Do Solar Lights Work in the UK? Four Real Installations Tested

Walk around any DIY store or garden centre and you will find solar lights selling for surprisingly little money. As engineers, that immediately raises questions. How much light do they actually produce? What happens in December? And why do two seemingly identical installations perform so differently?

So we stopped speculating and started measuring. Over the past few years Lightmaster has run four of our own solar powered installations as an ongoing research and development exercise, across two countries and four very different sets of conditions. This article sets out what we found, including the numbers that did not match the marketing.

The short answer to do solar lights work in the UK is yes, reliably, but only when the system is sized for the worst month rather than the best one.

Key points

• Shading beats wattage. On our Warwickshire test rig, a single east facing 20W panel generated almost three times more energy on a sunny morning than both panels managed at midday, because tree shading at noon mattered more than orientation.
• Winter output falls by roughly 4.5 times. Manufacturer data for our self contained PIR light shows 398Wh recoverable in July against just 88Wh in December, with usable brightness dropping from 279 lumens to 47 lumens.
• Panel tilt is seasonal, not fixed. Optimum tilt in South Warwickshire is around 14 degrees at midsummer and around 60 degrees in midwinter. A single fixed angle of about 37 degrees is the year round compromise.
• Portable systems underperform fixed ones. Not because the hardware is worse, but because nobody repositions them.
• Battery capacity is the usual failure point. Three 1W lights running six hours a night need only 18Wh a day, but surviving a week of December cloud needs a 12V 20Ah LiFePO4 bank and a 50W to 80W panel.

Why solar lighting performance is so hard to predict

Solar output depends on far more than the panel rating printed on the box. Location, mounting angle, shading, battery chemistry, weather patterns and operating temperature all shape the final result, and they interact. A 100W panel in the wrong place will lose to a 20W panel in the right one.

Context matters too. UK solar deployment is no longer niche: DESNZ deployment statistics recorded around 22.8GW of installed solar capacity across more than 2,076,000 installations by the end of June 2026, against a Clean Power 2030 Action Plan target of 45GW to 47GW by 2030. The technology is proven. The specification is where projects come unstuck.

It also helps to be honest about the resource we are working with. Met Office long term averages for 1991 to 2020 put the UK at roughly 1,403 hours of sunshine a year, which is under four hours a day averaged across the year. That is a workable resource. It is not a generous one, and it is very unevenly distributed across the calendar.

Four installations, four different answers

We compared four systems: a boat installation in the South of France, a fixed test rig at our South Warwickshire R&D site, a portable kit we use for architect CPD sessions, and a factory built self contained solar light with a PIR sensor.

1. Boat system, South of France

System: 2 x 100W Marlec SpectraLite SemiFlex panels, Marlec charge controller, 12V battery bank.

On paper a boat is an ideal solar environment. Water reflects light and southern France receives far more sunshine than the Midlands. In practice, the compromises are significant. Trees along rivers and canals shade panels constantly, and during 40°C summers most boat owners deliberately seek out shaded moorings.

Despite all of that, the system performs exceptionally well through summer, running two fridges continuously while keeping the battery bank fully charged.

What we learned: high sunshine hours can absorb a lot of installation compromise. Panel angle matters less in summer when the sun is high overhead, but winter performance improves dramatically once panel tilt is increased towards roughly 50 degrees.

2. Fixed test rig, South Warwickshire

System: 2 x 20W monocrystalline panels, 30Ah LiFePO4 battery, Victron charge controller.

This produced the most predictable and consistent results of the four. The panels face east and west, with deciduous trees shading the southern aspect.

The most instructive finding came from comparing the two panels directly. On sunny mornings the east facing panel generated almost three times more energy than both panels produced combined around midday. The reason was not orientation. It was that tree shading at noon was more damaging than pointing a panel away from south.

What we learned: shading has a greater impact on annual yield than panel wattage. A fixed installation consistently outperforms a portable one. Monocrystalline cells hold up well in less than perfect light.

3. Portable kit for CPD demonstrations

Portable solar sounds appealing because you can move it to wherever the light is. In reality that flexibility is usually its weakness. Panels get left flat on the ground, positioned badly, or forgotten entirely. Unless someone actively repositions them through the day, yields fall well below expectations.

This is our demonstration kit, which we use for CPD sessions with architects. It is genuinely useful for showing how quickly output changes when you move a panel, which is exactly the point.

What we learned: portable systems need active management to earn their rating. Incorrect panel angle is the single biggest source of lost output. Convenience nearly always costs performance.

4. Self contained residential light with PIR

Specification: 180mm x 180mm square body, integrated solar panel, 10 LEDs totalling 2.4W, PIR motion sensor, dual 2.6Ah lithium batteries, USB-C backup charging, IP66 weatherproof rating.

This category is the sensible compromise between cost, performance and installation effort, and it is where most domestic solar lighting questions end up. The intelligent controller varies brightness according to battery charge, seasonal conditions, weather patterns and motion detection activity, which is why a well designed unit behaves so differently from a cheap one.

Runtime at the three brightness settings works out at roughly five hours at 500 lumens, ten hours at 250 lumens, and 21 hours at 125 lumens.

In the South of France it performs strongly all year. In the UK, winter performance depends almost entirely on where you put it and whether it can see low angle sunlight.

What twelve months of real data looks like

This is the part that rarely appears on packaging. The table below shows month by month performance for the self contained unit, including recoverable energy, the length of night the light has to cover, and the brightness the controller can sustain.

Read that table sideways and the problem becomes obvious. July delivers 398Wh into a seven and a half hour night. December delivers 88Wh into a night nearly twice as long. The unit is being asked to do roughly twice the work with roughly a fifth of the energy, which is why sustained brightness falls from 279 lumens to 47 lumens.

That is not a fault. It is physics, and a good controller managing it sensibly. But it explains why so many people conclude solar lighting “does not work” after their first winter. They specified for the July column and lived with the December one. If lumen figures are unfamiliar territory, our guide to lumens, lux and watts puts these numbers into context.

The mistake almost everyone makes: undersizing the battery

Battery capacity, not panel size, is the most common failure point in solar lighting.

Take a modest scheme of three 1W lights running for six hours on a winter evening. Daily consumption is only 18Wh. On that basis alone you would want a 12V 7Ah to 10Ah battery minimum and a 30W to 50W panel.

But that assumes the sun turns up. If you want the system to ride out a week of poor December weather without going dark, the requirement rises sharply:

The lead acid bank needs roughly double the nameplate capacity for the same usable energy, because you cannot discharge it as deeply without shortening its life.

Winter changes everything

UK solar faces three compounding winter problems: low sun angles, short daylight hours and persistent cloud. They arrive together, which is what makes December so punishing.

Sun angle and panel tilt

In Shipston-on-Stour, midsummer sun elevation reaches around 61 degrees. In midwinter it falls to around 14 degrees. That single change alters how much atmosphere the light has to travel through and how obliquely it strikes a fixed panel.

If a system only has to work in summer, lay the panel closer to flat. If it has to work in January, stand it up steeply. If it has to do both on a fixed mount, 37 degrees is the compromise, and you accept that winter will be lean.

The same array in Agde will typically generate around 40% to 50% more electricity per year than one in South Warwickshire. Part of that is cloud free days. Part of it is atmospheric loss: at higher sun angles the light travels through less atmosphere before it reaches the panel.

What cold does to batteries

Temperature is the quieter half of the winter problem, and it catches people out because it affects storage rather than generation.

• AGM batteries lose roughly 20% to 30% of usable capacity in freezing conditions.
• Standard lithium cells cannot safely charge below 0°C without dedicated protection circuitry.
• Heated LiFePO4 batteries are the best cold weather lithium option, and worth the premium on any system that has to perform through a Cotswolds January.

What we would specify

Across all four installations the same three component choices kept coming out ahead.

Panel: monocrystalline. More efficient per square metre, noticeably better in weak and diffuse light, and typically rated for 25 to 30 years.

Charge controller: MPPT. Typically 20% to 30% more efficient than a traditional PWM controller, and the gap widens in exactly the low light conditions where you need every watt.

Battery: LiFePO4. Compact, efficient and long lived, with the caveat above about charging in the cold.

Add careful attention to panel position and you have the specification that produced our best results. Everything else is detail.

Plug-in solar: what changes on 27 August 2026

Anyone researching solar lighting this year will run into plug-in solar, and the legal position has just changed.

Until now, plugging a generating device into a standard socket sat outside UK wiring rules, regardless of how well built the hardware was. The obstacle was narrower than most people assume: BS 1363 did not permit a plug and socket to be used to connect generating equipment.

The Plugs and Sockets etc. (Safety) Regulations 1994 and Electricity Safety, Quality and Continuity Regulations 2002 (Amendment) Regulations 2026, better known as SI 2026/848, was made on 16 July 2026 and comes into force on 27 August 2026. From that date, certified plug-in solar devices can be sold and self installed in Great Britain. The key limits set out in the DESNZ Plug-in Solar Device Interim Product Specification are:

• Maximum inverter output of 800VA, with maximum AC output current of 3.5A.
• Up to 2,000W of panels behind that inverter.
• A plug straight into a fixed socket. No extension leads, no adaptors.
• Type approved anti islanding protection, so the inverter cannot energise a dead grid.
• DNO notification under G98 remains mandatory before you connect.
• Battery storage is explicitly excluded from this route and still needs a hardwired installation.

The government has confirmed that Currys, B&Q, Amazon and Lidl are working with DESNZ to bring compliant products to market. Worth noting that certified kits may be thin on the ground on day one, because manufacturers still need to get products, documentation, plugs and labelling through conformity assessment.

On the numbers, an 800W system in central England generates roughly 550kWh to 800kWh a year. Without a battery you will typically self consume 25% to 40% of that, and at the current Ofgem price cap unit rate of 26.11p per kWh for 1 July to 30 September 2026, that is somewhere in the region of £40 to £85 a year. Households at home during the day, running fridges, computers, routers and entertainment equipment on daytime generation, will do considerably better. Plug-in kits do not currently qualify for Smart Export Guarantee payments, so it is sensible to assume nothing for export and treat all savings as self consumption.

The biggest lever is behavioural. Run the appliances when the sun is out.

It is also worth knowing where the wider policy is heading. Under the Future Homes Standard, published as the Part L 2026 update to the Building Regulations for England, new homes will need solar PV equivalent to 40% of the dwelling’s ground floor area from 24 March 2027. Solar is becoming a default rather than an upgrade.

The verdict

Our testing showed that solar lighting works, and works well, when systems are correctly specified and installed. The problem is almost never the technology. It is unrealistic expectations.

Cheap solar lights fail because they are undersized, badly positioned, or asked to perform through a long British winter on a battery that was never going to manage it. The best results came from monocrystalline panels, MPPT charge controllers, adequate battery capacity, and genuine care over where the panel goes.

Solar also has real limits worth naming. It is excellent for remote positions with no cable route, for wayfinding and for motion triggered security. It is not the right answer where you need guaranteed output at a guaranteed level every night of the year, or where a scheme depends on precise dimming and colour consistency. Those jobs still want a mains supply and proper lighting control.

Any permanently installed exterior lighting, solar or mains, also needs to respect the same fundamentals: an appropriate IP rating for the position, sensible colour temperature, and shielded downward light in line with dark skies good practice. A solar light that spills upwards is still light pollution.

As with most engineering, the data tells the story. Solar can be highly effective, but only when the installation matches the environment it has to work in.

If you are weighing up solar against a mains scheme for a driveway, garden or approach, talk to our team. We are happy to tell you when solar is the wrong answer.

Frequently asked questions

Do solar lights work in the UK in winter?

Yes, but at greatly reduced output. Our monitored data shows a self contained solar light recovering 398Wh in July against 88Wh in December, with sustained brightness falling from 279 lumens to 47 lumens. A system sized for summer will disappoint in winter. A system sized for December will work all year.

What angle should a solar panel be in the UK?

For a fixed mount at around 52 degrees north, roughly 37 degrees from horizontal is the best year round compromise. The seasonal optimum is about 14 degrees at midsummer and about 60 degrees at midwinter, so an adjustable mount that you reposition twice a year meaningfully improves winter yield.

Is monocrystalline or polycrystalline better for solar lighting?

Monocrystalline, in almost every UK case. It is more efficient per square metre, performs better in weak and diffuse light, which describes most of the British winter, and typically carries a 25 to 30 year rating.

What size battery do I need for solar garden lights?

Three 1W lights running six hours a night use 18Wh a day, which suggests a 12V 7Ah to 10Ah battery as a bare minimum. To ride out a week of poor winter weather, size up to a 12V 20Ah LiFePO4 bank or a 12V 40Ah to 50Ah AGM bank, paired with a 50W to 80W panel.

Is plug-in solar legal in the UK?

From 27 August 2026, certified plug-in solar devices can be sold and self installed in Great Britain under SI 2026/848. Output is capped at 800VA with up to 2,000W of panels, the kit must plug into a fixed socket rather than an extension lead, DNO notification under G98 is still required, and battery storage is excluded from this route.

Does shading matter more than panel size?

Frequently, yes. On our Warwickshire test rig a single east facing 20W panel outproduced both panels combined at midday by almost three to one on sunny mornings, purely because of tree shading at noon. Moving a panel out of shade is usually cheaper and more effective than buying a bigger one.