How Much Runtime Is Your Solar Positioning Giving Away?

The 100% baseline

The most energy the array could collect at your location — ideally aimed (true south), at the best tilt for each month's sun, with no shade. Months are weighted by their real sunshine using the PVGIS Typical-Meteorological-Year grid, so brighter months count for more.

Your setup

A clear-sky plane-of-array model re-scores the array at the direction and tilt you picked, then applies a peak-hour shading factor. Because cells are wired in series, partial shade costs more than its area: light ≈ −8%, partial ≈ −28%, heavy ≈ −50%. The result is the percentage of the ideal energy you give up.

Why winter is worse

A low winter sun sits closer to due south for fewer hours, so the same aiming and tilt errors shave off a bigger share of the day's energy — which is why a summer tilt left on, or a shadow across the array, hurts most in the months you can least afford it.

Expected days in the field

The chart runs a day-by-day simulation for a reference platform — an 800 W array, a 460 Ah AGM bank (≈ 5.5 kWh usable at 50% depth-of-discharge), and a 100 W continuous load. Starting with a full battery in each month, it steps forward through the real upcoming months (daily charge minus load, battery capped at full) and counts the days until the pack is depleted — so a deployment launched in a strong month is still limited by the leaner months it runs into. A bar at 90+ rides into the next year. Green is optimal positioning, orange is your setup; the gap is what positioning costs you.

A planning estimate — real output also depends on weather, snow, soiling and mounting.

Where it comes from

The amount of sun at each location is from PVGIS — the European Commission Joint Research Centre's Typical Meteorological Year (TMY) dataset — on a 2° grid across the US (plus Alaska & Hawaii). Each point gives the average solar energy by month, in kWh per day per kW of panel, with standard system losses already applied. It's the same dataset behind our Solar Runtime Map.

Canada is covered too. Direct PVGIS data (from the North American NSRDB record) for the major Canadian regions and cities; the few most remote spots interpolate from the nearest points.

Does it include clouds and rain?

Yes. TMY is built from many years of real satellite and ground weather observations, so average cloud cover, haze and rain are already baked into the monthly numbers. That's why a cloudy region reads lower than a sunny one at the same latitude, and why every location dips in winter beyond just the shorter days — you're not seeing a clear-sky best case.

What it doesn't capture

TMY is a typical year, and the day counts use each month's average generation. So it reflects the normal climate of a place — not a specific forecast or a worst-case storm. A prolonged cloudy stretch will drain the battery faster than these averages show, so treat the days in the field as typical conditions, not a guarantee.

The positioning penalty

How much direction, tilt and shading cost you is computed from sun-angle geometry and applied on top of the weather-adjusted PVGIS baseline. Clouds affect a good and a bad setup equally, so they cancel out of the loss percentage and stay in the absolute kWh and days.

Runtime follows the margin, not the production

Each day the battery gains the difference between what the panels produce and what the load consumes (2.4 kWh/day for this platform). As long as production covers the load, the battery never drains — the runtime is unlimited, no matter how much extra is produced.

Below the line, the battery is a countdown

The moment production falls under the load, the battery bridges the gap, and runtime becomes usable battery ÷ daily shortfall. A shortfall of just half a kWh per day empties this platform's 5.5 kWh bank in about 11 days.

That's the cliff

So a percentage decrease in power is not an equal decrease in runtime. Losing 30% of production in a month with plenty of margin costs nothing; losing the same 30% in a month where the margin is thin drops runtime from unlimited to days. What matters isn't how much power positioning loses — it's whether the loss pushes production across the load line.