Is a peak sun hour the same as an hour of daylight?

No. Daylight counts photons you can read by; peak sun hours count energy at panel-rating strength. A 16-hour June day in Seattle compresses to about 5 peak sun hours; an 8.5-hour December day there is barely 1.

How many peak sun hours do solar panels need?

There is no minimum; there is only array size. At 3.5 hours you need about 30% more panels than at 4.5 for the same energy, and Germany runs a huge solar fleet at under 3. The question is whether the extra panels pencil at your electric rates, not whether the sun qualifies.

Peak sun hours, explained with actual numbers

A peak sun hour is a unit of energy wearing a time costume: one hour of sunshine at full standard strength, 1,000 watts per square meter. Phoenix banks about 6.6 of them on an average day, Seattle about 3.6, and that one number, more than any hardware choice, is what sizes a solar array.

The measurement, without hand-waving

Stand a perfect square meter facing the sun. At solar noon on a clear day, roughly 1,000 watts of light lands on it; that intensity is the "peak" in the name, and it is the same condition panels are lab-rated at. But mornings are weak, afternoons decay, clouds happen. So meteorologists add up the whole day's actual energy on that square meter and restate it as hours-at-full-strength. A day that delivers 4,500 watt-hours per square meter is a 4.5 peak-sun-hour day, whether it came as one glorious desert afternoon or twelve hours of Midwestern haze.

This is why the figure is so useful: it's the fuel gauge, not a weather description. Panel watts × peak sun hours = daily watt-hours, before real-world losses. A 400W panel under Denver's 5.5 hours makes 400 × 5.5 = 2,200 Wh DC, about 1.7 kWh AC after the losses our output guide itemizes.

What moves the number

Latitude sets the ceiling and clouds spend it. Yuma and Miami sit at similar latitudes; Yuma's dry air gets it 6-plus hours while Miami's afternoon thunderheads hold it near 5.2. Elevation helps (thinner air, less scattering; Denver outperforms its latitude), and so does winter snow reflection at the margins. The practical spread across the lower 48 runs from about 3.6 (Seattle) to 6.6 (Phoenix), which in hardware terms means a Seattle array must be about 80% bigger than a Phoenix array for the same annual energy. The full ranked table is on our sun-hours page.

Averages hide the seasons

Every figure above is an annual mean of a curve that swings hard. A 4.0-hour state might see 5.5 in June and 1.8 in December. Grid-tied systems shrug at this: net metering banks the summer. Off-grid systems live and die by the December number, which is why our off-grid guide designs for the worst month and treats the annual average as marketing. When you want the local, monthly truth for a real address, NREL's PVWatts serves it free from measured weather data, and any installer quoting you production numbers should be able to show you the equivalent.