Every residential solar quote comes with a headline number: the estimated annual kilowatt-hour output your system will produce. It is almost always higher than what your roof will actually deliver.
Not because installers are dishonest — most are not — but because the calculation starts from laboratory conditions and applies only some of the real-world penalties, and sometimes none of them. The result is a number that looks great in a sales presentation and disappoints in year two when you check your actual meter readings.
This article is the calculation sheet your installer probably did not give you. It covers peak sun hours, the standard testing gap, every derating variable that chips away at nameplate output, roof geometry penalties, and the degradation curve your panels will follow across their 25-year lifespan. By the end, you will be able to run the math yourself — for your specific roof, in your specific city, facing your specific direction.
Peak Sun Hours: The Most Misunderstood Metric in Solar
The phrase "peak sun hours" sounds like it should mean "hours of daylight." It does not. This is the single most common misconception in residential solar, and getting it wrong will make every other piece of the calculation wrong.
Peak Sun Hours vs. Daylight Hours
A peak sun hour is defined as one hour during which the solar irradiance on a flat surface averages 1,000 watts per square meter (1 kW/m²). This is the standard test condition (STC) intensity used to rate solar panels in laboratories.
In reality, the sun never maintains exactly 1,000 W/m² for clean clock-hour blocks. It rises gently, reaches a midday peak somewhere between 800 and 1,100 W/m² depending on your latitude and the season, and tapers off in the afternoon. Solar scientists take the total daily solar energy received at a location (measured in kWh/m²) and express it as an equivalent number of hours at the standard 1,000 W/m² intensity.
So if your location receives 5.5 kWh/m² of solar energy in a typical day, it has 5.5 peak sun hours — regardless of whether the sun was visible for 14 daylight hours. Phoenix, Arizona gets around 6.5 peak sun hours per day on an annual average. Seattle, Washington gets around 3.5. Both cities get roughly 14 hours of daylight in summer. Daylight hours tell you almost nothing about solar production potential.
How Peak Sun Hours Are Measured
The data comes from the National Renewable Energy Laboratory (NREL), which maintains a database called the National Solar Radiation Database (NSRDB). It synthesizes decades of satellite and ground-station measurements into a grid-level dataset that covers the entire continental United States at high resolution. When you give an installer your address, the serious ones use NREL's PVWatts calculator to pull location-specific peak sun hour averages. The less careful ones use a rough regional estimate that can be off by 10–20%.
Peak Sun Hours by U.S. City
| City | Avg. Daily Peak Sun Hours | Annual kWh/m² |
|---|---|---|
| Phoenix, AZ | 6.5 | 2,373 |
| Las Vegas, NV | 6.3 | 2,300 |
| San Diego, CA | 5.8 | 2,117 |
| Dallas, TX | 5.3 | 1,935 |
| Denver, CO | 5.5 | 2,008 |
| Atlanta, GA | 5.0 | 1,825 |
| Chicago, IL | 4.5 | 1,643 |
| Boston, MA | 4.7 | 1,716 |
| New York, NY | 4.7 | 1,716 |
| Seattle, WA | 3.5 | 1,278 |
| Miami, FL | 5.6 | 2,044 |
| Portland, OR | 3.8 | 1,387 |
These figures represent horizontal surface averages. A south-facing tilted panel will capture more than these numbers suggest — the exact gain depends on tilt angle and your latitude, which we will cover shortly.
STC vs. PTC: The Rating Gap Nobody Explains
When a solar panel's data sheet says "400W," that number was measured at Standard Test Conditions: 1,000 W/m² irradiance, 25°C cell temperature, and a specific air mass coefficient. These conditions exist in a laboratory. They almost never exist on your roof at the same time.
PTC — PVUSA Test Conditions — is a closer proxy to real-world output. It uses 1,000 W/m² irradiance but 20°C ambient air temperature and assumes a wind speed of 1 m/s, which better reflects actual operating conditions. A 400W panel rated at STC will typically produce 370–385W under PTC. That 3–4% gap is before any system-level losses.
The practical implication: when you see "10 kW system" on a solar quote, that is 10 kW at STC. Under real sky conditions on an average day, that same array is closer to 9.3–9.6 kW at its operational peak — and it will almost never operate at peak for the entire peak sun hour window.
The Formula, Step by Step
Converting your system's nameplate capacity and your location's peak sun hours into an estimated annual kWh output is a four-step process. Here is each step with the logic behind it.
Step 1 — Nameplate DC Output
Start with your system size in kilowatts (DC). This is the sum of all your panels' STC ratings. A system with 25 panels at 400W each is a 10 kW DC system.
DC Nameplate (kW) = Number of Panels × Panel Wattage (W) ÷ 1,000
Step 2 — Apply Peak Sun Hours
Multiply your DC nameplate by your location's average daily peak sun hours. This gives you the theoretical daily DC energy production at STC — under ideal conditions with no losses anywhere in the system.
Theoretical Daily DC Output (kWh) = DC Nameplate (kW) × Daily Peak Sun Hours
A 10 kW system in Denver (5.5 peak sun hours) produces a theoretical 55 kWh per day at this stage.
Step 3 — Apply the Derating Factor
This is the step most quotes underexplain. A derating factor is a single multiplier (between 0 and 1) that captures all the real-world losses that reduce output below the theoretical STC value. The California Energy Commission (CEC) and NREL's PVWatts use a default system derating of approximately 0.80 for a standard residential installation — meaning real output is roughly 80% of the STC theoretical maximum.
We will break this number down in the next section. For now:
Real Daily AC Output (kWh) = Theoretical Daily DC Output × Derating Factor
55 kWh × 0.80 = 44 kWh per day (realistic estimate for Denver)
Step 4 — Annualize
Multiply by 365 days for a single annual estimate. For a more precise figure, run the monthly calculation separately using each month's average peak sun hours and temperature, then sum — but the annual average method is accurate within 3–5% for most continental U.S. locations.
Annual Output (kWh) = Real Daily AC Output × 365
44 kWh × 365 = 16,060 kWh per year for a 10 kW system in Denver, with a 0.80 derating factor.
Derating Factors Broken Down
The 0.80 default derating factor in PVWatts is a composite of eight independent loss sources. Understanding each one matters because they are not equally inevitable — some are fixed physics, some are design choices, and some are maintenance variables you can control.
Temperature: The Biggest Surprise
Silicon photovoltaic cells lose efficiency as they heat up. Every panel's data sheet includes a temperature coefficient, typically expressed as a percentage of power loss per degree Celsius above 25°C. Modern monocrystalline panels carry a temperature coefficient of around -0.35% per °C (premium panels) to -0.50% per °C (standard panels).
On a hot summer day, a rooftop panel's cell temperature can reach 65°C — that is 40°C above the STC reference. At -0.40% per °C, that is a 16% efficiency loss on your hottest production days. The panels that are generating the most solar input are simultaneously losing the most efficiency to heat. This is counterintuitive and consistently catches homeowners off guard.
In Phoenix — where solar production potential is highest — summer cell temperatures routinely hit 70°C, producing temperature derating losses of 18–20% on peak days. The abundant sunshine partially compensates, but the temperature penalty is real and significant.
The Full Derating Stack
Each loss source multiplies the others — they do not simply add. The combined effect is what produces the overall system derating factor.
| Loss Source | Typical Range | Best Case | Worst Case |
|---|---|---|---|
| Inverter efficiency | 2–4% loss | 1.5% | 5% |
| Temperature derating | 5–10% loss | 3% | 18% |
| Soiling (dust, pollen, birds) | 2–5% loss | 1% | 10% |
| DC wiring resistance | 1–3% loss | 0.5% | 4% |
| AC wiring resistance | 0.5–1% loss | 0.3% | 2% |
| Module mismatch | 1–2% loss | 0.5% | 4% |
| Shading (tree, chimney, vent) | 0–15% loss | 0% | 30%+ |
| System downtime / availability | 0.5–2% loss | 0.3% | 4% |
A modern system with good siting, quality inverters, and minimal shading will realistically land around 0.84–0.87 total derating. A system with moderate shading, average temperatures above 30°C, and standard inverters is closer to 0.76–0.80. A system in a hot climate with significant shading, undersized wiring, and older inverter technology can fall to 0.68–0.72.
The PVWatts default of 0.80 is the midpoint of a range that actually spans ±10%. If your installer is quoting you a single number without explaining the derating assumptions, it is worth asking which specific losses they included and what values they assumed for each.
Roof Geometry: The Variables Your Quote May Ignore
The peak sun hour table earlier showed horizontal surface averages. Your panels are not horizontal — they are mounted at whatever tilt your roof pitch provides, and they face whatever compass direction your roof faces. Both of these factors significantly affect how many of those peak sun hours your panels actually capture.
Orientation: Which Direction Your Panels Face
In the Northern Hemisphere, a south-facing roof is the ideal orientation for solar panels. It maximizes exposure during the peak production window (roughly 9 AM to 3 PM solar time). The yield penalty for other orientations, compared to true south, looks like this on an annual average basis:
| Panel Orientation | Annual Yield vs. True South |
|---|---|
| True South (180°) | 100% — baseline |
| Southeast (135°) or Southwest (225°) | 95–97% |
| East (90°) or West (270°) | 80–85% |
| Northeast (45°) or Northwest (315°) | 65–72% |
| True North (0°) | 55–65% |
East/west split systems — where panels are installed on both the east and west sides of a roof — can sometimes outperform a pure south system on a household energy-matching basis, because they generate more power during morning and evening hours when usage is higher. But for raw annual kWh output, south-facing with optimal tilt is still the maximum.
Tilt Angle: Latitude Is Not Just a Geography Lesson
The optimal tilt angle for a fixed panel in the Northern Hemisphere is approximately equal to your latitude. A home in Denver at 39.7° latitude would ideally have panels tilted at roughly 38–40° off horizontal. Most residential roofs have pitches between 15° and 40°, which means they incidentally land close to optimal for many U.S. locations.
| Tilt vs. Optimal | Annual Yield Penalty |
|---|---|
| Within 10° of optimal | Less than 3% loss |
| 15° off optimal | 4–6% loss |
| 25° off optimal | 8–12% loss |
| Flat (0°) — commercial rooftops | 10–15% loss vs. tilted |
Flat commercial rooftop installations often use tilt-up racking to recover this loss. For residential rooftops, the existing pitch is typically accepted as-is, and the tilt penalty is factored into the design estimate — or should be.
Shading: The Silent Killer
Shading is in a category of its own because its impact is nonlinear. A small shaded area on one panel can reduce the output of an entire string, not just that single panel, depending on your inverter technology.
Traditional string inverters connect panels in series. The string's output is constrained by the weakest panel — exactly like old-style Christmas lights where one dead bulb killed the whole string. A single panel with 30% shading from a tree branch during peak hours can reduce the entire string's output by 20–30%.
Microinverters and DC optimizers (from manufacturers like Enphase and SolarEdge) solve this by isolating each panel electrically, so shading on one module no longer degrades its neighbors. If your roof has any shading — from trees, a chimney, a dormer, a neighboring building, or even a plumbing vent pipe — microinverters or optimizers are not a luxury upgrade. They are the difference between recovering 85% of your theoretical output and recovering 60%.
Worked Example: A 10 kW System in Three Cities
Let us run the full calculation for a 10 kW south-facing system at a 30° tilt, with a realistic 0.82 derating factor, in three different cities.
| City | Daily Peak Sun Hours | Daily DC Theoretical | After Derating (×0.82) | Annual kWh |
|---|---|---|---|---|
| Phoenix, AZ | 6.5 | 65.0 kWh | 53.3 kWh | 19,455 |
| Denver, CO | 5.5 | 55.0 kWh | 45.1 kWh | 16,462 |
| Seattle, WA | 3.5 | 35.0 kWh | 28.7 kWh | 10,476 |
The Phoenix system produces nearly 1.9× the output of the Seattle system with identical hardware. This is why "solar makes no sense in Seattle" is a claim worth examining carefully — a 10 kW system there still generates over 10,000 kWh annually, which covers roughly 75–80% of average Pacific Northwest household electricity consumption. The math just requires a larger system to hit the same coverage targets compared to the Southwest.
Now compare against an overly optimistic quote that uses no derating factor at all (100% efficiency, theoretical maximum):
| City | Quoted (No Derating) | Realistic (0.82 Derating) | Overstatement |
|---|---|---|---|
| Phoenix, AZ | 23,725 kWh | 19,455 kWh | +22% over |
| Denver, CO | 20,075 kWh | 16,462 kWh | +22% over |
| Seattle, WA | 12,775 kWh | 10,476 kWh | +22% over |
A quote that ignores derating will consistently overstate your expected output by 20–25%. That translates directly into a longer payback period than projected — and sometimes the difference between a system that makes financial sense and one that doesn't.
Year-Over-Year Degradation: The Math Nobody Shows You
Solar panels do not maintain their rated output indefinitely. They degrade — slowly but measurably — due to UV exposure, thermal cycling, and chemical changes in the silicon and encapsulant material. The industry standard warranty guarantees that panels will produce at least 80% of their rated output at year 25, which implies a linear degradation of approximately 0.8% per year.
Premium manufacturers (Panasonic, SunPower, REC) now offer warranties guaranteeing 92% output at year 25, implying a degradation rate closer to 0.5% per year. Standard panels from commodity manufacturers typically degrade at 0.7–0.9% annually.
| Year | Output at 0.5%/yr Degradation | Output at 0.8%/yr Degradation |
|---|---|---|
| Year 1 | 100% | 100% |
| Year 5 | 97.5% | 96.0% |
| Year 10 | 95.1% | 92.2% |
| Year 15 | 92.8% | 88.5% |
| Year 20 | 90.5% | 85.1% |
| Year 25 | 88.3% | 81.7% |
The cumulative production difference between a 0.5%/yr and a 0.8%/yr degradation rate over 25 years is roughly 3.5–4.5% of total lifetime output for a typical system — not a dramatic gap, but real enough to matter when calculating payback periods and net metering credits over a full system life.
A properly modeled 25-year production estimate should apply year-specific degradation factors, not a single flat annual output number. If your installer's proposal shows the same kWh output for year 1 and year 25, the model is incomplete.
What to Do With This Math
You now have enough to do three useful things before signing any solar contract.
Verify the peak sun hours figure. Ask your installer which data source they used. The answer should be NREL's NSRDB or PVWatts. A regional estimate or a number from a brochure is not adequate for a contract-level projection.
Ask for the derating factor breakdown. Request a line-item list of what losses were included: inverter efficiency, temperature derating, soiling, wiring, mismatch, shading, and availability. Add them up and confirm the composite factor. A number above 0.85 for a standard residential install in a warm climate should prompt questions. A number below 0.75 may indicate excessive shading that should be addressed before installation.
Run the formula yourself. Take the system size in kW, multiply by your location's annual peak sun hours per day, apply a 0.80 derating factor as a conservative baseline, and multiply by 365. Compare the result to the number in your quote. If the quote is more than 15% higher than your calculation, ask what assumptions produced the difference.
Solar energy is one of the few home investments where the underlying physics is transparent, the data is publicly available, and the calculation is accessible to any homeowner willing to spend thirty minutes with a spreadsheet. The gap between a confident solar decision and a regretted one is almost always a gap in understanding the math — not a gap in access to it.
