How to size a solar PV system from your electricity use
Size a home solar PV system from 12 months of electricity use: the sizing equation, a worked example, panel count, battery size and simple payback.
A home solar PV system is sized by dividing its yearly electricity use by the energy that one kilowatt of panels produces in a year at the site. For the example home in this article, which uses 9,050 kWh a year, the answer is 6.84 kW DC. That rounds up to 18 panels of 400 W, or a 7.2 kW array. Two inputs carry most of the uncertainty: the peak sun hours at the site and the system losses. The rest is arithmetic, which a spreadsheet handles well.
The home in the example is fictional. Its monthly values, installed cost and electricity price are assumed inputs, to be replaced with your own.
What equation sizes a solar PV system from electricity use?
DC size (kW) = annual use (kWh) × offset ÷ (annual peak sun hours × derate)
- Annual use is the electricity the home consumes in a year, in kWh. Use twelve months of bills or smart meter data, not one month multiplied by twelve.
- Offset is the share of annual use the system is meant to cover. An offset of 100% sizes the system to match a year of use on paper. A lower offset gives a smaller system.
- Annual peak sun hours is the sum, over the twelve months, of each month's daily peak sun hours multiplied by the days in that month. One peak sun hour is one hour at 1 kW per square meter, so a daily irradiance of 5.0 kWh/m² is 5.0 peak sun hours.
- Derate is the fraction of the panels' DC nameplate output that reaches the home as AC energy. It is
(1 − system losses) × inverter efficiency.
System losses cover soiling, wiring resistance, mismatch between panels, temperature and similar effects. A common default for system losses is 14%, and that is the starting point used here. A real site may differ, so check the value against a site-specific estimate. Inverter efficiency is the share of DC energy the inverter converts to AC. This article uses 96% as an example value.
The equation gives a DC nameplate size. Yield per kilowatt is annual peak sun hours × derate, in kWh per kW per year. Dividing annual use by that yield gives the number of kilowatts needed.
What inputs do you need to size a solar PV system?
Three inputs are needed: twelve months of electricity use in kWh, monthly daily peak sun hours for the site, and the rating of the panel in watts. The table gives example values for a fictional home in a mid-latitude climate. They show the method and are not measurements from any site.
| Month | Use (kWh) | Peak sun hours per day | Days |
|---|---|---|---|
| January | 780 | 2.4 | 31 |
| February | 700 | 3.1 | 28 |
| March | 720 | 4.2 | 31 |
| April | 660 | 5.1 | 30 |
| May | 640 | 5.9 | 31 |
| June | 760 | 6.3 | 30 |
| July | 900 | 6.2 | 31 |
| August | 940 | 5.8 | 31 |
| September | 780 | 4.9 | 30 |
| October | 680 | 3.8 | 31 |
| November | 690 | 2.7 | 30 |
| December | 800 | 2.2 | 31 |
What does a worked solar PV sizing example look like?
Work through the sizing one step at a time.
- Annual use. The twelve months add to 9,050 kWh.
- Annual peak sun hours. Each month's daily value multiplied by its days, then summed, gives 1,602.3 hours. January contributes 2.4 × 31 = 74.4 hours and June contributes 6.3 × 30 = 189.0 hours.
- Derate.
(1 − 0.14) × 0.96 = 0.8256. - DC size.
9,050 × 1.00 / (1,602.3 × 0.8256)= 9,050 / 1,322.9 = 6.84 kW. - Panel count. 6.84 kW × 1,000 / 400 W = 17.10 panels. Rounding up with
ROUNDUPgives 18 panels, an installed size of 18 × 400 W = 7.2 kW.
The 7.2 kW array produces 1,322.9 kWh per kW a year, or 9,525 kWh in total. That is 105.2% of the 9,050 kWh use, slightly above 100% because the panel count was rounded up.
Can an annual match hide monthly mismatches in a solar PV system?
An annual match can hide a monthly mismatch. The table compares each month's production from the 7.2 kW array, calculated as installed kW × peak sun hours × days × derate, with that month's use.
| Month | Use (kWh) | Production (kWh) | Production minus use (kWh) |
|---|---|---|---|
| January | 780 | 442.3 | −337.7 |
| February | 700 | 516.0 | −184.0 |
| March | 720 | 774.0 | 54.0 |
| April | 660 | 909.5 | 249.5 |
| May | 640 | 1,087.2 | 447.2 |
| June | 760 | 1,123.5 | 363.5 |
| July | 900 | 1,142.5 | 242.5 |
| August | 940 | 1,068.8 | 128.8 |
| September | 780 | 873.8 | 93.8 |
| October | 680 | 700.2 | 20.2 |
| November | 690 | 481.5 | −208.5 |
| December | 800 | 405.4 | −394.6 |
The array produces more than the home uses from March to October, a surplus of about 1,600 kWh in total. It produces less than the home uses in January, February, November and December, a shortfall of about 1,125 kWh. Measured month by month, production covers 87.6% of use. The annual figure of 105.2% is correct, but it describes a year in which the home sends energy to the grid in summer and buys it back in winter.
Solar output tracks the sun, which is strongest from spring into summer. Household use follows heating, cooling and occupancy, and in this example it peaks in July and August. Whether the summer surplus offsets the winter shortfall depends on the utility's export and net metering rules. Check the tariff before assuming a surplus is worth its retail value.
How do you size a battery for a solar PV system?
A battery sized for whole-home backup uses this formula:
battery (kWh) = daily use (kWh) × autonomy (days) ÷ depth of discharge ÷ round-trip efficiency
- Daily use is annual use divided by 365. Here, 9,050 / 365 = 24.79 kWh.
- Autonomy is the number of days the battery should carry the home without sun or grid power. This example uses one day.
- Depth of discharge is the share of the battery's rated capacity that may be used in each cycle. Manufacturers state it. This example uses 80%, or 0.8.
- Round-trip efficiency is the share of energy put into the battery that comes back out. This example uses 90%, or 0.9.
For one day of autonomy, the battery is 24.79 × 1 / 0.8 / 0.9 = 34.4 kWh. For half a day, the result is 17.2 kWh.
This sizing is for backup. A battery meant to store daytime surplus for use in the evening is sized from the evening load and the surplus available, which is a different calculation.
How do you estimate the simple payback of a solar PV system?
Simple payback divides the net cost of the system by its first-year savings. It ignores interest, maintenance, changes in electricity prices and the timing of cash flows, so it is a screening figure only.
- Gross cost. 7,200 W × $3.00 per W = $21,600. The $3.00 per W is an example installed cost.
- Annual savings. 9,524.6 kWh × $0.18 per kWh = $1,714.43. The $0.18 is an example retail price, and the calculation assumes each kWh produced is valued at that price.
- Simple payback. $21,600 / $1,714.43 = 12.6 years.
- With an incentive. If a rebate or tax credit covered 30% of the cost, the net cost would be $21,600 × (1 − 0.30) = $15,120 and the payback $15,120 / $1,714.43 = 8.8 years.
For US homeowners, the federal residential clean energy credit (Section 25D) was 30% under the Inflation Reduction Act, but Public Law 119-21, signed in July 2025, ended it for expenditures made after December 31, 2025. State, utility and local incentives vary, so enter only the incentives that apply to your system, and remember that a tax credit is worth what you can actually use against your tax. A financed system's monthly payment is a PMT calculation, covered in how to calculate a loan payment with PMT.
What factors does a solar PV spreadsheet leave out?
- Shading. Trees, neighboring buildings and terrain remove sun hours, and the loss varies by season. A monthly average for a location describes the location, not your roof.
- Orientation and tilt. The peak sun hours assume a particular direction and angle. A roof that faces away from the equator produces less than these figures suggest.
- Roof, permits and rules. The roof's structure and free area, permits, utility approval and export limits are not in a sizing sheet.
- Losses and decline. The 14% system loss is a default, and temperature, soiling and snow change it by site and season. Output also falls slowly over the life of the panels, so year-one production overstates later years.
Where do you get site data for a solar PV estimate?
- NREL PVWatts estimates monthly and annual production for an address, with tilt, azimuth and loss inputs. Use it to check the spreadsheet result.
- NASA POWER gives daily solar irradiance and weather data by location, which can be averaged by month to give peak sun hours.
- Your utility provides the twelve months of kWh on bills or smart meter downloads, and the tariff sets the export and retail values.
How do you lay out a solar PV sizing spreadsheet?
Keep inputs and calculations apart, as described in spreadsheet model design. Put the months in rows 2 to 13, with use in column B, peak sun hours in column C and days in column D. Put the assumptions and results in column I, with labels in column H. Then:
- Annual use, I2:
=SUM(B2:B13) - Annual peak sun hours, I3:
=SUMPRODUCT(C2:C13,D2:D13) - Derate, I7:
=(1-I5)*I6, with system losses in I5 and inverter efficiency in I6 - DC size in kW, I8:
=I2*I4/(I3*I7), with the offset in I4 - Panels, I10:
=ROUNDUP(I8*1000/I9,0), with the panel rating in watts in I9 - Installed DC size in kW, I11:
=I10*I9/1000 - Monthly production, E2:
=$I$11*C2*D2*$I$7, filled down to row 13 - Annual production, I22:
=SUM(E2:E13) - Simple payback in years, I24:
=I21/I23, where I21 is net cost and I23 is annual savings
Type the monthly values and the assumptions, and let every other cell be a formula, so that changing one assumption updates the panel count, production and payback together.
The solar PV system sizing calculator is the Sheet Reserve template for this calculation. Enter your own monthly use and sun hours before using any figure from it.