Peak Sun Hours: What the Number Actually Measures
Peak sun hours are a unit of energy, not daylight length. How irradiance, insolation and a derate combine to predict what a PV array will produce.
A peak sun hour is a unit of energy, not a duration of daylight. One peak sun hour is the energy delivered by an irradiance of 1,000 watts per square metre sustained for one hour, which is 1,000 watt-hours per square metre. The sun does not have to be at that intensity for the full hour, and usually is not. A day that spends two hours at 500 W/m² and one hour at 1,000 W/m² delivers two peak sun hours, even though the sun was usefully above the horizon for three hours. The number describes the resource, not the clock.
Irradiance and insolation are not the same quantity
Irradiance is power per unit area at a given instant, measured in watts per square metre. Insolation is energy per unit area accumulated over a period, measured in watt-hours per square metre or, in the units most often published, kilowatt-hours per square metre per day. Peak sun hours are insolation with the units divided through by the reference irradiance: 1 kWh/m²/day equals 1 peak sun hour. The two are numerically interchangeable, which is why a dataset may label its column either way.
This is the same 1,000 W/m² that appears in standard test conditions. STC fixes irradiance at 1,000 W/m², cell temperature at 25 °C and a defined spectral distribution. A panel rated at 400 W produces 400 W under those conditions and something else under any other. Peak sun hours borrow the irradiance figure from STC because it is a convenient round number near the upper end of what a clear sky delivers to a tilted surface at mid-latitudes, not because arrays spend their day at that level.
Why a long day and a short day can score alike
On a clear summer day at a mid-latitude site the sun may be up for fifteen hours. The irradiance rises from near zero at sunrise, peaks around solar noon, and falls back. The area under that curve, divided by 1,000 W/m², might come to five or six peak sun hours. On a clear winter day the sun is up for nine hours, the peak is lower because the sun is lower in the sky, and the total might be two or three peak sun hours. The summer day wins, but not by the ratio of daylight hours, because the shape of the curve matters as much as its width.
A hazy summer day can be long and still score poorly. Diffuse light spreads the curve and lowers the peak. A cold, crisp winter day at high altitude can score better than a humid summer day at sea level, because water vapour and aerosol scatter and absorb more of the beam. Latitude sets the ceiling; weather and air mass decide how close a given day gets to it.
Using the number with a derate
The arithmetic a designer uses is short. Daily energy in kilowatt-hours equals array size in kilowatts multiplied by peak sun hours multiplied by a derate factor. The derate is where the losses live: soiling, mismatch, wiring resistance, inverter conversion, and the gap between the cell temperature in the field and the 25 °C of STC. A derate of 0.75 to 0.85 is common for a grid-tied system, and the exact figure depends on the equipment and the mounting.
The temperature term deserves its own mention because it is the one people forget. A silicon cell loses roughly 0.3 to 0.45 percent of its output per degree Celsius above 25 °C, depending on the cell technology. A roof-mounted array in still air on a hot afternoon can run 30 °C or more above ambient cell temperature. That loss is already inside the derate; it is not a fault.
Peak sun hours are a property of the site and the tilt, not of the equipment. Public datasets publish them as long-term monthly averages, usually derived from satellite irradiance or ground stations. The reader should look up the figure for the specific location and tilt rather than accept a number quoted for a nearby city, because the difference between two sites twenty kilometres apart can be larger than the difference between two months.
Where the figure comes from
Datasets such as the Global Solar Atlas, PVGIS and the U.S. National Solar Radiation Database publish monthly and annual insolation for a given latitude, longitude and tilt. The output is usually in kWh/m²/day, which is peak sun hours. The conventions differ slightly: some report global horizontal irradiance, some report plane-of-array irradiance for a stated tilt and azimuth. Mixing the two is a common source of confusion, because a tilted plane in winter collects more than a horizontal one and the numbers will not match.
A command-line query to PVGIS, for example, returns a table of monthly values:
curl "https://re.jrc.ec.europa.eu/api/v5_2/PVcalc?lat=40&lon=-3&peakpower=1&loss=14&outputformat=json"The E_m field in the response is monthly energy in kilowatt-hours for a 1 kWp system at the stated loss; dividing by the days in the month and by the array size gives daily peak sun hours after that loss. The H(i)_m field is monthly plane-of-array irradiation in kWh/m², which is the raw insolation before any system loss. Reading the wrong field is how a designer ends up with a number that is either too optimistic or double-derated.
What people get wrong
The most common misreading is to treat peak sun hours as the number of hours the panels are generating. A proposal that says “five peak sun hours” is not saying the array runs for five hours. It is saying the day’s energy equals five hours at full STC irradiance. The array generates from first light to last, at a varying rate. The mistake is natural because the unit has the word “hours” in it and because a daily energy figure is easier to grasp as a block of time than as an integral.
The second misreading is to compare peak sun hours across tilts without noticing. A dataset queried for horizontal irradiance will report lower winter values than one queried for a 35-degree tilt at the same site. Neither is wrong; they answer different questions. A reader who checks a proposal against a horizontal dataset will conclude the design is overstated when it is not.
The third is to treat the annual figure as a daily guarantee. Peak sun hours are averages. A single overcast day can deliver a tenth of the monthly mean, and a battery or grid connection is what carries the load across it. The average is the right input for sizing over a year and the wrong input for predicting any particular Tuesday.
When a cause does not apply
| Observation | Likely cause | Does not apply when |
|---|---|---|
| Output lower than the proposal predicted | Derate assumed too low, or soiling | The comparison month was unusually cloudy; check against the long-term mean, not one month |
| Winter output much lower than summer | Lower sun angle and shorter day | The array is steeply tilted for winter; the gap narrows and can reverse at high latitude |
| Output drops on hot afternoons | Temperature coefficient | The array is well ventilated and the cells stay near ambient; the loss is smaller |
| Morning or evening output low | Shading, or orientation away from south | The array faces east or west by design; the curve is meant to be flatter |
| Two nearby sites report different figures | Different tilt, azimuth or dataset convention | Both queries used the same tilt and the same irradiance product |
Reading a production graph against the number
A daily production curve should peak near solar noon, be roughly symmetric on a clear day, and integrate to the array size times peak sun hours times the derate. A curve that is flat-topped suggests clipping at the inverter, which is a sizing choice rather than a fault. A curve with a notch at a fixed time each day suggests a shading object, and the notch moves as the sun’s path shifts through the year. A curve that is uniformly lower than expected across all conditions points at the derate assumption or at soiling, not at the weather.
Safety note: a PV array produces dangerous DC voltage whenever light falls on it and cannot be switched off at the panel. Work on the DC side, roof work and the mains connection are licensed trades in most jurisdictions. The reader should not open a combiner, probe a string or disconnect an energised connector.
Interconnection rules, export compensation and permitting are set by state and utility and change. Any figure a proposal quotes for exported energy should be checked against the current tariff and the local interconnection agreement.
Common questions
What are peak sun hours in simple terms?
They are a way of expressing the day's solar energy as hours at a fixed reference intensity of 1,000 W/m². Five peak sun hours means the energy received that day equals five hours at that intensity, not that the sun shone for five hours.
How are peak sun hours different from daylight hours?
Daylight hours count time above the horizon. Peak sun hours count energy, weighted by how strong the irradiance was. A long hazy day can deliver fewer peak sun hours than a shorter clear one.
Is peak sun hours the same as solar irradiance?
No. Irradiance is instantaneous power per area, in W/m². Peak sun hours are accumulated energy divided by 1,000 W/m², so they carry units of hours and describe a day or a month, not a moment.
Why does my system produce less than array size times peak sun hours?
Because the product ignores losses. Soiling, wiring, inverter conversion and cell temperature above 25 °C all reduce output, and a derate factor of roughly 0.75 to 0.85 is applied to account for them.
Where do I find peak sun hours for my location?
Public datasets such as PVGIS, the Global Solar Atlas and the U.S. National Solar Radiation Database publish monthly insolation for a given latitude, longitude and tilt. Check whether the figure is for a horizontal or tilted plane before comparing it to a proposal.
Read next
- Solar Panel Efficiency: What the Datasheet Number Means Understand what solar panel efficiency measures, why higher efficiency isn't always better, and when it truly matters for your roof.
- PV System Components: What Each Part Does Learn what makes up a photovoltaic system: panels, inverter, mounting, isolation, metering. Understand each component's role and failure modes.
- Solar DC or AC: How Panels Produce and Convert Power Solar panels produce DC electricity; homes use AC. Learn why conversion is needed, the losses involved, and why both currents are essential in a PV system.
- Time-of-Use Rates and Solar: How Peak Pricing Works Understand time-of-use electricity rates and their impact on solar savings. Learn how peak pricing, net metering, and battery storage interact.
- Where Solar Savings Actually Come From Understand solar savings: offset consumption, export credit, avoided rate increases. Learn how payback works without specific figures.