Sizing a Battery for Outage Hours, Not Sunshine Hours

Most sizing guides open with your location's peak sun hours. In a region where the grid fails on schedule, that number answers the wrong question. Sun hours tell you how fast a bank refills. Outage hours tell you how much you must store, and that is what sets capacity.

Start from the outage, not the resource

Real deficits vary by an order of magnitude between regions. Sub-Saharan Africa has over 600 million people without stable supply; South Africa loses more than 200 days a year and Nigeria more than 240. In South Asia, India's summer shortfall runs past 15 percent, while Pakistani cities lose 6 to 12 hours a day and rural areas approach 20. Vietnam's industrial parks run scheduled weekly cuts. Iraq and Kuwait see summer gaps above 31 percent, with as little as 12 hours of supply.

A system built for four hours of daily outage and one built for twenty are not variations of the same design. They differ by multiples.

The calculation, worked

Take the loads you must keep running during the cut. A fridge, lights, a router and a borehole pump might draw 0.5kW averaged across the outage window. Multiply by outage hours to get the energy you need, then divide by depth of discharge.

LiFePO4 safely uses about 80 percent of nameplate. Lead-acid only about 50 percent.

• 4 hours a day: 0.5kW × 4 = 2kWh, ÷ 0.8 = 2.5kWh LiFePO4 nameplate (5kWh of lead-acid for the same usable energy)
• 6 hours a day: 3kWh ÷ 0.8 = 3.75kWh
• 12 hours a day: 6kWh ÷ 0.8 = 7.5kWh
• 20 hours a day: 10kWh ÷ 0.8 = 12.5kWh

That last case is where solar has to carry nearly everything. If the outage runs 20 hours, the array gets about 4 hours of daylight to both run the house and refill a 12.5kWh bank. Panel wattage, not just battery size, becomes the binding constraint.

Two numbers that quietly shrink your bank

Motor surge. A fridge or pump pulls 3 to 7 times its running watts for a second at startup. Size the inverter's surge rating to that spike. A unit rated for steady load trips on the first pump start, and a tripped inverter delivers nothing regardless of how large the battery is.

Temperature. LiFePO4 loses about 10 to 15 percent of capacity at 0°C and 20 to 30 percent at -20°C. Add roughly 20 percent to the capacity you calculated, and confirm the BMS blocks charging below freezing, since charging a cold cell plates lithium and damages it permanently.

Transfer time decides whether your food survives

Capacity keeps the lights on. Transfer speed keeps motors alive. A 20 millisecond switch drops a fridge compressor for a beat and it can stall. A 10 millisecond transfer keeps it running. When you compare hybrid inverters, read transfer time, not just whether it has a battery port.

Predictable cuts are easier than random ones

Scheduled curtailment, like the weekly cuts in Vietnamese industrial parks, is easier to size than random failure because you know the window. Random outage regions need more margin: size for the worst week you have actually experienced, then add 20 percent.

Run the outage hours through the formula above and you have a number that reflects how the grid behaves where you live, instead of a generic rule of thumb built for a grid that never fails.