Skip to content
WattCurious

Do Solar Panels Work on Cloudy and Rainy Days?

Updated 21 September 20267 min readSolar Energy

Yes. A solar panel converts whatever light reaches its surface, not only direct sunbeams, so it keeps producing under cloud and rain. Heavy overcast typically leaves a tenth to a quarter of clear-sky output; bright cloud leaves far more. What the weather takes away is irradiance, not the panel's ability to work.

Key takeaways

  • Cells respond to photons, not to warmth or to direct beam. Under cloud nearly all the light arriving is diffuse, and diffuse light frees electrons exactly as direct light does.
  • Current tracks irradiance almost proportionally, so output falls roughly in step with how much light the cloud lets through — not off a cliff.
  • Rain costs little by itself, and often pays for itself by washing off the soiling that had been quietly costing output for weeks.
  • Broken cloud is not overcast: bright cloud edges can briefly push output above the clear-sky value before the next shadow arrives.
  • A weather dip follows the sky and recovers with it. A fault does not. That difference is the fastest diagnostic available.
On this page
  1. What cloud actually blocks
  2. How much output you lose under different skies
  3. Why cells respond to diffuse light at all
  4. What rain adds, and what it takes away
  5. The shape of a cloudy day
  6. Telling weather apart from a fault

What cloud actually blocks

Sunlight arrives at a panel by two routes. Direct beam travels in a straight line from the sun's disc. Diffuse light has been scattered by air molecules, water droplets and dust, and arrives from the whole sky dome at once. Add them together — plus whatever bounces up off the ground — and you have the irradiance the panel actually sees.

Cloud does not switch light off. It converts it. A droplet-laden cloud scatters the beam so thoroughly that the sun stops being a disc and becomes a bright patch, and eventually a uniform grey ceiling. Some of that scattered light continues downward and reaches the panel; some is reflected back to space. On a clear day the beam does most of the work. Under heavy overcast the beam is effectively gone and the panel runs entirely on diffuse light.

How light reaches a solar panel on a cloudy dayA rooftop solar panel under a partly clouded sky. Where the sky is clear, direct beam travels straight from the sun to the panel. Where a cloud lies in the path, the beam is scattered: part of it is reflected back to space and part continues downward as diffuse light arriving from many directions. A smaller component reaches the panel after reflecting off the ground.SunClouddroplets scatter the beamPaneldirect beamsome is reflected back to spaceDiffuse lightarrives from the whole skyreflected off the groundA cell cannot tell which direction a photon arrived from; only whether it carries enough energy to free an electron.
Cloud does not switch sunlight off — it scatters it. Some of the scattered light still reaches the panel, arriving from all over the sky instead of from one direction.

That is the whole reason panels keep working in bad weather: a cell has no way of knowing which direction a photon came from.

The three components of irradiance on a panelA tilted panel beneath a sky dome. Global irradiance on the panel is the sum of three parts: the direct beam from the sun's disc, reduced by the cosine of its angle to the panel; diffuse light arriving from across the whole sky dome; and light reflected from the ground in front of the panel. Under cloud the direct term collapses and the diffuse term dominates.sky domepanelnormal1. Direct beamfrom the sun's disc, scaled by cos of theangle between beam and panel normal2. Diffusefrom all over the sky dome3. Ground-reflectedgrass, gravel or snow in fronton the panel: direct x cos(angle) + diffuse + ground-reflected = what the cells receiveUnder heavy cloudterm 1 collapses andterm 2 does the work.
What the panel receives is the sum of three components. Cloud shifts the balance from the first to the second; it does not remove all three.

How much output you lose under different skies

Short-circuit current in a silicon cell rises and falls almost in proportion to irradiance. That makes the arithmetic refreshingly simple: let a quarter of the light through and you get roughly a quarter of the current, and therefore roughly a quarter of the power.

What different skies typically leave you
SkyWhat reaches the panelShare of clear-sky output
Clear, sun highMostly direct beam, with a modest diffuse component100% (reference)
Thin or high cloud, sun visible as a discBeam attenuated, diffuse risingroughly 60–80%
Bright overcastAlmost entirely diffuseroughly 20–40%
Heavy storm cloudDiffuse only, strongly attenuatedroughly 5–15%
FogDiffuse, scattered in all directions including upwardroughly 5–20%, highly site-dependent

Illustrative ranges for a fixed array with the sun reasonably high. Cloud thickness, cloud type, sun elevation and how much sky the array can see all move these figures, so treat measured data from your own site as the authority.

Typical share of clear-sky output under different skiesIllustrative bar chart comparing the share of clear-sky output a fixed array typically delivers under five conditions: clear sky at one hundred per cent as the reference, thin or high cloud at roughly sixty to eighty per cent, bright overcast at roughly twenty to forty per cent, heavy storm cloud at roughly five to fifteen per cent, and fog at roughly five to twenty per cent. Each bar is drawn as a range rather than a single value.0%20%40%60%80%100%Share of clear-sky output100%60–80%20–40%5–15%5–20%Clear skyThin orhigh cloudBrightovercastHeavystorm cloudFogIllustrative ranges, not measurements. Cloud type and thickness, sun elevation and how much sky the arraycan see all move these figures; measured data for the site is the only authority.
Illustrative, not measured: the point is the shape, not the exact bars. Output falls steeply under thick cloud but never reaches zero in daylight.

Notice what is missing from that table: temperature. Panels are rated at a cell temperature of 25 °C (77 °F), and a cool overcast day keeps them much closer to that than a hot clear one does. Some of the loss to cloud is quietly returned as a voltage gain, which is why the relationship between sunshine and output is never quite linear. That trade is the subject of why panels lose efficiency in hot weather.

Why cells respond to diffuse light at all

A photon frees an electron if it carries more energy than the semiconductor's band gap — about 1.12 eV for silicon. Energy depends on wavelength, not on direction of travel. A blue photon scattered twice inside a cloud carries exactly the energy it started with, so when it finally arrives at the cell it does the same job it would have done on a clear day.

Direction still matters, but only for how much light gets *in*. Light striking the glass at a shallow angle reflects more readily than light arriving head-on. Modules fight this with an anti-reflective coating and a textured cell surface, both of which work across a wide range of incidence angles — a design choice that pays off precisely when light is arriving from everywhere at once.

How diffuse light is absorbed by a solar cellSunlight enters through an anti-reflective coating and front contact fingers into a thin n-type silicon layer. Below it a depletion region carries a built-in electric field, and beneath that lies the thicker p-type base and the rear contact. An absorbed photon creates an electron-hole pair; the field sweeps the electron up toward the front contact and the hole down toward the rear, driving current through an external load.Light from every directiondirect or scattered: only the energy matters+   +   +   +   +−   −   −   −   −field Ee−h+loadcurrentAnti-reflective coatingand front contact fingersn-type layerphosphorus-doped, very thinDepletion regionthe built-in field lives herep-type baseboron-doped, bulk of the waferRear contactLayer thicknesses are exaggerated; a full wafer is roughly 0.15-0.2 mm (0.006-0.008 in) thick.
A cell cannot tell which direction a photon came from. What matters is whether the photon carries more energy than the band gap, and whether it gets through the glass.

What rain adds, and what it takes away

Rain arrives attached to thick cloud, so production drops during a shower — but blame the cloud, not the water. A film of water on glass reflects a little more light than dry glass and costs a few percent at most.

The compensation is soiling. Dust, pollen, salt and bird droppings build up continuously and reduce output by blocking light before it reaches the cell. In a dry spell this loss grows week by week, invisibly, because there is no sudden change to notice. A decent rain shower resets much of it in an afternoon.

How soiling accumulates between rain showersIllustrative chart of daily output over several weeks. Output declines slowly during each dry spell as dust accumulates on the glass, then steps back up towards the clean baseline after a rain shower. The rainy day itself produces less because of the cloud, marked below the axis, while the days that follow produce more than they would have without the wash.Output on a clear dayWeeksclean glass baselinerainrainraindust builds up; the loss is invisible day to dayeach shower recovers most of itIllustrative. Rate of soiling depends on dust, pollen, salt, birds and the tilt of the array; a steeper tilt sheds more.The shower itself is a low-output day. The fortnight after it is a high-output one.
Illustrative: soiling accumulates slowly and invisibly, then rain resets most of it. The rainy day itself produces less; the following week produces more.

Where rain does not help is when the deposit is greasy, cemented by long-dry mineral dust, or concentrated along the lower frame edge where water drains and evaporates. A residue line along the bottom of the glass is the classic sign that rain alone is no longer doing the job.

The shape of a cloudy day

A clear day produces a smooth arc: output rises with the sun, peaks near solar noon, falls away symmetrically. Overcast produces the same arc at a fraction of the height. Broken cloud produces neither — it produces a spiky mess, and one feature of that mess surprises people.

When the sun clears the edge of a bright cumulus cloud, the panel briefly receives the full direct beam plus unusually strong diffuse light reflected off the white cloud face. For a few seconds the irradiance can exceed what a clear sky would deliver at that moment, and output can spike above its clear-sky value. It is real, it is called cloud enhancement, and a good inverter simply clips or rides through it.

Array output through a clear, overcast and broken-cloud dayIllustrative line chart of array power from dawn to dusk. The clear-sky day is a smooth arc peaking at solar noon. The overcast day follows the same shape at a small fraction of the height. The broken-cloud day swings rapidly between the two, with brief spikes rising above the clear-sky arc where sunlight arrives past a bright cloud edge.Array powerdawnsolar noonduskcloud edges briefly lift output above the clear-sky arcinverter start-up thresholdIllustrative shapes. Overcast lowers the whole curve; broken cloud swings between both extremes within seconds,which is why a daily energy total says more about a cloudy day than any instantaneous reading.clear skybroken cloudovercast
Illustrative day curves. Broken cloud swings between the two extremes within seconds, and the brief spikes above the clear-sky arc at cloud edges are genuine.

Very low light has one more consequence. An inverter needs a minimum input voltage before it can start, so under deep overcast a small array may sit below the threshold and the inverter stays asleep. It will start late, may drop out during the darkest part of the day, and restart as the light improves. That behaviour is designed in.

Telling weather apart from a fault

The useful question is not "is it cloudy?" but "is the output consistent with the sky?" Four checks separate the two, and none of them need a meter:

  1. Does the loss track the weather? Weather losses vary day to day and recover when the sky clears. A fault holds the output down on a bright day too.
  2. Does the shape match? Cloud produces a ragged or uniformly lowered curve. A notch at the same time each day, drifting slowly with the season, is a shadow — see how partial shade affects solar panels.
  3. Do the strings agree? Two strings with the same orientation should fall and rise together under cloud. One string down while the other is fine is not weather.
  4. Is the drop proportional? Heavy overcast cutting output to a fifth is normal. A bright day at a fifth is not, and belongs in the systematic hunt described in why a system produces less power than expected.

The deeper point is that cloudy climates are not unsuitable for solar; they are simply lower-yield. What they change is not whether an array works but how much energy it delivers over a year — and, if the array is paired with storage, how much capacity is needed to ride through a dull week. That sizing question turns on the battery's chemistry as much as the weather, which is where the differences between LFP, NMC and lead-acid start to matter.

Frequently asked questions

Do solar panels work in the rain?

They do, at reduced output. The rain itself blocks very little light; the thick cloud that comes with it blocks a great deal. A wet panel also runs cool, which slightly raises its voltage, and rain washes away dust that was reducing output before the shower.

Do solar panels work in fog?

Yes, but poorly. Fog scatters light in every direction, including back upward, so only a small fraction of it reaches the panel. Fog that burns off by late morning usually costs less over the day than a full overcast that lasts until dusk.

Is a cloudy day the same as shade?

No, and the difference matters. Cloud dims the whole array evenly, so every cell produces a little less. Shade darkens part of the array while the rest stays lit, which forces mismatched cells into bypass and costs far more than the shaded area suggests.

Do panels produce any power at night?

No. Moonlight is roughly a millionth of the intensity of sunlight, which is far below the level at which a cell can drive a useful current, and an inverter will not start on it. Any output recorded overnight is a metering artefact, not generation.

Will the inverter shut down on a very dark day?

It may. Inverters have a minimum input voltage and power at which they start, published on the datasheet, and under deep overcast a small array can sit below it. The inverter waking late and sleeping early on a dark day is normal behaviour, not a fault.

Sources

Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.

Editorial Team

Research, drafting and review

Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.

Last reviewed 21 September 2026. How we research and review