How To Calculate Solar Panel Output (Watts → kWh, Day / Month / Year)

The frustrating thing about every "how to calculate solar panel output" article is that it gives you a formula like "Output = wattage × peak sun hours × 0.75" and stops there. Where does the 0.75 come from? Why not 0.77 or 0.86? Why does PVWatts give a different number? Why does your installer give yet another number? The single-line formula hides almost all the interesting physics — and most of the room for being wrong.

This article opens the formula up and shows you every term. I built my own 6 kW array on my own roof in 2024, so I have a personal stake in getting this number right. By the end you'll be able to estimate solar panel output for any system size, in any U.S. location, at any age — with the same methodology NREL uses, and the same assumptions every credible installer in the country runs.

The Master Formula

kWh per day = system kW × peak sun hours per day × derate factor

Three inputs, one output. Each one of them deserves its own section.

Input 1: System size in kW

This is the DC nameplate rating. For a single panel, divide its watts by 1,000 — a 400 W panel is 0.4 kW. For a system, add up the nameplate of every panel — twelve 400 W panels is 4.8 kW. Use the manufacturer datasheet number, not whatever your installer rounded to in marketing copy. The formula is calibrated against the STC rating (IEC 60904-3), not against the panel's real-world hot-day output, so don't pre-derate.

Input 2: Peak sun hours per day (PSH)

This is the equivalent number of hours per day during which solar irradiance averages exactly 1,000 W/m² — the same reference intensity used to rate solar panels at STC. The U.S. average is 4.98 PSH/day. Arizona is the highest at 6.54, Alaska the lowest at 3.17. To find your exact location's PSH, use our peak sun hours calculator — it pulls directly from NREL PVWatts v8 and the National Solar Radiation Database, the same dataset every U.S. solar installer uses.

A common mistake: confusing peak sun hours with hours of daylight. They are not the same. Daylight is how long the sun is above the horizon (10–16 hours depending on season). Peak sun hours is the integrated daily energy at the standard 1,000 W/m² intensity — usually 3.5 to 7 hours.

Input 3: Derate factor

This is the multiplier that converts the panel's lab-condition DC nameplate into real-world AC output at your meter. It accounts for everything that goes wrong between the cell and the wall socket. The PVWatts v8 default is 0.83. The most-quoted rule of thumb is 0.77, which builds in a small margin of safety. The next section breaks down exactly what this number contains.

Where The 14% Loss Comes From (PVWatts v8 Methodology)

NREL PVWatts v8 uses a default total system loss of 14%, which corresponds to a derate factor of 0.86 on the DC side. This isn't a single number — it's eight independent loss categories combined multiplicatively (PVWatts v5 Manual, Dobos 2014):

These don't add to 14% — they multiply. The combined formula is:

total loss = 1 − (1 − 0.02) × (1 − 0.03) × (1 − 0.02) × (1 − 0.02) × (1 − 0.005) × (1 − 0.015) × (1 − 0.01) × (1 − 0.03)

= 14.076%

That gives a DC derate of 0.86. Then the inverter is modeled separately (not in the 14% number) using a quadratic part-load efficiency curve based on California Energy Commission inverter test data. A modern string inverter has a CEC weighted efficiency of about 96%, so the combined DC + AC derate works out to:

0.86 × 0.96 = 0.83

That's the PVWatts v8 default annual derate, and it's the number you should use if you want your formula to match what every U.S. installer's PVWatts run will return.

Why "0.77" is the most-quoted number

The older PVWatts v1 (used until 2014) folded everything — including the inverter — into a single 0.77 default. That value got quoted in tens of thousands of articles before PVWatts v5 split the inverter out, and it stuck as the "rule of thumb." It's still a good practical default because:

1. It builds in a small margin of safety vs. PVWatts v8 (about 7% conservative)
2. It absorbs unknown future degradation
3. It tracks closely with what most installers will actually quote you

So in this article we'll use 0.83 for PVWatts-aligned numbers and 0.77 for conservative real-world planning. Both are defensible. The 0.75 number some older articles use is too pessimistic for modern equipment — it dates from the early 2010s when inverters were 92–93% efficient and string mismatch was higher.

Worked Examples

Let me work through three system sizes in three locations using the 0.83 PVWatts default, then redo one in 0.77 conservative so you can see the difference.

Example 1: 400 W panel in Phoenix, AZ (6.54 PSH)

0.4 kW × 6.54 PSH × 0.83 derate = 2.17 kWh/day

Multiply by 365 for the annual: 793 kWh/year.

A typical U.S. household uses 10,500 kWh/year, so a single 400 W panel covers 7.6% of an average home in Phoenix.

Example 2: 6 kW residential system in Boston, MA (4.70 PSH)

6 kW × 4.70 PSH × 0.83 derate = 23.4 kWh/day

Annual: 8,541 kWh/year — about 81% of average household use.

This is the most common residential configuration in the U.S. For Boston specifically, you'd want a slightly bigger system (8 kW, ~11,400 kWh/year) to fully offset usage.

Example 3: 10 kW system in Los Angeles, CA (5.61 PSH)

10 kW × 5.61 PSH × 0.83 derate = 46.6 kWh/day

Annual: 17,000 kWh/year — about 162% of average household use, leaving plenty of headroom for an EV (3,000–4,000 kWh/year) or electric heat pump.

Conservative redo: Example 2 with derate 0.77

6 kW × 4.70 PSH × 0.77 derate = 21.7 kWh/day → 7,924 kWh/year

That's 7.2% lower than the PVWatts default. If you want a number you can confidently underpromise on, use 0.77. If you want a number to compare against an installer's quote, use 0.83.

Temperature Derate (The Loss That Isn't In The 14%)

Notice that PVWatts's 14% number doesn't include temperature loss. That's intentional — temperature is separately baked into PVWatts's hourly model using each location's actual NSRDB temperature data. But if you're estimating output by hand, you need to account for it. Here's how.

Solar cell efficiency falls as cell temperature rises. The rate of fall is the temperature coefficient of P_max (β), measured in % per °C above STC. Modern panels have:

(Sources: LONGi Hi-MO 6, Maxeon 7, JinkoSolar Tiger Neo, REC Alpha Pure-R datasheets, 2024–2026.)

Cell temperature is typically 25–35°C above ambient air temperature when the panel is in full sun on a roof with good airflow underneath. So:

• A 30°C summer day → cell temp ~60°C → 35°C above STC → ~10.5% loss for TOPCon
• A 35°C summer day → cell temp ~65°C → 40°C above STC → ~12.0% loss for TOPCon
• A 15°C spring day → cell temp ~45°C → 20°C above STC → ~6.0% loss for TOPCon
• A 5°C winter day → cell temp ~30°C → 5°C above STC → ~1.5% loss for TOPCon (panels actually outperform STC on cold sunny days)

This is averaged into the PVWatts hourly model. If you're adjusting for a hot climate by hand, expect to lose another 8–12% on top of the 14% system loss during peak summer months. PVWatts already factors this in for the annual number — but its monthly breakdown is more accurate than any hand calculation.

Year 1 vs Year 25: The Degradation Curve

Solar panels lose a small amount of output each year. The standard tier-1 industry warranty:

• Year 1: 2–3% loss (light-induced degradation, "LID")
• Year 2 onward: 0.5%/year (linear degradation)
• Year 25: ~85% of nameplate output retained

Premium panels do significantly better:

To estimate output in year N:

year-N output = nameplate × (1 − year-1 LID) × (1 − annual rate)^(N−1)

For a Maxeon 7 in year 25: 1 × (1 − 0.01) × (1 − 0.0025)^24 = 0.99 × 0.9417 = 0.932 (93.2% of original).

For a generic tier-1 panel in year 25: 1 × (1 − 0.025) × (1 − 0.005)^24 = 0.975 × 0.8869 = 0.865 (86.5% of original).

For lifetime production estimates, use the average of year 1 and year 25 output as your annual figure. For a 6 kW system that produces 8,541 kWh in year 1, year 25 would be about 7,388 kWh (tier-1) or 7,950 kWh (premium). The 25-year lifetime average is about 7,965 kWh/year for tier-1 and 8,245 kWh/year for premium.

Solar Output Calculator

Skip the math and use the calculator. Set the wattage to your panel rating, set peak sun hours to your location's value (use the peak sun hours calculator to look it up), and read off daily, monthly, and annual production.

Here's how this works in plain English: take a 400 W panel in a 5.32 PSH location (the U.S. average for residential roofs), and the calculator gives you 1.77 kWh/day, 53.7 kWh/month, 645 kWh/year. Same formula as everything above — just with the slider doing the multiplication for you.

Solar Output Lookup Table (50 W → 15 kW At 5 PSH)

For quick reference, here's daily output for every common panel and system size, calculated at the U.S. residential median of 5 peak sun hours/day with derate 0.83.

To recompute for your location, multiply each value by (your PSH ÷ 5). Phoenix at 6.54 PSH? Multiply by 1.31. Boston at 4.70 PSH? Multiply by 0.94. Seattle at 3.95 PSH? Multiply by 0.79.

Solar Output Lookup Table (6 kW System By Top US Locations)

To make it concrete, here's what a 6 kW residential system produces in 12 representative U.S. cities. All numbers use the PVWatts v8 default derate of 0.83 and the actual NREL annual PSH for each city.

Average U.S. household use is 10,500 kWh/year (EIA, 2024). A 6 kW system covers most of an average home almost everywhere south of the 40th parallel, and a fraction of one in cloudy / high-latitude locations. If you live in Boston, Seattle, or anywhere in the Pacific Northwest, plan around an 8–10 kW system instead of 6.

Common Mistakes When Calculating Solar Output

After helping people on Reddit and forums for two years, these are the five most common errors I see:

1. Using daylight hours instead of peak sun hours

A 14-hour summer day in Phoenix is not 14 PSH. Phoenix gets about 7–8 PSH on a clear summer day. Daylight is the wrong number — only the integrated energy at the 1,000 W/m² reference matters.

2. Forgetting that PSH varies by month

The annual average works for annual estimates. For monthly estimates, you need the actual monthly PSH value for your location — December in Boston is 2.3 PSH, June in Boston is 6.1 PSH. Use the peak sun hours calculator to get the full 12-month array.

3. Confusing watts and kilowatts

A 400 W panel is 0.4 kW, not 400 kW. The formula uses kW. The most common version of this mistake produces a number 1,000× too big.

4. Comparing nameplate watts across different panel formats

A 400 W TOPCon panel and a 400 W HJT panel produce the same kWh per year at the same location, but the HJT panel will produce slightly more in hot climates because of its better temperature coefficient (about 2% better over a year in Phoenix). The difference is small enough that nameplate is usually the right comparison — but if you're in a hot climate, premium HJT or back-contact panels deserve a small bonus in your calculation.

5. Pre-derating an already-derated number

If your installer hands you "estimated annual production" of 9,500 kWh and you multiply that by 0.77 to "be conservative," you're double-derating. The installer's number is already AC-side, post-derate. Take it at face value.

Bottom Line

Three numbers, one formula:

kWh/day = kW × PSH × derate

Use 0.83 to match PVWatts (what your installer will quote). Use 0.77 to underpromise. Multiply by 365 for annual, by 30/31 for monthly. Look up your peak sun hours with the peak sun hours calculator. Look up your panel watts on the manufacturer datasheet. That's it. Everything else in this article is the why behind those three steps — useful when you want to defend a number to a contractor, sanity-check a quote, or estimate output in 2046 when your panels are 25 years old.

Keep Reading

• Peak Sun Hours Calculator — Find PSH At Your Location — the standard input to this formula
• Standard Solar Panel Sizes And Wattages (100W–600W) — for the kW input
• Solar Panel Watts Per Square Foot — the area-based companion
• STC vs NMOT Solar Panel Test Conditions — where the nameplate watts come from
• Average Peak Sun Hours By State — full state table
• How Much Power A 5 kW Solar System Produces
• How Much Power A 10 kW Solar System Produces
• Rooftop Solar Calculator — How Many Panels Fit On Your Roof