Why Is One Solar Panel Producing Less Than the Others? Panel-by-Panel Data From 269 LA Roofs
Panels on the same roof rarely match exactly. Across 5,864 panels with their own microinverters on 269 Los Angeles-area systems, 80% produced between 88% and 108% of their roof’s median panel over their lifetime. 12% of panels ran more than 10% below their neighbors, 5% more than 20% below, and 1% at under half. A gap of 10–20% is ordinary — usually a second roof face or partial shade. A panel at half its neighbors is worth a check.
On a system with microinverters, the monitoring app shows every panel separately, so one darker square on the array map is easy to spot and easy to worry about. Set against the median panel of its own roof over the energy its microinverter has recorded since the system was switched on — years of data per panel — a passing cloud or a single bad day does not move the comparison.
How far one panel strays from its neighbors
5,864 panels, lifetime energy as a share of the same roof’s median panel.
| Panel output, share of the roof median | Panels | Share of panels |
|---|---|---|
| under 50% | 57 | 1.0% |
| 50–70% | 85 | 1.4% |
| 70–80% | 168 | 2.9% |
| 80–90% | 388 | 6.6% |
| 90–95% | 485 | 8.3% |
| 95–100% | 1,684 | 28.7% |
| 100–105% | 2,081 | 35.5% |
| 105–110% | 428 | 7.3% |
| 110% or more | 488 | 8.3% |
https://cali-energy.com/research/one-solar-panel-producing-less-los-angeles. Aggregated data are free to use under CC BY 4.0 with a link to this page.The spread leans downward. 20% of panels were more than 5% below their roof’s median and 12% more than 10% below, typically on the shadier or worse-facing part of the array. Fewer ran ahead: 16% were more than 5% above the median, usually a smaller group of panels on a better-facing part of the roof than most of the array. Panels that age at slightly different rates can add a few percent either way.
Three patterns on a roof
| Pattern | Roofs | Share of roofs | What it usually means |
|---|---|---|---|
| Even: no panel below 90% | 116 | 43% | Single roof face, little shade. Small differences come from the panels and their exact spot on the roof. |
| Tail: 1–2 panels below 90% | 68 | 25% | A vent, chimney, tree branch or wire shading one spot; less often a weak panel or microinverter. |
| Group: 3 or more below 90% | 85 | 32% | Panels on a second roof face, or a shadow from a tree or neighboring house crossing part of the array. |
Measured roof by roof, the gap between a system’s 90th- and 10th-percentile panels was a median 13 points; a quarter of roofs spread 22 points or more. Different roof faces make different amounts of energy over a year; see east-west versus south-facing and shade and trees.
Truly low panels are uncommon. 20 of 269 roofs had at least one panel at under half of the median over its lifetime. That happens when a panel sits in deep shade, when a microinverter has been offline for a long stretch without being noticed, or when a panel or microinverter is failing — and only a look at the daily data and the roof tells which.
Same spread on every microinverter model
| Microinverter | Panels | p10 | Median | p90 |
|---|---|---|---|---|
| Enphase IQ8PLUS | 1,868 | 88% | 100% | 108% |
| Enphase IQ7+ | 1,461 | 90% | 100% | 107% |
| Enphase IQ8M | 1,036 | 89% | 100% | 107% |
| Enphase IQ8A | 477 | 92% | 100% | 107% |
| Enphase IQ7 | 295 | 87% | 100% | 111% |
| Enphase IQ8X | 201 | 90% | 100% | 106% |
Older IQ7 and newer IQ8 microinverters showed nearly the same spread, which is another sign that most of the difference between panels comes from the roof and the sun, not from the electronics. How much one panel makes in a day: output of one solar panel.
Check your own panels
Copy each panel’s energy for the same period from your app — a week, a month or lifetime.
Compare panels over days, not moments. A single live reading from each microinverter can differ widely between neighbors on the same roof and match poorly with the panel’s long-run energy; a passing cloud or the shadow of a chimney can reach panels one by one. A week of energy is enough to see a real pattern; a month is better. A panel that is low every day, all day, on the same roof face as its neighbors is the one to report. What the app shows and how to read it: solar monitoring apps; a step-by-step check: solar not producing.
Methodology
- Panels
- Every Enphase microinverter in our monitoring fleet that was active in October 2026, grouped by system: 5,864 panels on 269 residential systems with at least 8 original microinverters. 38 systems with fewer and 2 commercial systems are left out.
- Measure
- Each microinverter’s lifetime energy divided by the median of the original microinverters on the same system. Lifetime energy covers years of operation, including any stretches when that microinverter was offline.
- Original panels
- Microinverters whose warranty start date is within 45 days of the system’s most common date. Replacements and panels added later (136 microinverters) have recorded energy for a shorter time and are left out.
- Patterns
- Even: no panel below 90% of the median. Tail: one or two below 90%. Group: three or more below 90%.
- Not used
- Momentary power readings. A single live reading varies widely between neighbors on the same roof and agrees poorly with lifetime energy (correlation about 0.2), so it says little about a panel.
- Limits
- Roof faces, tilt and shade are not recorded per panel, so the patterns are not a diagnosis. All systems are Enphase.
Frequently asked questions
Is it normal for one solar panel to produce less than the others?
Yes, within limits. In the 5,864 panels we measured, 80% produced between 88% and 108% of their roof’s median panel, and 12% ran more than 10% below it, usually because of shade or a different roof face.
How much lower is too low for one panel?
Below 80% of its neighbors over weeks puts a panel in the lowest 5% we measured; below 50% in the lowest 1%. Either is worth a look, especially if the panel faces the same way as its neighbors.
Why are several panels on my roof lower than the rest?
A group of lagging panels usually means a second roof face or a shadow crossing part of the array. 85 of 269 roofs we measured showed that pattern.
Can I judge a panel from the live power reading in the app?
Not reliably. In our data, single live readings from neighboring panels differed widely and matched poorly with each panel’s long-run energy. Compare the energy each panel made over a week or a month.
Does the microinverter model change how panels compare?
Hardly. Enphase IQ7 and IQ8 microinverters showed nearly the same panel-to-panel spread in our data.
More research
Next steps
A panel that stays low?
Prepared by Cali Energy, October 15, 2026. Figures describe systems in our monitoring fleet and are not a guarantee of what any specific roof will produce; orientation, tilt, shade, equipment and weather change results. Not engineering, financial or utility advice. See our Content Disclaimer.