Micro, optimiser or string: the choice nobody explains

The three architectures, one sentence each

The string inverter is the classic solution: panels are wired in series, like fairy lights, and one unit converts all the energy. Optimisers keep the central inverter but add small electronics under each panel so every one works at its own maximum. Microinverters remove the central unit altogether: each panel carries its own inverter and works independently.

String: the sensible choice for a clean roof

When the roof is a single well-oriented face without shading, string is hard to beat: less electronics, fewer points of failure, simpler maintenance and a good price. Its weakness is its series nature: if one panel underperforms through shade or dirt, it drags the chain down. On a clean roof, that weakness barely shows.

Optimisers and micros: when they earn their price

The cases where per-panel electronics stop being a luxury and become the solution:

• Partial shading crossing the roof at certain hours: chimneys, aerials, the building next door. Each panel works on its own and the shadow stops multiplying.
• Roofs with two or three different orientations or pitches, where a single string would force panels living different conditions to behave as equals.
• Panel-level monitoring: with micros or optimisers we see every panel separately, which makes daily monitoring sharper and faults faster to locate.
• Future staged expansions: microinverters let the system grow panel by panel without redesigning strings.

What we look at before deciding

The decision comes from the shading study and the roof's geometry, not from a distributor's catalogue. On a clear south-facing roof we usually propose string; in old-town centres full of chimneys and party walls, per-panel electronics gain ground; and on complex roofs the answer is often mixed.

Whatever the architecture, we demand equipment from consolidated manufacturers with written warranties, and remember that beyond the manufacturer's warranty you are covered by Spain's three-year legal conformity guarantee on everything sold. That part does not depend on the brand: it depends on whoever sold it to you still being around. We plan on staying.

The edge cases where the general rule breaks down

In the old town of Xàtiva and in villages like Canals, la Llosa de Ranes or Genovés, terraced houses are the norm: small pitched roofs facing east and west, with a chimney, a stairwell hut and the neighbouring building's shadow late in the day. A well designed east-west roof produces between 82 and 85 per cent of what the same array would yield facing south, and in an area that PVGIS puts at 1,350 to 1,500 kWh per kWp per year, that is still a lot of energy. The real issue is not the total figure: it is that two orientations mean two different production curves living in the same installation.

A modern string inverter with two MPPT inputs handles east-west without drama if each roof face gets its own input. The true edge case is shade that moves: the neighbour's aerial, a cypress that has grown taller than anyone expected, the water tank next door. Partial shade sweeping across three panels between mid afternoon and sunset punishes the whole string unless there is panel-level electronics. That is where optimisers fitted only to the affected panels tend to be the sensible middle ground: they solve the problem where it exists without making the whole installation more expensive.

Then there is the case almost nobody raises at the start: expansion. If you know that in a few years you will want more panels, a battery or an electric car, the architecture you choose today opens or closes doors. Mixing panels of different power ratings on the same string carries a penalty; with micros or optimisers, each panel works on its own terms and mixing stops being a problem.

• Three or more usable orientations on the same roof
• Partial shade that shifts with the hours or the seasons
• A real plan to expand within a few years
• Existing panels you want to keep alongside new ones

The breakdown that comes with the years: where the electronics are when they fail

All power electronics fail eventually, and the catalogue never tells you where the device will be that day. A string inverter lives on the garage or utility room wall: you diagnose it from the ground, unbolt it with two hands and replace it with whatever the equivalent model is at the time. A microinverter lives under the panel, in one of the hottest spots of the house, and reaching it means climbing onto the roof, lifting the panel and working on a slope. On a village house with traditional curved tiles, that means a harness or scaffolding and a job that does not get done on an ordinary visit.

The simple arithmetic of failure points also counts: an installation with micros has as many electronic devices on the roof as it has panels; with a string, one on the wall. Serious manufacturers offset this with much longer warranties on micros, and it is true that a faulty micro only takes down its own panel while the rest keep producing, which a string cannot claim. Neither side is a trick: they are different trade-offs, and you should choose knowing what you are buying for the long run, not just for commissioning day.

The last detail is spare parts. By the time a replacement is due, the exact model on your roof will most likely no longer be made. Swapping a string inverter for its current equivalent is routine; replacing a discontinued micro in the middle of an installation requires the manufacturer to still exist and to have kept compatibility. That is why, when we recommend panel-level electronics, we look at the maker's track record as much as at the device.

Frequently asked questions