Do solar panels need cleaning? Usually not, and when it pays

Usually not. On a tilted roof, ordinary rain does the cleaning: REC and Canadian Solar both say so. The research figure for dirt, 3–5% of a year’s output, comes from large solar plants and is an upper bound for a rained-on roof. Cleaning becomes worth doing when you can see dirt on the glass, or when something on the roof (a bird colony, an overhanging tree, sea salt, city traffic film) is holding a layer of dirt that rain cannot shift. The way to tell “grimy-looking” from “actually losing money” is to compare your own daily kWh against what the weather allowed, not to look at the glass.

Do dirty solar panels lose output?

Dirt costs a tilted, rained-on roof little; the research figure of 3–5% of a year’s output comes from large solar plants. PVGIS’s default 14% system loss already includes some dirt, so a shortfall against a PVGIS estimate cannot simply be put down to dirt.

Are solar panels self-cleaning? What rain washes off, and what it does not

Sandia puts the split plainly: “some types of soiling can be cleaned naturally by precipitation while others require manual washing to remove”.

The manufacturers say the same thing from the other side. REC:

“REC panels are manufactured so that normal and regular levels of rainfall will clean them naturally if installed at a sufficient angle.”

Canadian Solar puts a number on “sufficient angle”: modules with at least 15° tilt “generally may not require cleaning (rain will have a self-cleaning effect)”. Canadian Solar also names what regular maintenance is meant to keep off the glass: “snow, bird droppings, seeds, pollen, leaves, branches, dirt spots, and dust.”

That list is the practical answer to the question. General atmospheric grime and dust go with the rain on a tilted roof. What stays behind is whatever sticks or lands in bulk: droppings under a perch or a nesting spot, a pollen film in spring under a birch or an oak, leaf litter and moss creeping up from the lower edge where water sits, salt near the coast, traffic film in a city. Which? names the same triggers for a deeper clean: birds roosting and soiling the panels, trees dropping leaves or pollen, salt near the sea, and inner-city traffic pollution creating a film.

On green growth, REC says that on the rear of a panel, “growth such as moss, mold or lichens” is “not generally a cause for concern”, though if you do clean the rear, take care not to damage the backsheet.

Snow is a different question from dirt, and REC answers it: “as snowfall generally occurs at times of the year when irradiation is at its lowest, clearing snow build-up from the panels is not essential for maximum performance.”

Solar panel cleaning cost: does a clean actually pay?

Whether a clean is worth paying for usually settles itself once you do the arithmetic.

Which?, 12 August 2026, puts cleaning kits with an extendable pole at typically around £100. It says professional cleaning is available through installers, and that it is worth speaking to a couple of companies and getting quotes so you know the going rate in your area. So use the quote you are given as the cost side of the sum.

Weigh an upper-bound loss of that size against a £100 kit or a quoted visit. A clean pays when your roof is one of the ones holding dirt that rain cannot shift, and you find that out from your production history, as in the last section, not from the look of the glass.

How to clean solar panels without damaging them

Both manuals agree, and both put safety first.

Before anything. REC: “Damaged panels can present a lethal shock hazard due to leakage currents. The risk of shock is higher when panels are wet. Before cleaning, thoroughly inspect panels for cracks, damage, and loose connections.” Canadian Solar says the same and adds that “The voltage and current present in an array during daylight hours are sufficient to cause a lethal electrical shock.”

Time of day. REC: “Panel cleaning should be carried out before reaching working temperature i.e., early morning and must only be cleaned with water at ambient temperature to avoid thermal shock.” Cold water on hot glass is the specific danger: “Applying cool, cold or lukewarm water to hot panels can cause components (e.g., glass) to shrink rapidly, potentially causing cracks and breakages.” Canadian Solar: conduct cleaning “avoiding the hottest hours of the day, in order to avoid thermal stress on the module.”

Water and implement. REC’s order of preference is explicit: “Ideally, de-ionized water should be used to clean the panel. Deionized water is water that has had the mineral ions and salts removed. If de-ionized water is not available, rainwater, tap water or diluted alcohol may be used as a secondary solution.” It permits “A standard flexible garden hose with domestic water pressure” to run water down the whole surface. Canadian Solar specifies “Water with low mineral content”, “Near neutral pH water”, a maximum recommended pressure of 4 MPa (40 bar), and a non-abrasive implement such as a sponge, used during the cool part of the day. Which? is blunter for homeowners: “Only use water to clear your solar panels. Detergents or abrasive materials could harm their surface.”

Never dry-rub. Canadian Solar: “Dirt must never be scraped or rubbed away when dry, as this will cause micro-scratches on the glass surface.” It also recommends avoiding rotating brush methods, “as they could create micro-cracks in the PV modules”, which matters if a contractor or a machine turns up with one.

If marks remain. REC allows “a mild biological and biodegradable washing-up liquid”, immediately rinsed with plenty of water, and rules out acid or alkali detergent. Canadian Solar rules out abrasive cleaners, de-greasers, unauthorised chemicals, and “cleaning corrosive solutions containing hydrofluoric acid, alkali, acetone, or industrial alcohol”. REC also notes it “does not provide warranty for the application of any sprays or treatment”, which covers the coatings sold as making panels self-cleaning.

Never stand on the glass. REC: “Never at any point stand on, walk on, lean on or apply pressure to the panels as this can cause both visible and non-visible damage to the panel.”

Your own manufacturer’s rules may differ from these two, so find its installation manual before you start.

Working at height: the part that actually hurts people

HSE’s guidance is written for people who do this for a living, and the standard it sets them is the right yardstick for whether to go up at all.

HSE’s roof work page states that “Almost one in five deaths in construction work involve roof work”, that the main causes are “falling from roof edges or openings, through fragile roofs and through fragile rooflights”, and that “All roofs should be treated as fragile until a competent person has confirmed they are not.” It also requires that those doing the work “must be trained, competent and instructed in use of the precautions required.”

On ladders, HSE says they “can be a sensible and practical option for low-risk, short-duration tasks, although they should not automatically be your first choice”, and that “The law calls for a sensible, proportionate approach to managing risk”. REC’s own cleaning guide says the same in one line: “When working at heights, use fall and personal protection measures throughout the cleaning process.”

The practical reading: an extendable pole and a hose from the ground is a different activity from standing on a roof, and the roof is the point at which the sensible answer is a quote from a professional.

Solar panel cleaning companies, robots and drones

The manufacturer rules apply whatever is holding the brush. REC states that pressure or steam cleaners, high pressure hoses, knives, blades and metallic sponges “is not permitted on REC panels and will invalidate the warranty”, and Canadian Solar recommends avoiding rotating brush cleaning. Ask any contractor which of those they use before they quote, and speak to a couple of companies and get quotes, as Which? advises. Neither REC nor Canadian Solar addresses domestic cleaning robots or drones, so there is no manufacturer word on how well they work, but the rotating-brush and pressure limits above are the questions to put to whoever is selling you one.

How often should you clean solar panels — and when not to

Neither manual sets an interval. Do not clean:

• When you cannot see dirt. REC’s own trigger is visual: “to optimize electrical output, it is recommended to clean the panels when dirt can be seen on the glass surface.”
• In the middle of a hot day, for the thermal-shock reason above.
• When the glass is cracked or the wiring looks disturbed. That is an electrical fault to report, not a cleaning job. If a storm or hail is the likely reason, solar panels after a storm sets out what to record before anything is tidied away.
• For snow, where REC says clearing it is not essential for maximum performance. If you do clear it, both manuals say use a soft, non-abrasive brush.
• Because a shortfall appeared and dirt is the first explanation offered. That one deserves a number first.

Let your own production history answer it

Which? gives homeowners this trigger: “Monitor how much your solar system generates. If you see an unusual drop, check with your installer what the reason could be.”

That is doable with what you already have, and it is the only way to separate soiling from the things that look identical from the ground: a string that has stopped, a dead optimiser, shade from a tree that has grown, an inverter derating in heat.

The method, in order:

1. Get daily kWh, not a live power reading. Your manufacturer’s app or your monthly report gives a day-by-day total; look for the daily or monthly energy view rather than the live power dial. A single midday glance is not enough; you need a column of daily totals.
2. Pick the days in question and comparable days. Take the fortnight you are worried about, the same fortnight last year, and a fortnight from before whatever changed.
3. Check what the weather allowed on each of those days. A dull fortnight and a soiled array look identical in a bar chart until you put irradiance next to them. The step-by-step is in how to check your solar production against the weather.
4. Look at the shape of the decline. Soiling builds up and then steps back up after heavy rain. A fault does not recover on its own when it rains. If the gap is flat, or appeared overnight, or clears the moment the sun moves past a particular hour, dirt is not the explanation. Start at solar panels producing less than expected.
5. If you have never established a baseline at all, do that before you buy a cleaning kit: how do I know if my solar panels are working and the solar panel health check cover the readings to take and where to find them.

The reason to do the comparison rather than the clean is that the alternative is the pattern in this owner’s post, a shortfall explained away for months by weather and age while a whole string sat dead behind an inverter that reported nothing:

Since installation the generation figures have been slightly lower than before but I put this down to cloudier weather and deterioration of the solar panels. However when I checked the inverter I noticed that whilst string 2 was working fine, string 1 had the correct voltage (360v) … but the ampage was zero and hence not generating any output. The inverter appeared to be working fine and had no error messages.

Dirt is the cheapest explanation to reach for and the easiest to check. Check it with the numbers, and if the numbers say the array is doing what the weather allowed, the grime on the glass is costing you less than the ladder would.

What the daily check says about dirt on a real roof

On a 6.09 kWp Enphase roof in Almere that Solmiro checks every day, September 2026 came to 468 kWh against the 467 kWh the weather allowed on 28 comparable days: 100 %. There was no shortfall for dirt, or anything else, to explain. Whole years on the same roof differ more, 927 kWh per kWp in 2024 against 1,026 in 2025, so a yearly total tells you little about the glass.

Single days are noisier. That September the ratio ran from 66 % on 18 September to 217 % on 2 September. Taken alone, the 18th is the kind of dip that gets put down to grime. It was not a fault, and the roof was back above 100 % on the following comparable days. This is why Solmiro never alerts on one day: a shortfall counts only when it holds over several comparable days.