Solar panels on a shaded roof: when it still works and how to price the loss

Read the shade number, not the tree

Homeowners usually describe shade by its source: the oak to the south, the neighbor's second storey, the chimney. Installers describe it with a percentage, and the percentage is what prices the system. Shade reports produced on site with a sky-imaging tool, or remotely from aerial and lidar data, usually present three related figures.

Solar access is the share of the sunlight that would reach a point on the roof that still reaches it after nearby obstructions are counted. A point with 90 percent solar access loses a tenth of its possible sunshine to shadows over the year. The tilt and orientation factor (TOF) describes how much of the ideal resource the roof plane's own angle and direction capture, before any shade. The total solar resource fraction (TSRF) is the share of the ideal solar resource that the location receives once both shading and the plane's tilt and orientation are counted. Energy Trust of Oregon's guide to sky-imaging reports shows all three on the same page and asks installers to copy the TSRF onto incentive applications.

Two details in that program's rules are worth borrowing. First, Energy Trust asks for the skyline with the lowest TSRF on the array, not the average, and uses that value to estimate annual production. An average hides the worst corner, and the worst corner is where the electrical damage starts, as a later section explains. Second, its thresholds differ by method. Its 2023 solar electric system requirements set a TSRF of 75 percent or greater at all points on the array for string inverters when shade is measured on site, allow individual modules below 75 percent only when they are electrically isolated with microinverters or other approved technology (those modules then earn no incentive), and require 80 percent or greater on each plane when the analysis is remote.

Where shade hides inside a production estimate

A shade report and a production estimate are separate documents, and the link between them is one input. In NREL's PVWatts calculator, now served at pvwatts.nlr.gov, shading is one of ten categories in the system losses calculator, defined as the reduction in incident solar radiation from shadows caused by objects near the array, such as buildings or trees. The current PVWatts help text says the default shading value of 1 percent represents an array with little or no shading, and that shading analysis tools can determine a loss percentage for shading by nearby objects. The calculator's default total system loss is 14 percent.

The older PVWatts Version 5 manual, published in 2014, used a 3 percent default for shading inside the same 14 percent total and explained why: surveys of installed systems described as unshaded still averaged roughly 3 percent shade loss. It adds that for a system shaded by nearby trees or structures, the value should be increased using external shading software or an on-site survey. The manual also notes that the total is not the sum of the parts; each loss multiplies the energy left after the others.

The practical test follows directly. If your roof has visible shade and the quote's production model shows default or near-default losses, the estimate is describing a different roof. Ask for the loss table behind the estimate and for the shade report it came from. The production estimate guide walks through the rest of the model inputs; for shade, the one question is whether the measured loss made it into the number.

Turning a shade percentage into a price

Shade does not change what the system costs to install. It changes how many kilowatt-hours that cost buys, so the cleanest way to price it is cost per first-year kilowatt-hour: the gross price divided by expected first-year production. The table below holds the price at this site's benchmark comparison line, $2.95 per watt DC, or $2,950 per kW (the U.S. Department of Energy's Q1 2025 residential benchmark, 2024 dollars, checked October 30, 2026), and varies two things: the unshaded production factor for the area and the annual shade loss on the array.

Read across a row and the shape is clear: the penalty grows faster than the shade percentage, because the price is divided by what is left. A 25 percent loss raises cost per kilowatt-hour by a third, not a quarter. Read down a column and another pattern appears: a sunny site with 20 percent shade can still produce cheaper energy than an unshaded roof in a cloudier climate. At 1,600 kWh per kW and 20 percent shade the figure is $2.30, below the $2.46 for an unshaded roof at 1,200.

The same arithmetic converts directly to payback. Simple payback is the price divided by the yearly value of the energy, so shade lengthens it in exactly the same proportion it raises cost per kilowatt-hour. An 8 kW system at $23,600 producing 1,400 kWh per kW, with every kilowatt-hour valued at an illustrative 20 cents, makes 11,200 kWh and pays back in about 10.5 years. The same system with 19 percent shade makes 9,072 kWh and takes about 13.0 years. Whether that is acceptable depends on what a kilowatt-hour is worth at your meter, which the 2026 break-even test works through with no federal credit in the math.

Why a small shadow can cost more than its size

If shade only removed the light that falls on the shaded cells, a shadow covering 5 percent of the array would cost about 5 percent of its output. On many systems it costs more, because of how panels are wired. Cells inside a panel are connected in series, and panels on a conventional string inverter are connected in series too. Current through a series chain is limited by its weakest member, so one shaded cell can drag down the cells around it.

Panels limit that damage with bypass diodes. NREL's 2012 shading testbed report notes that typical 60-cell modules include three bypass diodes in the junction box, each protecting 20 cells in series, a unit the report calls a submodule. When shade on a cell is heavy enough, its diode begins conducting and the current routes around that third of the panel, sacrificing its output to protect the rest. The report observes that in a system where one string is shaded and another is not, as little as 25 percent opacity shade on a single cell could be enough to start that module's bypass diode conducting. A leaf-sized shadow can therefore switch off a third of a panel.

At the string level the effect compounds. A string inverter looks for the best operating point for the whole string; shaded modules with diodes conducting pull the string's voltage away from what the unshaded modules want, and the inverter settles on a compromise. That is why shade on two or three panels in a long string can cost more than the same shade spread thinly across the array, and why the Energy Trust rules apply the TSRF threshold at every point on a string-inverter array rather than on average.

What panel-level electronics recover, and what they cannot

Microinverters and panel-level DC converters, together called module-level power electronics, give each panel its own operating point, so a shaded panel no longer sets the terms for its neighbors. NREL's testbed measured how much that is worth. Comparing a microinverter system with a string inverter on identical 8 kW arrays under shade patterns weighted to match surveyed homes, it found the microinverter increased production by 3.7 percent under light shading (a site with about 7 percent annual shade), 7.8 percent under moderate shading (about 19 percent) and 12.3 percent under heavy shading (about 25 percent). The report's conclusion summarizes each score as a recovery of around half of the overall loss due to shade.

Around half is the important phrase. Panel-level electronics recover the mismatch loss, the energy that unshaded panels lost because a shaded neighbor dragged the string down. They cannot recover the sunlight that the shadow actually blocked, and they cannot help cells inside a shaded panel that a bypass diode has already switched off. A panel in deep shade under a microinverter still produces very little; it simply stops costing the panels next to it. The test was a single side-by-side comparison with 2012 equipment, so treat its percentages as an order of magnitude rather than a forecast for current hardware.

Electronics also cost money and add parts on the roof. Whether that trade pays depends on roof complexity, service access and replacement plans as much as on shade, which is why the choice belongs with the inverter decision. EnergyBS compares the two architectures in its microinverters versus string inverters guide, and the inverter replacement guide on this site prices the inverter event most systems meet once.

If the electronics add more to the price than the recovered energy is worth over the years you will own the system, the string design still wins on cost. Ask each installer to price both designs on the same roof and to show the production estimate for each, then compare cost per first-year kilowatt-hour rather than headline output.

Design moves that beat equipment

The cheapest shade fix is usually to stop putting panels in it. Fixed system costs such as permitting, the inverter and the electrical work do not shrink in proportion when you remove a few panels, so the price per watt of a smaller system rises slightly. But a panel in heavy shade produces so little that removing it often lowers the cost of the energy you keep.

Pricing the shaded plane separately

Inputs for an illustration: an 8 kW design at $23,600. Six kilowatts sit on a south plane with 3 percent shade loss and 2 kW on a west plane with 30 percent loss, all at 1,400 kWh per kW before shade. The south plane makes 8,148 kWh and the west plane 1,960 kWh, 10,108 kWh in total, or $2.33 per first-year kWh. The installer quotes the 6 kW south-only version at $18,600, a slightly higher $3.10 per watt because the fixed costs are spread over fewer watts.

The south-only system costs $2.28 per first-year kWh. The west plane, priced as the $5,000 difference, costs $2.55 per kilowatt-hour it adds. With energy valued at an illustrative 20 cents, the south plane alone pays back in about 11.4 years and the west panels as an increment take about 12.8. Whether to keep the west panels is now a clear question: is energy at $2.55 per first-year kWh worth buying? Sometimes it is, if your usage needs it and exports are paid well. The point is that the shaded plane gets its own price instead of hiding inside the average.

Other moves before buying electronics

• Shift the layout. Moving rows up or along a plane, away from a chimney or vent stack, can take modules out of the worst shadow without losing capacity.
• Use a different plane. An east or west plane in open sky can produce more than a south plane under trees. Compare measured TSRF by plane, not compass direction alone.
• Size down. A smaller array on the good plane often beats a larger one spread over poor ones; the sizing guide explains why export rules reward this.
• Look at the ground. On a property with open land, a ground mount can be placed where the trees do not reach; the ground versus roof comparison prices the extra site work.
• Consider the trees last, not first. Removing or trimming a tree changes shade, cooling and sometimes local permit obligations. It deserves its own arithmetic: removal cost against the production gained, not a promise from the design software.

Shade that changes after installation

A shade report measures the site on the day it was taken. Trees grow, neighbors build, and the sun's path shifts through the year, so the same roof can lose more in winter, when the sun sits low and shadows run long, than in summer. A shade report with monthly solar access figures shows that pattern; a single annual percentage hides it. If most of the shade falls in winter, its effect on annual production may be smaller than the shadows suggest, because winter days produce less energy anyway.

Growth is the harder risk to price. A young tree at the edge of the shadow line today may cross it within the life of the panels, and nothing in the current report captures that. Ask the installer whether the design accounts for growth, and where a tree on your own property is the problem, decide now whether it stays. A tree on a neighbor's land is a different matter. Some states have solar access or easement laws, and the U.S. Department of Energy's homeowner guide notes that heavy tree shade can make rooftop panels a poor fit, pointing those homeowners toward alternatives such as community solar. DSIRE lists solar access policies by state; the HOA guide explains how to read those laws and find the statute that applies to you.

Shade also changes how you judge performance later. A system that misses its estimate in its first winter may be performing as designed if the model used monthly shade data, or underperforming if the model assumed default losses. Keep the shade report with the system file, beside the production estimate, so a later comparison has something to compare against.

Reading the shade pages of a quote

Most of the work is asking for documents that already exist. A design team that measured shade has the files; one that did not will have to say so.

1. Ask for the shade report itself, on site or remote, with the measurement points marked on a roof plan.
2. Find the lowest TSRF or solar access value on the array, and the modules it applies to. Compare it with an outside reference such as Energy Trust's 75 percent on-site and 80 percent remote thresholds.
3. Find the shade or system loss figure in the production estimate and check that it reflects the report rather than a default.
4. Divide the gross price by the first-year production to get cost per first-year kWh, and check it against the table above for your production factor.
5. If the design uses microinverters or DC converters at each panel to justify shaded modules, ask for a priced version without the worst modules and compare the two on cost per first-year kWh.
6. Put the result into the payback estimator with the shaded production figure, not the brochure figure.