String Inverters vs Microinverters vs Hybrid

What the inverter actually does

An inverter has two jobs, and buyers usually only hear about the first. The obvious one is conversion: panels produce direct current, your house runs on alternating current, and the inverter bridges that. The one that decides your annual yield is maximum power point tracking.

A panel's voltage and current trade against each other along a curve, and only one point on that curve produces the maximum power. That point moves continuously as light and temperature change. An MPPT circuit hunts for it many times a second. The critical design question is how many independent MPPT circuits your system has, because every panel sharing an MPPT is forced to operate at the same point — and the panel in the worst condition drags the rest towards it.

That single fact explains the whole comparison below. If you want the underlying physics, our guide on how solar panels work covers series wiring and why current is limited by the weakest cell in a chain.

String inverters

The traditional arrangement. Panels are wired in series into one or two "strings", the string runs down to a single inverter box mounted on a wall, and that box does all the conversion and tracking for the whole array. Most residential string inverters have one or two independent MPPT channels, so you can put a south-facing string on one channel and a west-facing string on the other.

What it gets right. It is the cheapest topology by a clear margin. There is one device to install, one to service, and it lives at eye level in a garage or on an outside wall rather than on the roof. Conversion efficiency is the highest of the group — good units reach 97 to 98% weighted efficiency — because a single large converter is inherently more efficient than many small ones.

What it gets wrong. Shading. Because the string shares one MPPT, a single panel in shadow pulls the operating point of every panel connected to it. Bypass diodes limit the damage but do not remove it. It is also a single point of failure: when the inverter dies, the entire array stops. And per-panel monitoring is not available, so a gradually failing module can go unnoticed for years — you only see the array total.

String plus power optimisers

The middle path, and the one most often left out of comparisons. A power optimiser is a small DC-DC converter fitted behind each panel. It does the per-panel MPPT locally, then hands conditioned DC to a conventional string inverter which handles only the DC-to-AC conversion.

This buys most of what microinverters offer — per-panel tracking, per-panel monitoring, module-level shutdown for firefighter safety — while keeping one central conversion stage and its efficiency. It costs less than a full microinverter system and more than a bare string. The trade-off is that you still have a central inverter as a single point of failure, and you now have roof-mounted electronics as well, so you have added components without removing one.

In many markets this is the pragmatic default for a moderately complex roof.

Microinverters

Each panel gets its own complete inverter mounted behind it, converting to AC right there on the roof. Panels then feed a shared AC bus rather than a DC string. Every panel has its own MPPT and reports its own output.

What it gets right. Shade tolerance is genuinely per-panel: a shadow costs you only that panel's output. Multiple roof orientations need no planning at all, because there are no strings to keep matched. Failure is graceful — one dead microinverter costs one panel, not the array. Monitoring is per-panel, so you can see a specific module underperforming. And because conversion happens at the panel, there is no high-voltage DC on the roof, which simplifies compliance with module-level rapid-shutdown requirements and reduces DC arc-fault risk. Warranties reflect the design intent: 20 to 25 years is standard, against 5 to 12 for a typical string inverter.

What it gets wrong. Cost — commonly 20 to 30% more for the inverter portion of the system. Conversion efficiency is slightly lower, typically 96 to 97%, because many small converters lose more than one big one. And when a unit does fail, it is on the roof, under a panel, which means access equipment and labour rather than a wall-mounted swap.

Hybrid — a different question, not a third option

"Hybrid" is often listed alongside string and micro as though it were a competing topology. It is not. A hybrid inverter is a string inverter that can also charge and discharge a battery, managing panels, battery and grid in one unit. The topology question (how many MPPTs, where they sit) and the storage question (can this system run a battery) are independent.

The reason it matters at purchase time is sequencing. If you fit a plain grid-tied inverter now and want storage in three years, you add a separate battery inverter and AC-couple it — which works, but means paying for a second conversion stage and losing a few percent to the extra round trip. Buying hybrid up front costs more than a plain string inverter but less than the two-stage path, provided you actually add the battery.

The comparison, in one table

Read the efficiency row carefully, because it is the one people over-weight. The one-point difference between 98% and 97% is worth about 1% of your annual generation. A single panel shaded for three hours a day on a string system can cost far more than that. Topology decides shading behaviour; conversion efficiency is a rounding error next to it.

How to decide

Do a shading assessment first, then let the answer fall out. Stand where the array will go and look for anything that casts a shadow — trees, chimneys, satellite dishes, parapet walls, taller buildings — at morning, midday and late afternoon, and remember that winter shadows are much longer than summer ones. An installer with a solar pathfinder or a phone-based sun-path app can quantify it in half an hour.

Then answer the storage question separately. If a battery is a firm plan rather than a vague ambition, specify a hybrid inverter now — but check that the model's battery voltage window and usable capacity match the storage you actually intend to buy, because "battery ready" on a datasheet is not the same as compatible with a specific pack.

Choose the topology from your shading survey, not from a brochure. Then choose hybrid or not from whether a battery is a decision or a daydream.

Whichever you pick, the inverter is the component most likely to need replacing during the life of the array, so put its replacement cost into your financial model. Our payback guide shows what omitting it does to a payback figure.

Common questions

Are microinverters worth the extra cost?

On a shaded, multi-orientation or hard-to-access roof, yes — the recovered generation and the graceful failure behaviour justify it. On a single-plane unshaded roof, a string inverter produces almost the same annual yield for less money, and the premium mostly buys you monitoring.

How long do solar inverters last?

Plan for a string inverter to be replaced once in the panels' lifetime, somewhere around year 10 to 15; warranties run 5 to 12 years. Microinverters are typically warranted 20 to 25 years, but replacing one means getting back on the roof and lifting a panel.

Do I need a hybrid inverter if I might add batteries later?

Not required, usually cheaper. A hybrid manages panels and battery in one unit. Adding an AC-coupled battery to a plain grid-tied system later also works, but you buy a second inverter and lose a little efficiency to the extra conversion.

Can I mix inverter types on one house?

Technically yes — a string inverter on the clean south roof and microinverters on the awkward shaded section is a legitimate design. It complicates monitoring, commissioning and warranty administration, so it is worth doing only when one roof section is genuinely problematic.

What size inverter do I need relative to my panels?

Slightly smaller than the array is normal and intentional. A DC-to-AC ratio of about 1.1 to 1.3 is common, because panels rarely hit their rated output, so a marginally undersized inverter runs in its efficient band more of the time and clips only a few peak hours a year. Sizing the inverter to the full DC rating usually wastes money.