Does Cleaning Solar Panels Actually Increase Output?

A solar cell only works on the light that reaches it. Everything on the glass either absorbs that light or scatters it, and a film of dust does both across the whole face of the module at once. That is the entire mechanism. There is no argument in the industry about whether soiling reduces output, only about how much. So the honest version of the answer starts with a yes and then gets careful, because the size of the gain is where most of the marketing goes wrong.

Published soiling research in dry climates tends to land in a similar place. Averaged across a full year the loss looks modest, often just a few percent, because winter rain resets it. That annual average hides the shape of the curve. A panel in February after a week of storms is close to clean. The same panel in late September, five months into a rainless stretch, is carrying everything that has landed on it since spring. The end of the dry season is where the real number lives, and it is commonly put somewhere around five to ten percent.

What it is not is a fixed figure you can be quoted over the phone. Tilt changes it, because a steep panel sheds dust and a nearly flat one holds it. Orientation changes it. So does what sits upwind, how long it has been, whether birds have a perch above the array, and whether your dirt is valley dust or coastal salt. If somebody guarantees you a specific percentage before looking at the array, they are quoting the worst case in the literature and hoping it is yours.

The rain calendar is the whole problem. From roughly May through October, San Luis Obispo County gets close to nothing in the way of measurable rain. That is five or six months of accumulation with nothing to interrupt it. Plenty of the country never sees a soiling problem worth naming, because a thunderstorm every ten days keeps the glass roughly honest. Here the dust just keeps stacking, a gram at a time, so the array never looks bad enough on any single day to make you do something about it.

Then there is what is in the air. In the Santa Maria valley and out toward Oceano, disking and harvest lift very fine soil that travels a long way and settles into an even, stubborn film. Around Paso Robles, Templeton and Creston the same happens with vineyard rows and harvest traffic on unpaved lanes, and it peaks in late summer, exactly when your array is already at its dirtiest point of the year. An array beside an active dust source is a different case from one on a suburban street.

The coast gets a slower version with the same ending. The marine layer through Cambria, Cayucos, Morro Bay, Los Osos and Avila Beach does not deliver enough water to rinse anything, but it keeps the dust damp, and salt in the air cements it in place. Coastal arrays often look cleaner than inland ones from the driveway and are not. Different mechanism, same result on the meter.

This is the part most owners have backwards. Rain wets dust and moves it. It does not take it off the property. A light shower on a panel carrying five months of dry season buildup makes mud, that mud runs down the face of the module, and the sun dries what is left into a hard rim along the bottom edge of the frame. The frame lip is what holds it there. Plenty of arrays read lower on the week after a light rain than they did the week before it.

That rim sits right across the lowest row of cells, and cells in a panel are wired in series, so the weakest row sets what the rest can pass. A stripe of dried mud along the bottom inch costs more than the same amount of dirt spread evenly over the module would. It is also the last thing to come off, because it has been baked in place rather than laid down loose.

The honest counterweight: a proper multi day soaking storm does genuinely help. A January system in San Luis Obispo County that has been rained on for a week is in reasonable shape, and that is why cleaning in the middle of winter is usually the least useful time to do it. The wash that matters is the one before the dry season and the one after harvest dust has settled.

Dust spreads its loss thinly and evenly. A bird dropping is opaque, and it lands on one cell. Because the cells are in series, that string can only carry as much current as its worst cell allows, so a blocked cell drags on everything wired with it. This is why an array that looks basically clean can still be down noticeably, and why a handful of spots can cost more production than a whole summer of dust.

Panels are built with bypass diodes for exactly this reason. A module is usually split into two or three sections, each with a diode that routes current around a section that has gone bad. That protects the rest of the array by taking the bad section out of service. So one opaque spot the size of a quarter can cost you a third of that panel. Microinverters and optimizers keep the damage contained to the affected module rather than the whole string, which helps, but they do not hand the missing output back.

There is a durability angle too. A cell that is blocked but still carrying the current the others are producing gets driven backwards, and it turns that power into heat instead of electricity. Concentrate that in one small area for long enough and you get hot spotting, which can discolor the encapsulant or crack the cell underneath. That damage does not clean off. It is the reason a droppings problem is worth dealing with sooner than a dust problem, even though the dust is what you can see from the ground.

Purified water is water with the minerals filtered and stripped out of it, so when it evaporates off the glass it leaves nothing behind. That is the whole point of it. Tap water and inland well water around Paso Robles, Creston and Templeton is the opposite: it dries into calcium spots that sit on the glass and shade the cells, so you have traded a dust problem for a mineral one and made the array harder to clean next time.

Soap is the mistake we see most often, and it is an understandable one. A surfactant does lift dirt, but it also leaves a very slight film as it dries, and that film is just tacky enough to hold the next round of dust. The array gets dirty faster than it would have if you had left it alone. On the Central Coast, where the next round of dust is guaranteed and the next real rinse might be four months out, that is a bad trade.

Nothing abrasive touches panel glass either. There is an anti reflective coating on the front surface, it is thin, and scratching it costs you output permanently to save ten minutes on a stubborn spot. Baked droppings get soaked and softened and lifted over several passes instead. Nobody should stand on a module, ever, because the cells crack out of sight under a boot and turn up months later as dead spots in your production. We work arrays with water fed poles, ladders, or a lift when the pitch or the height calls for one.

Here is the part the industry does not push, because it lets you check the claim. Almost every system installed in the last decade came with production monitoring, either built into the inverter or sitting in an app on your phone. That data is the only honest source for what soiling is costing you specifically, and it is free. You do not have to take a number from a website, ours included.

Read it like with like or it will lie to you. Do not compare July against December. Compare a clear day this September against a clear day last September, or run the simple version: pick a cloudless day, note the peak output around solar noon, get the array cleaned, then check the next cloudless day. Use a few days on each side rather than one, because heat matters too. Panels lose a little output when they are hot, so a 100 degree afternoon in Paso Robles reads lower than a mild one at the same sun angle.

What you get out of that is a number that belongs to your roof instead of somebody's brochure. If your array comes back two percent, twice a year cleaning probably does not pay for itself and you should know that. If it comes back eight or nine, the math works easily and you have proof rather than a promise. Either way you are deciding on your own data, which is the position we would rather have a customer in.