Solar Battery Bank Calculator

Size an off-grid or backup battery bank from daily energy use and days of autonomy, with depth of discharge and round-trip losses accounted for — in watt-hours, amp-hours and batteries.

Last checked against the code

Solar Battery Bank Calculator

Units
Battery count(optional)

Fill these in to turn the capacity above into a number of batteries. Leave them blank and the answer stays in watt-hours and amp-hours.

Note: This sizes energy, not power. A bank of the right watt-hours can still be the wrong bank: it also has to deliver your peak load without exceeding the inverter rating or the BMS discharge limit, and something has to be able to recharge it. Array sizing, charge controllers and inverters are separate questions this page does not answer.

How this is calculated

Note: Energy over time, not power. This is the storage member of the site's three "size a power source" calculators, and it measures a different quantity from the other two.Transformer sizing is kVA off an existing utility supply.Generator sizing is watts — instantaneous power, decided by the worst transient. This page is watt-hours: how much energy is stored, and how long it has to last. A bank that comfortably delivers your peak power can still be flat by 3 a.m., which is a failure mode neither of the others has.

The calculation is one multiplication and three divisions, and the divisions are where the interesting part lives — each one makes the bank bigger, and two of them are losses people leave out entirely.

Start with energy, not power. Daily consumption times days of autonomy gives the watt-hours that must come out of the bank between charges. Note the units: a 100 W fridge whose compressor runs a third of the time uses 800 Wh a day, not 2,400. Adding up nameplate wattages is the fastest way to oversize a system by a factor of three.

Divide by depth of discharge, and watch what happens. This is the input that dominates the answer. A battery you are only willing to take to 50% has to be twice the size of your usable requirement; at 80% it is a quarter larger. Depth of discharge is not a hard limit but a life decision — every cycle taken deeper wears the cells faster, and the curve is steep. The gap between the two chemistries is almost entirely this number.

Divide by round-trip efficiency. Energy enters through a charge controller, sits in cells that lose a little to their own chemistry, and leaves through an inverter. Ten to twenty percent of what went in does not come back out, and the bank has to be larger by that fraction to deliver what was asked of it.

Finally divide by the bus voltage to get amp-hours, which is how batteries are actually sold. This step is a unit conversion rather than a loss: the stored energy is the same at 12 V or 48 V, but the amp-hour figure is four times larger at 12 V — and so is every current in the system.

Worked example

One full calculation with real numbers, so you can follow along and check the tool by hand.

An off-grid cabin measured at 4,000 Wh a day, designed to runtwo days through cloud with no charging. LiFePO4 cells at their usual 80% depth of discharge, a system round-trip efficiency of90%, on a 48 V bus.

Usable energy. 4,000 × 2 = 8,000 Wh. That is what has to come out of the bank between charges, and it is the only figure in this calculation that describes the load rather than the hardware.

Depth of discharge. 8,000 ÷ 0.80 = 10,000 Wh. The extra 2,000 Wh is capacity you pay for, install, and deliberately never touch.

Round-trip efficiency. 10,000 ÷ 0.90 ≈ 11,111 Wh. Another 1,111 Wh, this time energy that goes in and never comes back out.

Amp-hours. 11,111 ÷ 48 = 231 Ah at 48 V.

So an 8,000 Wh requirement buys an 11,111 Wh bank — 39% more nameplate capacity than the load actually consumes, before anyone has rounded up to a product. On lead-acid at 50% the same load needs 370 Ah, roughly 60% more again, and several times the weight.

Now turn it into batteries, carefully. With 12 V 100 Ah units: 231 Ah needs three in parallel, but each parallel string must first reach 48 V, which takes four in series. That is 3 × 4 = 12 batteries, not three. Dividing 231 by 100 and buying three is the single most common way this calculation goes wrong, and it leaves the system at a quarter of the intended capacity.

Visual comparison

The same energy, two chemistriesUpdates with your load, autonomy, efficiency and bus voltage — only the depth of discharge differs between the bars, and it produces a 60% difference in nameplate capacity. The highlighted bar is your current selection. Real lead-acid efficiency is also lower than lithium's, so an honest comparison is wider still than this shows.
The same energy, two chemistries
ChemistryBarCapacity needed (Ah)Depth of discharge
Lead-acid37050% DoD
LiFePO423180% DoD — your selection

Source: Both at their conventional depth of discharge, everything else held equal

Reference tables

The first table is the two chemistries side by side. Read it before deciding anything, because the depth-of-discharge column is doing more to your bank size than any other number on this page, and the cycle-life column is what justifies it.

The second table is the worked example at each of the three bus voltages, and it exists to make one point: the watt-hour column does not move. Stored energy is a property of the bank, not of how you wire it. Only the amp-hour figure changes, and with it every current, conductor and fuse in the system — which is the real argument for 48 V over 12 V.

Lead-acid against LiFePO4
ChemistryUsual DoDRound tripCycle lifeIn practice
Lead-acid (flooded, AGM or gel)50%85%500–1,000 cycles at 50%Cheap per nameplate watt-hour and expensive per usable one. Half the capacity is unavailable by design, the bank is heavy, and discharging deeper to reclaim it trades cycle life away quickly.
LiFePO4 (lithium iron phosphate)80%95%3,000–5,000 cycles at 80%More expensive per nameplate watt-hour and usually cheaper per usable one over its life. Takes a deeper discharge, weighs far less for the same usable energy, and holds voltage better under load.
Source: Typical figures for a first-pass sizing. Manufacturer-dependent — the datasheet for the battery you are buying beats this table.
The worked example at each bus voltage
Bus voltageNameplate energyWhCapacityAhAs 100 Ah units
12 V11,11192610 × 100 Ah strings
24 V11,1114635 × 100 Ah strings
48 V11,1112313 × 100 Ah strings
Source: 4,000 Wh/day, 2 days autonomy, 80% DoD, 90% efficiency. Computed, not transcribed.

Notes and exceptions

Capacity is not the only constraint. A bank of the right watt-hours still has to deliver your peak load without exceeding the inverter's rating or the BMS discharge limit, and lithium banks in particular are often limited by continuous discharge current rather than by energy. Check the peak draw against both before settling on a design.

Something has to refill it. Array sizing is a separate calculation driven by insolation at your latitude and season, panel orientation and shading, and it frequently ends up being the binding constraint rather than the bank. A bank sized for three days of autonomy that takes five days of good sun to recharge is not a three-day system.

Temperature moves both figures. Lead-acid loses meaningful capacity below about 20 °C — a bank sized at room temperature can be well short on a winter night, which is exactly when the autonomy is needed. Lithium is less affected on discharge but many BMS units refuse to charge below freezing at all, which turns a cold snap into a system that discharges and never recovers.

Parallel strings are not free. Cells in parallel strings share current according to their internal resistance, and mismatched or ageing strings share it unevenly. Beyond three or four parallel strings the usual advice is to move to higher-capacity cells at a higher bus voltage instead — which is another argument for 48 V, since it needs a quarter of the parallel strings that 12 V does for the same energy.

Common mistakes

  1. Sizing a lead-acid bank at 100% depth of discharge

    The nameplate is not the usable figure. Cycle life falls off steeply past about 50% on lead-acid — a bank taken to 80% daily can be finished in two years where a 50% one lasts five, and most warranties are written against a stated depth. Half of what you bought is deliberately unavailable.

  2. Forgetting round-trip efficiency entirely

    Energy goes in through a charge controller, sits in cells that lose a little, and comes back out through an inverter. Ten to twenty percent never returns. A bank sized on usable watt-hours alone is short by exactly that, and it shows up as running out earlier than the arithmetic promised.

  3. Confusing series voltage with parallel capacity

    Series adds volts and leaves amp-hours alone; parallel adds amp-hours and leaves volts alone. Four 12 V 100 Ah batteries in series make a 48 V 100 Ah string, not 400 Ah. Dividing your required amp-hours by one battery’s rating gives strings, not batteries — and on a 48 V bus that is out by a factor of four.

  4. Sizing from connected wattage instead of daily energy

    A 100 W fridge does not use 2,400 Wh a day, because the compressor runs perhaps a third of the time. Adding up nameplate wattages and multiplying by 24 produces a bank several times larger than anything needed. Measure or estimate duty cycle, or read a meter.

  5. Comparing amp-hours across different bus voltages

    A 12 V 100 Ah battery holds 1,200 Wh; a 48 V 100 Ah battery holds 4,800 Wh. Amp-hours only compare within one voltage, which is why quoted prices per amp-hour are close to meaningless across systems. Watt-hours is the figure that travels.

This tool provides planning estimates. Always verify final values against your local code and a licensed electrician.

Frequently asked questions

What size battery bank do I need for 4 kWh a day?

With two days of autonomy on LiFePO4 at 80% depth of discharge and 90% round-trip efficiency, about 11,100 Wh of nameplate capacity — roughly 231 Ah on a 48 V bus. The same requirement on lead-acid at 50% depth of discharge needs about 370 Ah, because a larger share of the bank is capacity you own but are not allowed to use.

How is this different from sizing a generator?

A generator is sized in watts, for instantaneous power, and its worst case is the half-second a motor starts. A battery bank is sized in watt-hours, for energy over time, and its worst case is the last hour before dawn. The two are independent: a bank that comfortably starts your well pump can still be flat overnight, and a generator that runs all night may still stall on that pump.

Why can’t I use the full capacity of a battery?

You can, once or twice. Depth of discharge is a life decision rather than a hard limit — every cycle taken deeper wears the cells faster, and the relationship is steep rather than proportional. Lead-acid is conventionally held to 50% and LiFePO4 to 80% because those points sit where cycle life is still long. A deeper discharge saves money once and costs it repeatedly.

Does wiring batteries in series increase capacity?

No. Series wiring adds voltage and leaves amp-hours unchanged: four 12 V 100 Ah batteries in series give 48 V at 100 Ah. Parallel wiring does the opposite, adding amp-hours at constant voltage. Both increase watt-hours, because watt-hours are volts times amp-hours — which is why watt-hours is the honest way to compare two banks and amp-hours is not.

Should I build a 12 V, 24 V or 48 V system?

48 V for anything house-sized. The stored energy is identical at any bus voltage, but the current is not: the same 5 kW draws over 400 A at 12 V and about 104 A at 48 V. That difference decides conductor size, fuse ratings, charge controller cost and how much energy is lost as heat in the wiring. 12 V survives mainly where the appliances are natively 12 V, as in vehicles and boats.

How many days of autonomy should I design for?

One to three for most systems, and the number is a judgement about weather and consequences rather than a calculation. Each extra day costs the same as the last, while the chance of needing it falls away sharply — which is why past about three days it is usually cheaper to add a generator for the rare long stretch than to buy storage that sits full for most of the year.