The call usually comes in spring or in autumn, and it is always some version of the same sentence. The bill went up, or the app shows a number that looks wrong, and the owner wants to know whether the array is broken. Solar panels not producing what the owner expects is a symptom with at least five different causes, and only two of them are equipment failures.
Working through them in order saves a service call about half the time, and when it does not, it means the technician arrives already knowing what to look at.
Solar panels not producing: the five causes, in the order to check them
First, check that the expectation is right. Output in central Ohio swings enormously across the year. A shorter day, a lower sun angle and more cloud mean that comparing December against June tells you nothing except that it is winter. The only comparison that means anything is against the same month in a previous year, and in a system’s first twelve months that comparison does not exist yet. The design estimate is a model built on typical weather. A cloudier than usual month will come in under it, and that is the model being a model rather than the array being broken.
It is also worth separating two different complaints. A high electricity bill can be a consumption change rather than a production failure. A new heat pump, an electric vehicle, a teenager home from school, a hot summer. Check what the system produced before assuming it produced less.
Second, look at the surface. Snow is the obvious one and it clears itself. Dust, pollen and agricultural dust during a dry spell can build up meaningfully on an array with a shallow tilt, where rain runs off without washing. Bird droppings and leaf litter accumulating in a corner are more damaging than uniform dust, because a heavily shaded cell drags down more than its own share.
Third, look at shade that was not there before. Trees grow. The shade study done at installation is a snapshot, and five years later a maple that cleared the array in the afternoon may not. A neighbour’s addition, a new garage, even a taller chimney can do the same. If output has fallen gradually over several years and only in certain hours of the day, this is where to look first.
Fourth, the equipment. Covered below, because this is where the diagnostics matter.
Fifth, degradation. Modules do lose a little output over time, and it is real. It is also slow and gradual. If production fell noticeably between one month and the same month a year earlier, degradation is not the explanation, because degradation does not arrive in steps.
What the monitoring is actually telling you
The single most useful distinction on this whole subject is whether your system reports per module or only as a whole.
A string inverter system reports the string. If one module is shaded, soiled or faulty, you see a lower total without knowing which module caused it, and depending on the wiring one bad module can drag the whole string.
A system with microinverters or module level electronics reports each module separately. Now the fault is visible: one module low, all others normal, and the location is known before anybody goes on the roof. If your array has module level electronics for rapid shutdown, that data usually exists whether or not anyone ever turned the reporting on.
The habit worth building is checking the shape of a day rather than the total. A clean production curve rises, peaks and falls smoothly. A curve that is smooth until two in the afternoon and then steps down is shading. A curve that is jagged all day is cloud. A curve that stops mid morning and never restarts is a fault.
Set the alerts up. Most owners never do, and an inverter that dropped offline in April gets noticed in July.
The equipment faults worth knowing by name
Inverter failure. The inverter is the electronic heart of the system and it has a shorter service life than the modules. On a string system its failure stops everything. On a microinverter system one unit failing takes one module offline, which is much less painful and much harder to notice without monitoring.
Ground fault or arc fault trip. Solar systems have detection for both, and when either trips the system shuts down deliberately. That is the protection working. What it means is that something in the direct current wiring needs investigating, and it is not a fault to reset repeatedly until it stays on.
Connector and wiring problems. The direct current connectors on a roof sit through Ohio weather, freeze and thaw and ultraviolet exposure for decades. Corrosion, water ingress and mismatched connector brands mated together are all real failure modes, and they usually show up as intermittent output before they show up as none.
Rapid shutdown device failure. The module level electronics that de-energise the roof can themselves fail, taking their module out.
All four are solar repair work, and none of them is a rooftop errand for a homeowner.
The thing to check that is not on your roof
After a grid outage, a grid tied system has to see stable utility power for a set period before it reconnects. That is normal and automatic. What is not always automatic is recovery from a fault condition that occurred during the disturbance, and some systems need a manual restart at the inverter that nobody performed because nobody knew to.
If production stopped on the day of a storm and never resumed, that is the first thing to check. The relationship between the array and the grid is covered in our guide to grid tied solar in Ohio.
Also worth checking: whether a meter change or a billing arrangement change happened. A bill that changed shape without production changing is a billing question, not an array question.
Why this is not a job to do yourself
Between the modules and the inverter, a solar array carries direct current at voltages that are dangerous, and unlike the alternating current in your house it does not switch off when the sun is up. There is no breaker that makes a roof safe in daylight. Direct current arcs also behave differently from alternating current arcs and are harder to extinguish.
Add the roof itself, the fall risk, and the fact that walking on modules damages cells invisibly, and the sensible boundary is clear. Read the monitoring, look at the array from the ground with binoculars, check the inverter display, and stop there. Everything past that is electrical work at height.
What a good service visit includes
It should start with the monitoring history rather than with a ladder, because the data usually names the suspect. It should check the inverter for logged fault codes rather than only for its current state. It should include a visual inspection of the modules, the racking and the roof penetrations, since a leak at a mount is a roofing problem you want found early. It should check the direct current connectors and the wire management, because sagging cable that has rested on a roof for a decade is a known failure path. It should confirm the rapid shutdown function actually works. And it should end with a production comparison against the same period in previous years rather than against the design estimate.
Routine solar maintenance covers most of that on a schedule, which is how the inverter that failed in April gets found in April. If your array is producing less than it used to, start at the contact page and bring your monitoring history, or read more about the solar systems we work on.
Frequently asked questions
Why is my solar production lower this year than last year?
Work through it in order. Compare the same months rather than consecutive ones, because seasonal swing in Ohio is large. Then consider weather, since a cloudier year genuinely produces less. Then look for shading that has changed, which usually means trees that have grown. Then check the monitoring for a module or an inverter that is not reporting normally. Module degradation is real but slow, so it does not explain a drop you noticed in one year.
Do solar panels need to be cleaned in Ohio?
Usually rain does the work, and on a normally tilted roof array cleaning is rarely the answer people hope it is. It matters more on a shallow tilt, where water runs off without washing, and during a long dry spell. Localised soiling is worse than uniform dust: bird droppings or leaf litter sitting on part of a module drag it down more than a thin even layer across the whole array. If cleaning is needed it is done from the ground or by someone equipped to be on a roof, not by walking on the modules, which damages cells in ways you cannot see.
My system stopped producing after a power cut. What happened?
A grid tied system disconnects during an outage by design, so it cannot energise lines that utility crews may be working on. When power returns it waits to see stable voltage and frequency for a set period, then reconnects on its own. What sometimes does not clear by itself is a fault logged during the disturbance, and some inverters need a manual restart that nobody performed. If production stopped on the day of a storm and never came back, start there.
How do I know if one panel has failed rather than the whole system?
It depends on what your system reports. With microinverters or module level electronics you can see each module separately, so a single failure stands out immediately. With a string inverter you only see the string total, and a single bad module shows up as a lower overall number with no indication of where. If your array has module level rapid shutdown devices, per module data often exists even when nobody has enabled the reporting, which is worth asking about.
Can I troubleshoot my solar array myself?
Read the monitoring, check the inverter display for fault codes, and look at the array from the ground. Stop there. The wiring between the modules and the inverter carries direct current at dangerous voltage and it is live whenever the sun is on the array, with no breaker that makes the roof safe in daylight. Direct current arcs are also harder to extinguish than the alternating current arcs most people are used to. Add the fall risk and the fact that walking on modules damages cells invisibly, and everything above ground level belongs to a qualified technician.