Three inverter architectures dominate UK commercial solar in 2026: pure string, string-with-DC-optimisers, and pure microinverter. Each has a defensible place. Choosing wrong on a 250 kW commercial install can cost 8-15k pounds in unnecessary capital or 8-12 percent in unrealised yield over 25 years. The decision is not about picking the "best" architecture in the abstract — it is about matching architecture to roof, shading, occupancy, and serviceability profile. This page lays out how each architecture works, the cost and performance economics, and a practical decision matrix.
How each architecture works
String inverter. Panels are wired electrically in series to form strings of typically 12-24 panels. Multiple strings run in parallel to a single central inverter unit which converts the combined DC power to AC. The inverter is the only intelligent device — it tracks the maximum power point, manages grid synchronisation, and provides monitoring. Modern string inverters are 96-99 percent peak efficient and span 5 kW to 250 kW per unit.
String inverter with DC optimisers. A hybrid architecture. Panels are still wired in series strings to a central inverter, but each panel has a small DC-to-DC converter (the optimiser) attached behind it. The optimiser performs per-panel maximum-power-point tracking and conditioning before passing the regulated DC up the string. The string inverter runs at fixed voltage rather than tracking, simplifying its job. Net result: per-panel optimisation with central-inverter cost economics. SolarEdge and Tigo dominate this category.
Microinverter. Each panel has its own miniature grid-tied inverter mounted directly behind it. The microinverter converts panel-side DC to grid-frequency AC at the panel level. The output of all panels then runs as AC down a trunk cable to the building distribution. There is no central inverter — the architecture is fundamentally distributed. Enphase dominates this category.
Cost — where each architecture sits in 2026
| Architecture | Inverter cost per kW | Total install cost difference (vs pure string) |
|---|---|---|
| Pure string | 60-150 pounds per kW | baseline |
| String + DC optimisers | 110-230 per kW (inverter + optimisers) | +8-15 percent |
| Pure microinverter | 250-400 per kW | +15-25 percent |
The cost gap on a 100 kW commercial system: pure string total install around 95k pounds, string-plus-optimisers around 105-110k, pure microinverter around 110-118k. The headline gap is 15-25 percent of total install cost. Whether the additional capital is justified depends on yield uplift, monitoring value, and serviceability profile.
Performance — when does the architecture matter
Three scenarios produce a meaningful yield uplift from optimisers or microinverters versus pure string.
Shading. A string with one fully shaded panel loses roughly 80-90 percent of that string output. The same shading pattern with optimisers or microinverters loses only the shaded panel power — typically 4-8 percent of the array. Real-world UK commercial shading sources: chimneys, dormers, parapet walls, adjacent buildings, plant rooms, mast structures, vegetation. A site with no shading benefits zero from optimisers; a site with 15 percent of array shaded for 2-3 hours per day benefits 8-12 percent annual yield uplift from optimisers.
Multiple orientations. Commercial buildings frequently have multiple roof faces — east-and-west pitched roofs, complex hipped configurations, or south-facing facade installations alongside roof panels. Pure string inverters require a separate MPPT input per orientation (most modern commercial string inverters offer 2-6 MPPT inputs). Optimiser systems treat each panel independently, allowing arbitrary mixing of orientations on the same string. Yield uplift from this on a multi-orientation roof: 3-7 percent.
Panel mismatch. Even on a uniform roof, manufacturing tolerances mean panels vary by 2-4 percent on power output. With string inverters, the lowest-performing panel sets the string current. With optimisers each panel runs at its own MPP. Real-world mismatch uplift is small — typically 1-2 percent — and rarely justifies optimisers on its own.
Soiling and ageing. Differential soiling and panel-to-panel degradation produce small mismatches over time. Same logic as manufacturing mismatch. Marginal benefit.
Reliability and serviceability
String inverter failures are typically sudden and system-level. A 50 kW commercial string inverter that fails takes the whole array offline. Repair: 1-2 days for a like-for-like swap, typically 100-150 pounds per kW for the new unit. Mean time between failure (MTBF) on Tier-1 string inverters is around 100,000 hours — roughly 11 years of continuous operation. So most commercial string inverters fail once over a 25-year asset life and replace once.
DC optimisers are passive DC-to-DC converters. SolarEdge published data shows annual failure rates below 0.5 percent over 25-year tracking. With 25 panels per 10 kW system, that means 25 optimisers per 10 kW. At 0.5 percent annual failure, an installation might see 1-2 optimiser swaps per 25 kW per decade. Cost per optimiser swap (parts plus labour plus access): 250-500 pounds. Lifetime optimiser maintenance cost on a 100 kW system: roughly 2-5k pounds over 25 years.
Microinverters are full grid-tied power-electronics devices, one per panel. Enphase published data on IQ8 series suggests 0.05 percent annual failure rate. Same panel count as optimisers (25 devices per 10 kW). Cost per microinverter swap (parts plus labour plus access): 350-700 pounds. Lifetime microinverter maintenance cost on a 100 kW system: roughly 3-7k pounds over 25 years.
The kicker: access cost dominates. A wall-mounted ground-floor string inverter swap is fast and cheap. A microinverter swap behind a panel on a 30-metre warehouse roof needs MEWP or scaffolding and a roof crew. On large commercial systems with high access cost, string wins on total lifetime maintenance cost despite individual device reliability being similar.
Monitoring — what data each architecture gives you
String inverters give whole-array monitoring. Total daily kWh, total current, total voltage, peak power. The data is sufficient for commercial energy reporting and SEG export metering. Faulty panels show up only as a slow degradation in overall array yield over months or years.
DC optimisers and microinverters give per-panel monitoring. Each panel reports voltage, current, power, and cumulative yield individually. The monitoring portal (SolarEdge, Tigo, Enphase) flags underperforming panels within hours. This matters most on systems above 200 panels where eyeballing array-level data is too coarse to identify single-panel issues. For sub-100-panel systems, per-panel data is rarely actioned by the building owner.
Per-panel monitoring also helps with warranty claims. If a panel underperforms its warranty curve, per-panel data is the evidence the manufacturer needs. With string-only data, isolating a single faulty panel for warranty action is much harder.
Brand examples for each architecture
Pure string inverters. Sungrow SG series (commercial mainstream, cost-competitive), Huawei FusionSolar SUN2000 (smart-grid features, AI optimisation), SMA Sunny Tripower (German engineering, premium long-term support), Fronius Symo and Tauro (Austrian, Tauro field-replaceable cards). For UK commercial work in 2026 our default specification is Sungrow or Huawei on cost-driven projects, SMA on premium long-life projects, Fronius Tauro where serviceability matters most.
String + DC optimisers. SolarEdge HD-Wave residential and SolarEdge Synergy commercial (the dominant DC-optimiser brand globally), Tigo TS4 family (retrofit-friendly, vendor-agnostic — works with any string inverter). SolarEdge for new commercial installs is our default optimiser specification. Tigo we use for retrofitting optimisers onto existing string installations.
Pure microinverter. Enphase IQ8 and IQ10 series (the dominant brand globally — 50+ million microinverters shipped). APsystems also offers commercial microinverters but at lower UK volume. Hoymiles is a Chinese alternative. For UK commercial 2026 microinverter specifications we default to Enphase IQ8H-72-2 commercial.
UK 2026 commercial market split
Approximate UK commercial install volume by architecture across our portfolio in 2025-2026:
- Pure string: ~70 percent
- String with DC optimisers (SolarEdge or Tigo): ~25 percent
- Pure microinverter (Enphase): ~5 percent
The string dominance reflects the fact that most UK commercial systems are unshaded south-facing flat or pitched roofs above 100 kW where the cost-and-reliability case for string is decisive. The optimiser share is concentrated on shaded sites, conservation buildings, and 50-150 kW systems where per-panel monitoring is genuinely actionable. Microinverter share is concentrated on small-commercial (under 30 kW) and aesthetic-driven projects.
The decision matrix
| Site characteristic | Recommended architecture |
|---|---|
| Unshaded south-facing flat or pitched roof, single orientation, 100 kW+ | Pure string |
| Multi-orientation roof (east-west, hipped, facade-plus-roof) | String + DC optimisers |
| Heavy or partial shading (chimneys, dormers, plant, vegetation) | String + DC optimisers or microinverter |
| Listed building, conservation area, heritage | String + DC optimisers (rapid shutdown compliance) |
| Sub-30 kW small-commercial | Pure microinverter or string + optimisers |
| 500 kW+ industrial, simple unshaded roof | Pure string (cost and reliability) |
| High-value asset where per-panel monitoring matters | String + DC optimisers |
| Site with poor roof access for future maintenance | Pure string (lower maintenance cost-per-fault) |
Hybrid pragmatic specification we sometimes use
For a complex roof with one shaded section and one clean unshaded section, we sometimes split the system: pure string on the unshaded section, string-plus-optimisers on the shaded section. The customer pays the optimiser premium only on the panels that benefit from it. Net cost saving on a 200 kW mixed-roof system versus full-array optimisers is around 6-9k pounds with similar yield uplift on the shaded portion.
Authority and reference sources
Engineering Recommendation G99 inverter requirements at Energy Networks Association. SolarEdge published reliability data at SolarEdge. Enphase published reliability data at Enphase. MCS commercial certification at MCS.
Related decision pages
For inverter brand-by-brand comparison see best commercial solar inverters. For replacement and end-of-life see solar panel inverter replacement. For monitoring detail see solar panel monitoring. For panel selection see solar panel manufacturers UK and tier-1 solar panels UK. For system survey see commercial solar survey. For business case see are commercial solar panels worth it.
Common questions
What is the cost difference between string and microinverter on a commercial system?
String inverters typically cost 60-150 pounds per kW for the inverter unit alone in 2026. DC optimisers add roughly 50-80 pounds per kW on top of a compatible string inverter. Pure microinverter systems cost 250-400 pounds per kW for the inverters plus ancillary AC trunk-cable. On a 100 kW commercial system, string is around 12-15k for inverters total, optimiser-string is 18-23k, and microinverter is 30-40k. The price gap narrows as system size grows because microinverter pricing is per-panel (constant cost per kW) while string inverter pricing improves with size (declining cost per kW).
When should I specify microinverters or DC optimisers over plain string?
Three scenarios make optimiser or microinverter the right call. Heavy or partial shading where a string of panels is shaded for more than 1-2 hours per day on average — without optimisers, one shaded panel cuts the whole string output. Multiple roof orientations where panels face different directions or have different tilts (a north-east facade plus a south facade on the same building) — string inverters need separate strings per orientation, optimisers and micros do not. Listed buildings, conservation areas, or aesthetic-driven projects requiring per-panel monitoring and rapid-shutdown compliance for fire-safety reasons. For unshaded south-facing flat or pitched commercial roofs above 100 kW, plain string almost always wins on cost and reliability.
How does shading affect string versus microinverter performance?
On a string-inverter system, panels are wired electrically in series to form a string. The current through the whole string is limited by the lowest-performing panel. If one panel in a 12-panel string is fully shaded, all 12 panels are pulled down to the shaded panel current — losing roughly 90 percent of that string output. With DC optimisers each panel is individually voltage-and-current-managed so a single shaded panel only loses its own production. With microinverters each panel is independently grid-connected so one shaded panel has zero impact on the others. Real-world shading impact: a 5-10 percent shaded array using string runs at 70-80 percent of expected; the same array using optimisers runs at 92-95 percent.
What about reliability — which architecture has fewer faults?
String inverters have one or two large devices. They fail at system level — when the inverter fails, the whole array is down until repair. But there are very few moving parts to fail. Optimisers have one device per panel — typically 25 panels per 10 kW system means 25 devices per 10 kW — but each device is a passive DC-DC converter with no fans, no electrolytic capacitors, no IGBTs. Failure rates are very low, sub-1 percent over 25 years on SolarEdge published data. Microinverters have one inverter per panel — 25 devices per 10 kW each running grid synchronisation, AC conversion, fans, capacitors, and IGBTs. Failure rates are higher than optimisers but published Enphase data suggests 0.05 percent annualised failure for IQ8 series. Net: per-system uptime is similar across architectures; per-panel risk profile differs.
How does access cost change the reliability calculation?
A 5 kW string inverter ground-mounted in a commercial plant room is replaced in half a day at low cost. A 25-kW string inverter wall-mounted on the side of a 30 metre warehouse needs an MEWP and a half-day install. A failed microinverter behind a panel on a 500 kW commercial roof needs a roof crew, MEWP or scaffolding access, and 2-4 hours of labour to swap one device. The cost-per-fault is much higher on microinverters and optimisers despite their per-device reliability being equivalent or better. On large commercial systems where roof access cost dominates, string usually wins on lifetime serviceability cost.
What monitoring do I get from each architecture?
String inverters give whole-array monitoring — total kWh, total current, total voltage. Sufficient for commercial billing and SEG metering. Faulty panels show up only as a degraded array yield over time, not a per-panel fault. DC optimisers give per-panel monitoring — every panel reports voltage, current, power, and yield individually, and faulty panels are identified within hours. Microinverters give the same per-panel granularity. For commercial sites with 200+ panels, per-panel monitoring is genuinely useful for proactive maintenance and warranty claims; for sub-50-panel systems the additional data is rarely actioned.