed by the manufacturer's instructions and local code — treat the diagrams here as explanatory, not as an approved installation design, and see the safety fundamentals.

The BMS and protection

Most lithium batteries include a battery management system (BMS) that "monitors the voltage and temperature of the cells during both charging and discharging, and if dangerous values are exceeded it cuts off the power supply." It defines the safe operating window and is a hard limit, not a suggestion — do not bypass it to reach reserve capacity or to charge outside its temperature range.

Beyond the BMS, every battery needs system-level protection, because any battery can push very large current into a short circuit. Overcurrent protection, disconnects, correct cabling, and — for some chemistries — ventilation are part of the design, not optional extras. High-energy lithium cells that are damaged, overcharged, or overheated can enter thermal runaway; flooded lead-acid gives off hydrogen while charging and needs ventilation. These are reasons the safety cluster exists; this pillar stays at the level of why protection matters, not how to install it.

State of charge vs state of health

Two "how much" questions are easy to confuse:

  • State of charge (SoC) — how full the battery is right now, like a fuel gauge.
  • State of health (SoH) — how much of its original capacity it still has after ageing.

A monitor's SoC percentage is an estimate, not a direct measurement of health, and a fully charged but aged battery holds less energy than the same battery when new. What a monitor actually measures, and how to read it, is covered in what a battery monitor measures.

Questions to answer before you choose

Work these out before comparing products:

  • How much energy must the battery deliver between charges, from a real daily load profile?
  • How much power (peak current) must it supply at once?
  • How deeply do you intend to cycle it, and how long must it last?
  • What temperatures will it live and charge in?
  • What system voltage (12 / 24 / 48 V) suits the loads and wiring?
  • What space, weight, and budget constrain the choice?
  • What protection, ventilation, and installation rules apply locally?

Answer these and the specification and chemistry choices largely make themselves — and you will be comparing batteries on the things that matter, not on nameplate numbers alone.

Key takeaways

  • A battery stores chemical energy and returns DC; its behaviour depends on temperature, use, and age, not just its nameplate.
  • Energy (how much) and power (how fast) are separate ratings — size both.
  • The specifications interact: depth of discharge trades against cycle life, high current and heat reduce and stress capacity, and heat plus high charge accelerate calendar ageing.
  • No chemistry is universally best; match it to loads, temperature, weight, lifespan, and budget, and keep the BMS and system protection in place.

Where to go next