Build a monthly solar-production table without hiding the assumptions

A complete hypothetical twelve-month example shows day weighting, seasonal variation and leap years.

By Kyle Bell · Original examples and linked primary references · November 17, 2026

An annual solar total can hide when electricity is produced. A monthly table exposes the assumptions behind that total and lets a household compare the shape of production with its own electricity use. The table is most useful when every resource value has a known source or is explicitly labeled as an illustration.

The following is a fictional teaching profile, not a forecast for a city or ZIP code. It assumes a 6 kW DC array and a 20% total derate for a non-leap year. The calculation for every row is array kW × daily peak sun hours × days in the month × 0.80.

A twelve-month worked example

Estimated monthly energy for the 6 kW example
MonthAssumed sun hours/dayDaysEstimated energy
January2.531372.0 kWh
February3.028403.2 kWh
March4.031595.2 kWh
April5.030720.0 kWh
May5.531818.4 kWh
June6.030864.0 kWh
July6.031892.8 kWh
August5.531818.4 kWh
September5.030720.0 kWh
October4.031595.2 kWh
November3.030432.0 kWh
December2.531372.0 kWh

The twelve outputs add to 7,603.2 kWh. June and July use the same resource input, but July produces 28.8 kWh more because it has one additional day in this arithmetic model. A chart should show that difference rather than forcing equal months whenever the daily resource is equal.

Weight the annual average by days

The profile contains 1,584 equivalent sun hours across the year after multiplying each monthly daily average by its day count. Dividing 1,584 by 365 gives about 4.3397 hours/day. Multiplying that weighted annual average by 365 days reproduces the same annual result.

A simple average of the twelve monthly inputs gives about 4.3333 hours/day. Using that unweighted number for the full year gives 7,592 kWh, which is 11.2 kWh lower. The difference is small in this invented example, but day weighting is easy to implement correctly and avoids an unnecessary inconsistency.

Treat leap years and missing values explicitly

For a leap year with this same monthly profile, February has 29 days. The extra day adds 6 × 3 × 0.80 = 14.4 kWh, bringing the total to 7,617.6 kWh. Do not multiply every month by a single 366/365 adjustment when the extra day belongs to February.

A blank monthly input is unknown, not zero sunlight. The calculator should ask for the missing value or offer a clearly labeled flat scenario. A real zero is different: it can be an intentional scenario input and should produce zero for that month without breaking the annual table.

Compare equivalent outputs

When comparing against another tool, determine whether its numbers represent DC array energy or AC inverter energy. SAM's results documentation lists those as separate outputs. This calculator's combined-loss estimate is intended as an approximate delivered-energy scenario, so it should not be compared with an unadjusted DC figure as though they were identical.

Export the twelve inputs, calendar year, day counts, loss percentage and calculated outputs together. An energy-only CSV is difficult to reproduce. A useful saved result should allow someone to recreate every row with a pocket calculator and understand why the annual figure differs from a more detailed weather simulation.

Keep learning

Solar losses and shading: combine assumptions without counting them twice