The short version
- A dark monitoring app does not always mean a dark system — check your electric bill before you assume the panels quit.
- A string system can lose an entire string to one bad connection; micro-inverters and optimizers usually confine a failure to the panel it happened on.
- Arc fault and ground fault codes are stop signs — shut the system down and call rather than clearing them repeatedly.
- Critter damage under the array is a common cause of one dead panel, and repairing the wire without closing the gap invites a repeat.
- Compare production to the same month last year, not to last month — Pennsylvania winters cut output all on their own.
First question: are your solar panels not producing, or just not reporting?
Solar panels not producing is a symptom, not a diagnosis, and the first fork in the road is whether you have a power problem or a reporting problem. Your monitoring platform depends on a small communication device — an Enphase IQ Gateway, the Wi-Fi radio built into a SolarEdge inverter, or a cellular modem — that has to reach your home router. Change the Wi-Fi password, replace the router, or switch internet providers, and the data stops while the panels keep right on working.
Check something outside the app. On a clear day around noon, walk out to the inverter and look at it. A running inverter almost always shows a green status light. Some models also display current output on a screen, though many current inverters have dropped the screen in favor of LEDs and a phone app — and on a micro-inverter system there is no central inverter at all, just the gateway. Then look at your electric bill. If your usage charges jumped in the same month the app went dark, you have a real production problem. If the bill looks normal, you probably have a communication problem: annoying, but not urgent. We walk through those solar monitoring problems separately.
Check your breakers and disconnects before you call anyone
A grid-tied solar system can be switched off in three or four different places, and any one of them stops production. There is a solar breaker in your main service panel, usually labeled PV or Solar. There is an AC disconnect, often a small gray box near your meter, required so utility crews can isolate your system. And there is a DC disconnect at or on the inverter. A thunderstorm or a brief utility outage can leave any of them open.
Look for a breaker that is not fully in the ON position; tripped breakers often sit in a mushy middle spot. You can reset one, once. If it trips again right away, stop and call: repeated tripping means something downstream is faulted, and forcing it is how small problems become expensive ones. Restart sequences differ by manufacturer, so follow your inverter's manual instead of guessing. And never open equipment or touch DC wiring: a solar array is live whenever sun is on it, even with every switch off.
Reading your inverter: status lights and the fault codes you should never reset
Colors follow a rough convention across brands: green means running, amber or orange means partially working or not communicating, red means stopped. The details matter, though. On an Enphase gateway, solid green on the network light means it is talking to the cloud, while solid orange generally means it reached your home network but has no internet — a monitoring symptom, not a production one. Confirm the specific meaning in your manual before you draw conclusions.
Whatever the screen or the LEDs say, photograph it. Fault codes are often transient, and a picture of the exact code and the time it appeared saves an hour of guesswork later. If you see arc fault or ground fault language, treat it as a stop sign. Those two codes exist specifically because DC arcing and insulation breakdown are fire risks, and clearing them over and over hides a hazard instead of fixing it.
- Grid fault, or AC voltage or frequency out of range: the inverter is protecting itself and often restarts on its own.
- Arc fault (AFCI): the inverter detected arcing on DC wiring. Shut the system down and call a professional.
- Isolation or ground fault: insulation has broken down somewhere. This needs a technician with a meter, not a reset.
- Over-temperature: common on hot wall mounts and in tight, unventilated spaces; usually an airflow or mounting-location problem, though on models that have a cooling fan it can be a failing fan.
- No DC input at dawn and dusk is normal — inverters wake up and shut down with the available light.
One dead panel or the whole array? Your system's design decides
Systems fall into two broad families, and which one you have decides what a single failure costs you. A string system wires panels in series into one central inverter. Modern panels carry bypass diodes, so heavy shade on one module usually costs that module's output plus a little mismatch loss, not the whole string. What does take a string down is an open circuit: a corroded connector, a pinched conductor, a failed junction box. If you have one box on a wall and nothing but panels on the roof, you probably have a string system, and one bad connection can zero out everything behind it.
Module-level electronics behave differently. Micro-inverters (Enphase, APsystems) convert DC to AC right under each panel, and optimizers (SolarEdge) condition each panel before feeding one central inverter. When a single unit fails you lose that panel's production, and the app can usually tell you exactly which one. Two catches: with optimizers, the central inverter is still a single point of failure, and rooftop electronics live in heat, water and squirrel country, which is hard duty for any electronics. If your app is flagging one dark panel, our walkthrough on a failed micro-inverter covers what replacement involves.
A roof replacement can also quietly kill production. If the crew that removed and reinstalled your array was not trained on solar, mismatched connectors, unseated plugs, or a pinched conductor can leave a string dead or a handful of panels dark — sometimes months later, when a connector finally corrodes through.
Critters, storms, shade and snow: what is happening under the array
The space between your panels and your roof is warm, dry, sheltered, and full of chewable wire, which makes it excellent habitat. Squirrels and chipmunks are the usual culprits across Central Pennsylvania; pigeons and starlings nest and pack the cavity with debris. Most of this damage is invisible from the ground, so a production drop with no obvious cause usually has to be diagnosed from a ladder.
Two practical notes. First, do not climb up to clear snow. Panels are slick, and most Pennsylvania snow slides off within a few days as the glass warms. Persistent soiling in a climate with our rainfall typically costs only a few percent a year, so if production fell off a cliff, dirt is not your answer. Second, if critters caused the damage, repairing the wire without closing the gap just buys you another repair later. Critter guard mesh around the array perimeter closes it, and installing it is one of the services we offer. Whether a rodent chew is covered is a question for your own policy paperwork.
- Chewed conductors: one dead panel, an arc fault, or production that comes and goes on windy days.
- Nesting debris: blocks airflow under the panels and packs twigs and droppings into racking and gutters.
- Storm and hail damage: cracked glass, delamination, or a bent rail after a squall line moves through.
- Shade that grew in: the maple that was ornamental in 2014 now clips your morning production.
- Snow and soiling: pollen, farm dust and snow all cost output, though rain usually handles the dirt here.
When low solar production is a utility problem, not a roof problem
Grid-tied inverters are required to shut down when the grid goes down. That is anti-islanding protection, and it exists so your system cannot backfeed a line a utility crew is working on. Without backup equipment, your solar goes dark during an outage even at noon. Adding a battery and a backup panel is the usual fix; some newer grid-forming micro-inverters can also run priority loads off daytime production with a system controller and no battery, though that only works while the sun is on the roof.
Inverters also disconnect when grid voltage drifts too high. The interconnection standards they are built to allow only a limited band above nominal voltage, so a home at the end of a long rural feeder can trip offline midday and recover by evening. If your production data shows a sawtooth — strong mornings, repeated midday drop-offs, recovery later — high line voltage is a genuine suspect, and it is worth reporting to PPL, Met-Ed, PECO, or whichever utility serves your address. And if a newly installed system has never produced at all, the problem may be paperwork: utilities issue Permission to Operate only after inspection, and a finished system can sit unauthorized for weeks.
Normal seasonal decline versus an actual fault
A Pennsylvania system makes dramatically less power in December than in June: shorter days, a lower sun angle, and far more cloud cover. It is easy to compare this month to last month, see the cliff, and panic. Compare the same month year over year instead, which takes the season out of it. Panels themselves lose roughly half a percent a year, and whole systems measure a bit worse once inverter drift and soiling are counted, so a healthy ten-year-old system typically still makes somewhere in the low-to-mid nineties percent of what it did new. Treat anything in that range as normal aging, not a fault.
Real faults look different from slow decline. They show up as a step change, with production normal on Tuesday and half of that on Wednesday. Or as a flat zero on one panel while its neighbors read normally. Or as daily output that ends abruptly at one in the afternoon. The shape of the curve matters more than the number, so pull a week of data from your app if you can.
What to write down before you call about low solar production
A Solar Triage Call goes much faster when you have gathered a few things first. None of it requires a ladder or a meter — it is on a label, a screen, or in an old email.
With that in hand, a phone conversation can often narrow the cause before anyone drives to your house. Sometimes that is the whole fix: a gateway that lost the Wi-Fi password does not need a house call. When it does need one, we handle solar panel repair and diagnostics on any brand and any installer's work, including systems whose original company is long gone.
- The date production last looked normal, and whether the drop was sudden or gradual.
- A photo of the inverter screen or status lights, including any fault code and when it appeared.
- Inverter and panel brand and model, copied off the label on the equipment.
- Who installed it, roughly when, and whether you own the system, lease it, or have a PPA.
- Anything that changed recently: roof work, tree growth, a new router, a storm, an electrical panel upgrade.
Not sure what your system is doing?Start with a free Solar Triage Call — a phone consult with a repair tech, no truck roll required.
Book Your Triage Call