Sizing a Residential Battery Bank: The Math That Actually Matters

Sizing a battery bank is one of the few places in a residential energy project where a spreadsheet beats a gut feeling. Oversize it and you pay for capacity you never cycle; undersize it and the inverter shuts down on the first cloudy week. The method below is the one I use before touching any datasheet: four numbers, in order, no magic.

1. Start from daily energy, not peak power

List every load with its power draw and daily run time, then multiply: watts × hours = watt-hours. A small off-grid setup might look like this: LED lighting 40 W × 5 h = 200 Wh, a 24 V water pump 60 W × 1 h = 60 Wh, phone and laptop charging ≈ 150 Wh, a ventilation fan 90 W × 4 h = 360 Wh. Total: roughly 770 Wh per day. Be honest about duty cycles — a fridge compressor does not run 24 hours, and a pump rated 60 W rarely draws that continuously.

Peak power matters too, but it sizes the inverter, not the battery. Energy (Wh) sizes the battery; power (W) sizes the inverter. Mixing the two up is the most common beginner mistake.

2. Nominal vs. usable capacity

A battery's nameplate capacity is not what you get to use. Usable energy is nominal capacity × depth of discharge (DoD). LiFePO4 chemistry comfortably allows 90–95% DoD with thousands of cycles; classic lead-acid should stay around 50% unless you enjoy replacing batteries. If your loads need 770 Wh/day and you want two days of autonomy on LiFePO4 at 90% DoD, the math is 770 × 2 / 0.9 ≈ 1,711 Wh nominal — so a 24 V 100 Ah pack (2,560 Wh nominal) covers it with headroom, while a 12 V 100 Ah pack (1,280 Wh) does not.

Autonomy days are a design choice, not a law of physics. One day of autonomy with a generator backup is often cheaper than three days of batteries.

3. C-rate: can the bank deliver the watts?

Capacity in Wh tells you how long the energy lasts; the C-rate tells you how fast you may take it out. A 100 Ah LiFePO4 cell rated for 1C continuous discharge can deliver 100 A — at 24 V that is 2,400 W. But the real limit is usually the BMS: many budget 100 Ah packs ship with a 50 A or 100 A BMS, capping you at 1,200–2,400 W regardless of what the cells could do. Size for your peak load: peak watts / pack voltage ≤ BMS continuous current. If the numbers don't fit, add packs in parallel (capacity and current both scale) rather than pushing one pack past its rating.

4. Don't forget inverter losses — or the BMS

An inverter is typically 90–95% efficient, and it also draws idle power (often 10–30 W) around the clock. Add ~10% to your daily Wh for conversion losses, and count the inverter's standby draw as a 24-hour load in your audit. As for the BMS: it is not optional equipment. It protects against overcharge, deep discharge, overcurrent and temperature extremes, and it keeps series cells balanced. A lithium bank without a BMS is a chemistry experiment, not a power system.

A worked example

• Daily consumption (audited): 1,400 Wh
• Autonomy target: 2 days → 2,800 Wh usable needed
• LiFePO4 at 90% DoD → 2,800 / 0.9 ≈ 3,111 Wh nominal
• Inverter losses (+10%) → ≈ 3,420 Wh nominal minimum
• System voltage 24 V → 3,420 / 25.6 ≈ 134 Ah → a 24 V 150 Ah bank (or 2× 24 V 100 Ah in parallel)
• Peak load 800 W → 800 / 25.6 ≈ 31 A → any 100 A BMS handles it comfortably

daily_wh = 1400      # audited loads, Wh/day

autonomy_days = 2
dod = 0.90           # LiFePO4 usable depth of discharge
inverter_eff = 0.90
pack_voltage = 25.6  # 8S LiFePO4 nominal

nominal_wh = daily_wh * autonomy_days / (dod * inverter_eff)
required_ah = nominal_wh / pack_voltage
print(f"Nominal: {nominal_wh:.0f} Wh  ->  {required_ah:.0f} Ah at {pack_voltage} V")
# Nominal: 3444 Wh  ->  135 Ah at 25.6 V

These are rules of thumb for planning, not a substitute for product datasheets or local regulations. Anything grid-tied — and anything above extra-low voltage in some jurisdictions — deserves a qualified electrician's sign-off. But if you walk into that conversation with the four numbers above already computed, the conversation goes much faster.