Inverters, Wiring & Balance of System

Strip it to essentials and the inverter has two duties, one obvious and one subtle.

The obvious duty is conversion. Modules produce DC - a steady, one-way flow. The building's wiring and the grid use AC - current that reverses direction fifty times a second (in India, 50 Hz). The inverter uses fast electronic switching to synthesise a clean AC waveform from the incoming DC, matched precisely to the grid's frequency and voltage and phase, so the solar power can merge seamlessly with the building's supply. Modern inverters do this very efficiently - typically converting the large majority of the DC they receive into usable AC - but never perfectly: a few percent is always lost as heat, which is one reason the inverter needs a cool, ventilated home.

The subtle duty is maximum power point tracking (MPPT). A PV array does not deliver a fixed output; at any instant it has a curve of possible operating points, and on that curve there is one 'knee' - the maximum power point - where voltage multiplied by current is greatest. That point is not fixed: it drifts with irradiance (more or less sun), with temperature (hotter cells push it), and with shading. If the inverter simply held a fixed voltage, it would sit off the knee most of the time and leave energy on the table. So the inverter runs an MPPT algorithm that continuously nudges its operating voltage to stay on the knee as conditions change - a small, ceaseless optimisation that meaningfully raises yield over a day.

Why this matters for BIPV is the catch hidden in MPPT: a tracker optimises whatever is connected to it as a group. If a single MPPT input serves a whole string of modules and part of that string falls into shade or faces a different direction, the tracker can only find one compromise point for the mismatched group - and the shaded or ill-matched modules drag the healthy ones down. On a clean, uniform, unshaded roof that is fine. On a BIPV facade that is partly shaded by a neighbour, or that wraps around corners so modules face several ways, one big MPPT over the lot is exactly the wrong tool. The fix is to give mismatched parts of the array their own optimisation - which is precisely what the three inverter arrangements in the next section are about.

There are three common architectures, and the choice between them is one of the most consequential in a shaded or complex array.

String inverters are the traditional, economical default. Modules are wired in series into one or more 'strings', and each string (or a couple of strings sharing MPPT inputs) runs to a single central inverter box. It is cheap, simple, efficient and easy to service in one place. Its weakness is mismatch: because a string shares one operating point, its output tends to be pulled toward its weakest module - the most shaded, dirtiest or hottest one. On a uniform unshaded roof this barely matters; on a partly shaded or multi-orientation array it can waste real energy.

Power optimisers keep a single central (string) inverter but add a small DC electronic device on each module that conditions that module's output before it joins the string, so the string can run at its best point while each module contributes its own maximum. This is 'module-level power electronics' (MLPE) done on the DC side. It recovers much of the loss from shade and mismatch, adds per-module monitoring, and can help with safety shutdown - at extra cost and more components on the roof or facade.

Micro-inverters go furthest: a tiny inverter is fitted to each module, converting that module's DC to AC right there, so every module has its own MPPT and runs completely independently. Shade on one module affects only that module. Micro-inverters give the best tolerance of shading and mixed orientations, module-level monitoring, and an all-AC array (no high-voltage DC running through the building) - but they are the most expensive per watt and put more electronic units up on the array where access is hard.

The design rule of thumb (and it is only that - the binding choice is the engineer's) is: the more uneven the array, the more per-module electronics earn their keep. A clean, uniform roof is happy with a string inverter. A BIPV facade that is partly shaded, wraps several orientations, or where every last unit matters, is a strong candidate for optimisers or micro-inverters. BIPV also often *has* many small, awkward, differently-lit modules by its very nature, which pushes it toward module-level solutions more often than plain rooftop BAPV does.

Between the modules and the inverter, and beyond it, runs the wiring - and wiring is never merely 'connecting things up'. Cable has resistance, and resistance turns some of the precious generated power into waste heat along the way. The longer and thinner the cable, the greater the loss; keep DC runs short and adequately sized and the loss is small, let them sprawl thin and long and it quietly eats yield for the life of the system. In a BIPV facade, where cable may snake a long way through the envelope to reach a plant room, this is a real design consideration - short, well-sized, well-routed runs are not a detail, they are performance. (The exact cable sizing is an engineering calculation, and it is the engineer's; the design point is to give the runs short, planned, serviceable paths.)

A second wiring-adjacent idea is the DC-to-AC ratio (sometimes called inverter sizing or the 'ILR'). Designers often, deliberately, pair an array with an inverter whose AC rating is somewhat *smaller* than the array's peak DC rating - because the array only rarely hits its full nameplate (that needs perfect, cool, noon conditions), so a slightly smaller inverter runs closer to its efficient range most of the time and costs less, at the price of 'clipping' a little peak output on the very brightest moments. This is a genuine optimisation with a real trade-off, and the right ratio depends on the site, the orientation and the tariff. It is flagged here so a designer recognises the term and understands that a 'smaller' inverter is often a sensible choice, not an error - but the actual ratio is the engineer's to set.

The broader point is that the wiring and inverter sizing are where a chunk of the gap between nameplate and delivered energy lives (we quantify that gap in 2.4). A gorgeous array on lazy wiring and a badly sized, badly cooled inverter will underperform a plainer array that got these unglamorous choices right. For the designer, the actionable parts are spatial and coordinative: short protected cable routes, a cool ventilated accessible home for the inverter, and an early conversation with the engineer about arrangement and sizing - especially for any array that is shaded, split across orientations, or integrated into the envelope.

Everything in a PV installation that is not the module is the balance of system (BOS): the inverter, all the DC and AC cabling and connectors, isolators and switchgear, protection and earthing, the mounting or the integration substructure, the metering and the monitoring. It is easy to dismiss as plumbing. It is, in fact, where much of the cost sits and where much of the real-world performance is decided - and in BIPV it can be the majority of both.

Start with cost. On a plain roof, standard racking and short runs keep BOS to a moderate share. In a BIPV facade the 'mounting' is a bespoke weather-tight substructure engineered to hold the modules *and* keep the building dry; the cabling is long and concealed; the modules are custom. The non-module cost can meet or exceed the module cost - a core reason BIPV must earn its place against cheaper BAPV, and a number a designer must carry honestly from the first budget rather than discover on site.

Then performance and durability. A cheap module on a badly designed BOS still performs badly: undersized cable wastes power; a sun-baked inverter derates and dies young; poor connectors are a leading cause of faults and even fires; inadequate earthing and protection are a safety failure, not a cost saving. Good BOS is quiet - it just lets the array deliver, safely, for decades. And monitoring, often treated as optional, is what turns an invisible underperformance into a fixable fault: without it, a string that has quietly failed can lose energy unnoticed for months.

For BIPV specifically, the BOS also carries the integration burden. The substructure must do double duty (structure plus weatherproofing); the cabling must be hidden yet inspectable; the whole assembly must be serviceable without dismantling the facade. These are architectural problems as much as electrical ones, and they are where the designer and engineer must collaborate closely from early on.

The honest summary: the module is the star, but the balance of system is the supporting cast that determines whether the show works. Respect it in the budget, the space plan and the detailing - and defer its binding parts (protection, earthing, isolation, sizing, connection) to the engineer, the manufacturer's data and the codes.