Solar System Size Calculator: From Daily Usage to Panel Count
28 kWh a day: that's a realistic total for a mid-size home once central air, a couple of computers, and the everyday appliance mix are all added up. It's also the number every solar quote eventually has to translate into an actual panel count before a homeowner can hold it up against their own roof and budget. The math in between (daily usage, how much usable sun the site actually gets, how much the system loses along the way, and the wattage of the specific panel being quoted) is exactly what separates a 12-panel system from a 22-panel one.
The calculator below runs that chain live: change any input and the required array size, panel count, and a rough inverter range recompute immediately.
For context, residential solar installs in the US commonly land somewhere in the 5-10 kW range, so a calculated requirement in the high single digits, like the worked example below, sits well inside that typical band rather than at an unusual extreme. A larger load (an all-electric home with a heat pump and an EV, for instance) can push the calculated array size well past that range, which is one reason the same formula is worth re-running with a household's own numbers rather than reusing someone else's result.
- Required array size
- 6.73 kW
- Panel count
- 17
- Inverter range
- 6.7 to 8.4 kW
How daily usage becomes a required array size
Four numbers feed the sizing formula, and each one comes from a different source. Daily usage in kWh usually comes straight off a utility bill, or from a room-by-room estimate like the one on this site's home energy use page. Peak sun hours describes how much usable sunlight a specific site gets in an average day, a very different figure from hours of daylight, covered in the FAQ below. System losses bundle together everything that keeps a panel from delivering its full rated output in the field: inverter conversion loss, wiring resistance, dust and soiling on the glass, and heat derating on hot days. Panel wattage is simply the rating stamped on the specific panel model being priced.
Combining the first three gives the array size the site needs, in kilowatts:
Dividing by (1 − losses) rather than multiplying by it is what makes the required array bigger than the raw daily-usage number: if a fifth of the array's rated output never reaches the meter, the array has to be built a fifth larger than loss-free arithmetic alone would suggest.
From array size to a panel count and inverter range
Panel count follows directly: multiply the required array kW by 1000 to get watts, divide by the panel's own wattage, and round up: a fraction of a panel isn't a purchasable unit.
Inverter sizing is a separate, looser step. As a rough rule of thumb, a string or hybrid inverter is typically sized somewhere between 100% and 125% of the array's kW rating. The calculator shows that as a band rather than a single number, because the right figure inside that band depends on the specific inverter's own spec sheet and on whatever interconnection rules the local utility applies.
That band exists because of a design trade-off installers call the DC-to-AC ratio, or inverter clipping. An inverter rated right at 100% of the array's DC kW leaves little headroom for the panels' factory tolerance or for brief output above standard test conditions, while one rated toward the low end of the band will clip off a sliver of the array's peak output on the sunniest hours of the sunniest days. Many installers accept a small, deliberate amount of clipping as a reasonable trade for a smaller, less expensive inverter, which is part of why the accepted range spans a full 25 percentage points rather than pointing to a single number.
Worked example: a 28 kWh/day home
28 kWh/day usage, 5.2 peak sun hours (a typical continental-US average: real values run roughly 3-6 depending on location), 20% system losses, 400 W panels.
- 1
Apply the array-size formula
kW = 28 / (5.2 × (1 − 0.20))
- 2
Solve
28 / 4.16 = 6.73 kW
- 3
Convert to panels at 400 W each, rounding up
ceil(6730.8 / 400) = 17 panels
- 4
Read the inverter band, 100-125% of array kW
6.73-8.41 kW
This is a planning-level estimate, not a substitute for a site assessment from a licensed solar installer: real sun hours, losses, and available roof area all need site-specific confirmation before a system is finalized.
Why the real number moves around this estimate
Peak sun hours are not a fixed physical constant for a location: they shift with the season, with typical local cloud cover, and with the specific roof plane a panel actually sits on. The default used in the calculator above is a rounded, continental-US-style average meant to get a first estimate in the right neighborhood; a real design should start from a site-specific figure, ideally one that already accounts for the roof's own orientation and tilt rather than a flat regional number.
System losses move around too. A well-ventilated roof in a mild climate loses less to heat derating than a dark, poorly-ventilated one in a hot climate; a system with a shorter or cleaner wire run loses less to resistance than one with a long run to a distant subpanel; and a panel array that gets regularly rinsed by rain loses less to soiling than one in a dusty or low-rainfall region. The 20% figure used above is a common, typical combined estimate for all of these effects together. A real installer's design may reasonably land a few points on either side of it.
Panel output also drifts slowly downward over the system's lifetime as the panels themselves degrade. The FAQ below covers roughly how much, and how that factors into whether an installer pads the initial sizing to compensate.
Finally, the math above answers "how big does the array need to be," not "will it physically fit." A common panel footprint, including standard racking spacing, works out to roughly 20 square feet per panel, so the 17-panel worked example would need on the order of 340 square feet of usable, unshaded roof area, before accounting for setbacks, vents, chimneys, or any section of roof that falls in shadow for part of the day. Available roof area, shading, and local permitting rules can all cap the achievable array size regardless of what the daily-usage math alone suggests.
Peak sun hours: rough regional averages
Peak sun hours vary by location, season, and weather. The figure that matters for a real design comes from an actual site assessment, not a national table. The regional averages below only illustrate the size of that spread; treat every number here as a rounded, illustrative estimate, not a value to design a system around.
| Illustrative US region | Rough peak sun hours |
|---|---|
| Desert Southwest | ~6.5 |
| Southern California / Southwest | ~6.0 |
| Texas / Gulf Coast | ~5.0 |
| Southeast | ~4.7 |
| Mid-Atlantic / Northeast | ~4.0 |
| Pacific Northwest | ~3.4 |
Rough, illustrative averages only, not measured values, and not a substitute for a site-specific solar assessment.
Panel count at different daily-usage levels
Holding panel wattage, sun hours, and system losses fixed at the worked-example values (400 W panels, 5.2 sun hours, 20% losses), panel count scales in a near-linear way with daily usage. These figures are for comparison only: the calculator above recomputes from whatever sun-hours, losses, and panel-wattage inputs actually apply to a real site.
| Daily usage | Required array | Panel count (400 W) |
|---|---|---|
| 15 kWh/day | 3.61 kW | 10 panels |
| 20 kWh/day | 4.81 kW | 13 panels |
| 25 kWh/day | 6.01 kW | 16 panels |
| 28 kWh/day | 6.73 kW | 17 panels |
| 35 kWh/day | 8.41 kW | 22 panels |
| 40 kWh/day | 9.62 kW | 25 panels |
Reference estimates only, not measured values. Confirm any real system against a site-specific sun-hours figure and the installer's own loss assumptions.
Sizing the loads a solar array feeds
Questions
Solar system size FAQ
Sizing questions that come up once a rough panel count needs to turn into an actual system design.
What does net metering mean, and how does it change system sizing?
Net metering is a billing arrangement where excess solar production sent back to the grid earns a credit against electricity drawn later, and whether, and how generously, a utility offers it can change how big a system makes financial sense to build. Where net metering is generous, oversizing an array slightly to bank summer surplus against winter usage can make sense; where it's absent or compensated at a much lower rate, sizing closer to actual daily usage avoids paying for panels whose extra output is nearly worthless. Rules vary enormously by utility and by state or country, so the local net-metering policy is worth confirming before finalizing a system size.
How does panel degradation affect sizing over the system's lifetime?
Solar panels lose a small amount of output capacity every year. Commonly cited manufacturer figures are in the neighborhood of 0.3-0.5% a year, so a system sized exactly to today's usage will fall slightly short of that usage a decade or two later. Some installers size the initial array a few percent above the calculated requirement specifically to offset this drift, while others plan on it and treat any shortfall as a gradual multi-year creep rather than a step change. Either approach is a judgment call for the installer, not something the daily-usage sizing formula above accounts for on its own.
Why aren't peak sun hours the same as daylight hours?
Daylight hours count every hour the sun is above the horizon, including early morning, late afternoon, and overcast periods when very little usable energy actually reaches a panel; peak sun hours compress all of that variable daylight into an equivalent number of hours at a fixed reference intensity. A location might get 14 hours of daylight in summer but only 5-6 peak sun hours, because most of those daylight hours deliver far less than the reference intensity. Using daylight hours in the sizing formula in place of peak sun hours would badly undersize the array.
Is battery storage sized separately from the solar array?
Yes, battery storage and array sizing answer two different questions and are normally calculated separately. Array sizing (the calculation above) asks how many panels are needed to generate enough energy over a full day. Battery sizing asks how many kWh of storage are needed to cover a specific outage duration or a specific evening and overnight usage window, and depends on load priorities, desired backup duration, and the battery's own usable-capacity spec, none of which feed into the array-sizing formula above.
How do roof orientation and tilt change the sun-hours figure to use?
The peak sun hours figure used in a real design already has to account for a specific roof's orientation and tilt, not just its geographic location. A south-facing roof close to the site's latitude angle captures close to the regional maximum, while an east- or west-facing roof, a flat roof, or a steeply pitched one can lose a meaningful share of that figure. A solar installer's site assessment typically models this loss directly for the actual roof planes being proposed, which is why the regional averages further down this page are a starting point for a rough estimate, not a number to plug in for a specific roof without adjustment.
Does a bigger panel wattage always mean fewer panels for the same roof?
For a fixed required array size, yes: a higher-wattage panel needs fewer units to reach the same total kW, which can matter when roof space, not budget, is the limiting factor. It isn't automatically the cheaper or better choice, though: higher-wattage panels are often physically larger and can cost more per panel, so the real trade-off is between panel count, available roof area, and total cost per installed watt, not wattage alone.
How much does cloud cover or shading change the sun-hours estimate?
Both reduce the usable sun-hours figure below a location's clear-sky average, and shading in particular can cut output far more than its area alone suggests, because a single shaded panel on some system designs can throttle the output of an entire string. Persistent cloud cover is usually already baked into a regional peak-sun-hours average since it reflects typical weather, not clear-sky conditions, but site-specific shading from trees, chimneys, or neighboring buildings is not, which is another reason a rough regional number is a planning starting point rather than a final design input.