Solar Panel and Array Size Calculator

Turn your monthly kWh into an array size, panel count and roof area, using real peak sun hours and a loss stack that compounds properly.

Panels
Roof area
Annual output
Losses

Where the losses go

These compound rather than add — each takes its cut of what survived the last.

Peak sun hours are not daylight hours

This is the misunderstanding the whole subject rests on. A peak sun hour is one hour at 1000 W/m² — the intensity a panel is rated at — and a day's total sunlight expressed in those units. Seattle has sixteen hours of daylight in June and averages about 3.4 peak sun hours across the year. Anyone reasoning from daylight hours will size an array at roughly a third of what they need.

  • The annual average hides the seasonal swing, and the swing is enormous. Boston gets 5.9 peak sun hours in June and 2.5 in December. Grid-tied that is fine, because summer export pays for winter import where net metering exists. Off-grid it is not — you size for the worst month or you go dark in January, which is why off-grid arrays look absurdly oversized in summer.
  • Shading is the thing no calculator can model and the thing that most often ruins an array. A single shaded panel drags a whole string down unless the system has per-panel optimisers or microinverters, so one chimney matters far more than a few degrees of orientation.
  • Roof area is not panel area. Real installations need clearance around the array for fire access and mounting, so plan on noticeably more roof than the modules alone occupy.
  • Panels degrade about half a percent a year, leaving roughly 88% after 25 years. Warranties typically guarantee near 85% at that point, so this is the ordinary expectation rather than a fault.
  • These sun-hour figures are broad regional averages. An actual site wants NREL's PVWatts or the equivalent national dataset — microclimate and local shading move the number more than latitude does.

How to use

  1. Enter your monthly electricity use — it is on the bill.
  2. Pick your location, or type peak sun hours from PVWatts if you have them.
  3. Choose grid-tied to size on the annual average, or off-grid for the worst month.
  4. Read the array size, panel count and roof area, then check the loss breakdown.

Frequently asked questions

What are peak sun hours?

One peak sun hour is one hour at 1000 watts per square metre — the intensity a panel is rated at — and a location's figure is its whole day of sunlight expressed in those units. It is not daylight hours, and the difference is enormous. Seattle has sixteen hours of daylight in June and averages about 3.4 peak sun hours across the year, so anyone reasoning from daylight would size an array at roughly a third of what they need.

Why do the losses matter so much?

Because they take about 18 percent off before anything reaches your meter, and they compound rather than add. Inverter efficiency, panel temperature, soiling, wiring, mismatch between modules and system downtime each take a cut of what survived the previous stage. Temperature is usually the largest single item — panels lose roughly 0.35 percent per degree above 25 Celsius, and a dark roof in summer runs far hotter than the air does.

Should I size on the annual average or the worst month?

Grid-tied, the annual average is right, because the grid carries the difference and summer export pays for winter import where net metering exists. Off-grid it is the worst month or nothing — there is no grid to borrow from in January. That is why off-grid arrays are two or three times the size of a grid-tied one for the same usage, and why they look absurdly oversized in summer.

How much roof will I need?

Roughly two square metres per 400 watt panel, so a 7 kilowatt array is about 36 square metres of modules. Plan on noticeably more roof than that: real installations need clearance around the array for fire access and for the mounting hardware, and only genuinely unshaded south-facing area counts at all. Larger panels help — a 500 watt module in the same footprint as a 350 watt one is 40 percent more output from the same roof.

How much does shading cost me?

Far more than most people expect, and it is the one thing no calculator can model. A single shaded panel drags an entire string down to its output unless the system has per-panel optimisers or microinverters, so one chimney or one growing tree matters more than several degrees of orientation. If anything shades the roof at any point in the day, that is worth a proper site survey before anything else.

Do panels wear out?

They degrade slowly rather than failing. About half a percent a year is typical, leaving roughly 88 percent of the original output after 25 years, and most manufacturers warrant somewhere near 85 percent at that point. Inverters are the part that actually needs replacing — expect one replacement over the life of the array, which is a real cost worth including in any payback estimate.

How accurate are these sun-hour figures?

They are broad regional averages good enough to plan around and not good enough to sign a contract on. An actual site wants NREL's PVWatts in the United States or the equivalent national dataset elsewhere, because microclimate, local shading and roof orientation move the number more than latitude does. Treat the answer here as the size to expect a quote near, rather than the size to order.

🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.