Solar proposals present system size as a conclusion: 8.4 kW, 22 panels, done. It is actually the result of a four-input calculation you can run on a napkin, and running it yourself is the fastest way to spot a system that is oversized to inflate the contract or undersized to hit a payment target.
The calculation
System size in kilowatts equals your annual kilowatt-hours, divided by the annual production per kilowatt of installed capacity in your location. That second number is what varies, and it comes from three things: sun hours, orientation and tilt, and system losses.
In Central Texas, a well-oriented rooftop array typically produces somewhere in the range of 1,300 to 1,500 kWh per year for each kilowatt of installed capacity, after real-world losses. A shaded or poorly oriented array produces less. That range is the single assumption most worth interrogating in any proposal.
So a home using 15,000 kWh a year, at 1,400 kWh per kW, needs roughly 10.7 kW of panels to offset all of it. At 400 watt panels, that is about 27 panels. If a proposal for that home shows 34 panels, ask what production estimate it is using and why it is lower than the regional norm. Shade is a legitimate answer. Silence is not.
Where the losses come from
Panels are rated at standard test conditions that do not exist on a roof in August. The gap between nameplate and real output is normal and predictable:
- Temperature. Panel output falls as cell temperature rises. On a Texas rooftop in summer this is the largest single loss, commonly in the range of 8 to 12 percent annualized.
- Inverter efficiency. Converting DC to AC costs a few percent.
- Soiling. Dust and pollen, typically 2 percent, more if you are near agriculture or construction.
- Wiring and mismatch. A few percent more.
- Shading. Highly site specific and the one that can be enormous.
Together these usually total 14 to 20 percent. If a proposal shows a derate meaningfully better than that, it is optimistic and the production numbers downstream of it are too.
Orientation and tilt
South-facing is best for total annual production. West-facing produces less overall but produces it later in the day, which matters more than it used to under rate plans with time-varying pricing and matters a great deal if you are trying to cover a late-afternoon air conditioning peak.
East-facing produces a morning-weighted curve. North-facing on a pitched roof in Texas is generally not worth installing unless the roof is nearly flat. A proposal that fills a north face to reach a bigger contract number is a proposal to look at carefully.
Tilt matters less than people assume. Between roughly 15 and 40 degrees the annual difference is small. Do not pay for a tilt-mounting structure on a pitched roof to chase it.
Shade is measured, not guessed
Ask whether the installer performed a shade analysis and ask to see the output. Tools that model the sun path against a photograph or a lidar model of your roof produce a per-panel annual solar access percentage. That is the document you want.
Also ask about trees that will be larger in five years. A pecan that clips the array edge in the afternoon today will clip more of it later, and a shade study of today's canopy quietly assumes the tree stops growing.
Choose the offset deliberately
Most proposals target 100 percent of annual usage. That is a reasonable default, but it is a choice, not a law:
- Under 100 percent makes sense when exported energy is credited well below what you pay for imports, because every exported kilowatt-hour is worth less than the one it displaced. Sizing to cover daytime consumption rather than annual total can produce a better return.
- Over 100 percent makes sense when you expect load to grow: an electric vehicle, a heat pump replacing gas heat, an addition, a pool.
Some utilities and some interconnection agreements limit system size relative to historical usage. Ask whether such a cap applies before signing for an oversized array.
Read the production guarantee, not the estimate
Proposals present a year-one production estimate and a 25-year cumulative total with an annual degradation assumption, usually around 0.5 percent a year. Both are projections.
The question worth asking is whether there is a production guarantee, and if so, what happens when actual production falls short. A real guarantee names a percentage of the estimate, a measurement method, and a remedy, typically a cash payment for the shortfall. A guarantee with no remedy is marketing.
Two sanity checks before you sign
First, divide the proposal's year-one production estimate by the system size in kW. If that number is far above the regional range, the estimate is aggressive. Second, check the panel count against the roof plan and confirm every panel is on a face you agreed to. Both take two minutes and both have caught real problems.