How we size your system

We trust sources in this order, and the report says which one we used:

1. The rating label on the appliance, read from your photo. Volts × amps if watts aren't printed.
2. A named model from our catalogue. Every entry is taken from a manufacturer spec page.
3. Physical size — litres, HP, BTU, screen inches, bulb count — converted through published bands.
4. A typical value for that appliance type. Lowest confidence, and we flag it.

We then apply an efficiency multiplier: old units ×1.2, standard ×1.0, and inverter or energy-saver units per that model's own rating.

You pick a plain choice — "Evenings only", "Runs all the time" — and each choice maps to a fixed number of day hours, night hours and days per week. The engine only ever reads those three numbers. The wording is there to make the question answerable without a meter.

weekFactor = daysPerWeek / 7
dayWh      = watts × dayHours   × weekFactor
nightWh    = watts × nightHours × weekFactor
dailyKwh   = sum(dayWh + nightWh) / 1000

This is week-averaged, so a washing machine used twice a week contributes two sevenths of its run-day energy.

continuousW = sum(watts × quantity)
inverterKw  = max(1, continuousW × 1.25 / 1000)

applianceSurge = watts × quantity × surgeMultiplier
surgeW         = largest applianceSurge + second largest

Surge multipliers are per appliance type — fridge ×4, air conditioner ×5, water pump ×4, resistive loads like an iron ×1. We assume the two heaviest motors can start at the same moment and no others do. Because PowerPods are rated at PF 1, the recommended kVA equals the recommended kW, snapped up to the next standard step.

Sized from the night load on days those appliances actually run, not the week average — the battery has to survive a typical night, not an average one.

nightKwh   = sum(watts × nightHours) / 1000
gridCredit = gridHours >= 8 ? nightKwh × 0.5 : 0
pvCredit   = min(nightKwh, pvKwp × 4.2 × 0.5)
batteryKwh = max(2, ceil((nightKwh − gridCredit − pvCredit) × 1.2))

Three deliberate conservatisms:

• Only one recharge source is credited per day.
• A generator is never credited, even if you own one.
• Under 8 grid hours a day, grid is treated as unplannable and credited zero.

usableKwh     = batteryKwh × 0.8
autonomyHours = usableKwh / average night kW
batteryAh     = batteryKwh × 1000 / 48

Where the grid-hours figure comes from. If you type your own number, we use it. If you don't, we use a state average taken from the DisCo band feeder mix in that franchise area: 12 h/day for states dominated by commercial metro feeders (Lagos, Rivers, FCT), 9 h/day for mixed urban and semi-urban areas, and 6 h/day where Band D and E feeders dominate. With no state selected we use the national default of 9 h/day. This is a planning assumption, not a measurement — supply varies feeder by feeder, and the report always states which figure was used and whether it was assumed or entered.

pvKwp      = max(0.5, dailyKwh / 4.2)
panelCount = ceil(pvKwp × 1000 / panelWatts)

Panel count is indicative. The real count depends on roof area, orientation and shading, which only a site visit settles.

We walk the range from smallest to largest and take the first model that satisfies all four limits at once: continuous kW, surge kW, battery kWh and solar input kWp. Whichever limit ruled out the next size down is printed on your report as the deciding factor, so you can see the single number that set your size.

Each appliance carries a confidence level from how its watts were obtained. The report's overall confidence is load-weighted: an uncertain 2 kW air conditioner moves it far more than an uncertain 7 W bulb.

• Peak sun hours are one national figure, not per-state irradiance.
• Diversity is not modelled beyond the two-motor surge rule, so continuous load runs conservative — real homes rarely run everything at once.
• Grid supply is your stated average — or, if you didn't state one, a state-level assumption. It is never a measured feeder profile.
• Cable runs, earthing, protection devices, roof area and shading belong to the physical site assessment.
• The 24-hour load shape is derived from your day/night split and appliance behaviour, not from metered interval data.