Four versions of the same worry account for a great deal of the doubt about solar in Britain: winter, cloud, snow and darkness. Three of them have reassuring answers and one does not, and the useful thing is to be precise about which is which, because an installer who waves all four away is as unhelpful as a sceptic who treats them all as fatal.
Winter: much less, and not nothing
A British array generates far less in December than in June, because the days are short and the sun is low, and the winter months are a small fraction of the annual total. What that does not mean is that the panels stop. They work through the winter at a reduced rate, and cold itself is not the problem; panels are slightly more efficient when cold, and it is the length and angle of the daylight that cuts the output rather than the temperature. This is also why the annual figure, not a winter one, is what a payback calculation has to be built on.
Cloud: reduced, because diffuse light still counts
Panels respond to light rather than to direct sunshine, so an overcast day produces a fraction of a clear one rather than nothing at all. This is the reason solar works at all in a climate like Britain's, and the reason a modelled annual figure for a British roof already has the cloud in it. A quote that assumes clear sky conditions is not describing this country, which is one of the things worth checking when comparing three of them.
Snow: a short interruption, not a season
Snow lying on a panel blocks it, and a panel under snow produces very little. In most of Britain that is a handful of days a year, the panels are smooth and pitched so the snow tends to slide once the surface warms, and it happens in the weeks when the array was producing least anyway. It is not a reason to change a decision, and clearing snow off a roof by hand is a much worse idea than waiting.
Night: nothing, and that is what a battery is for
This is the one with no reassuring answer. Panels generate nothing in darkness, and any product sold as generating at night is not doing what it claims. What changes the picture is storage: a battery charged during the day is what carries a household through the evening peak, which is exactly when demand is highest and when electricity bought from the grid costs the most. That is the honest case for storage, and it has nothing to do with the panels working after dark.
Other questions people ask
Do solar panels work on cloudy days?
Yes, at a reduced rate, because panels respond to light rather than to direct sunshine and diffuse light on an overcast day still counts. A modelled annual figure for a British roof already includes the cloud, which is why the annual number rather than a good day is what a payback should be built on.
Do solar panels work at night?
No. Panels generate nothing in darkness. A battery charged during the day is what covers the evening, which is when a household uses the most and when grid electricity is dearest, and that is the real argument for storage.
Does snow stop solar panels working?
While it lies on them, largely yes, and in most of Britain that is a few days a year in the weeks when the array was producing least anyway. Panels are smooth and pitched, so snow usually slides off once the surface warms. Going up to clear it by hand is far more dangerous than waiting.
Are solar panels less efficient in cold weather?
No, the opposite: panels are slightly more efficient when cold. Winter output is low because the days are short and the sun is low in the sky, not because of the temperature, which is a distinction worth holding on to when somebody blames the cold for a poor figure.