Designing a BESS for a Commercial Site: What to Settle Before You Pick a Battery

Battery energy storage is often discussed as a product decision — which chemistry, whose cells, what warranty. In our experience it is an engineering decision that has already been made several steps earlier, by the site's tariff, its load profile and the point where the system connects. Get those right and the equipment selection becomes almost mechanical. Get them wrong and no amount of good hardware rescues the project.

Start with the tariff and the interval data

A commercial BESS earns its keep in one of a few distinct ways: shaving peak demand to reduce demand charges, time-shifting on-site solar generation for self-consumption, arbitraging between tariff periods, providing backup to selected loads, or — where the market allows it — offering services back to the grid. Each of these pays differently, and each implies a different machine. Before we size anything, we want metered interval data at the finest resolution available, ideally covering a full year of seasonal variation, together with the tariff structure in full: how the demand charge is measured, over what averaging interval, in which windows, and whether any ratchet applies. A design derived from twelve monthly bills instead of interval data is a guess wearing a spreadsheet.

Power and energy are two separate specifications

A storage system carries a power rating in kW and an energy rating in kWh, and the ratio between them — the C-rate the system is designed to sustain — matters more than either figure alone. Demand-charge management is largely a power problem: you may need substantial kW for a comparatively short window, several times a month. Solar self-consumption is an energy problem: moderate power, but hours of it, cycled daily. Two systems with identical kWh but different C-rates cost differently, are cooled differently, and age differently. Specifying the ratio backwards is the most common and most expensive early error we encounter.

Where it connects, and to what

Where photovoltaics are present, the storage can be DC-coupled — sharing a converter with the array, which allows it to capture energy that would otherwise be clipped at the inverter — or AC-coupled, which is simpler to retrofit onto an existing installation and keeps the two assets independent of one another for maintenance and availability. Beyond that, the point of common coupling determines protection coordination, metering, and the grid-code compliance the utility will require. Those approvals belong at the start of the design, not at commissioning, because they can constrain inverter selection and export behaviour.

Backup is a separate decision from storage

A BESS does not automatically provide backup. Operating a site as an island requires a grid-forming inverter, a transfer arrangement, and an explicit decision about which circuits remain energised and which are shed. It is also worth separating the two failure modes: riding through a sub-second disturbance and surviving a multi-hour outage are different problems. For loads that genuinely cannot see a break — control systems, medical and telecom equipment, instrumentation — a dedicated DC-UPS or AC-UPS at the load remains the right instrument, with the storage system sitting behind it rather than replacing it.

Environment, safety and the room it lives in

In a hot climate, thermal management is not an accessory. Cell ageing accelerates with temperature, and a system that derates in the middle of summer is smaller in practice than the one on the datasheet. Siting therefore has to account for ambient conditions, ventilation, clearances, fire detection and suppression strategy, and the separation distances required by the applicable standards and the local authority having jurisdiction. Plan the maintenance and replacement route at design time as well: modules eventually come out, and it is far cheaper to leave a path for them than to discover later that there isn't one.

Read the warranty as an operating envelope

Storage warranties are typically expressed as some combination of cycles, energy throughput, calendar years and a retained-capacity guarantee at end of term — always subject to conditions on temperature, depth of discharge and charge and discharge rate. Those conditions are not fine print; they are the operating envelope you are committing the site to for a decade. If the commercial model requires two cycles a day and the warranty assumes one, that gap will surface as a capacity shortfall long before anyone reopens the contract. Where augmentation is anticipated, it should be designed in from the outset, both electrically and physically.

Then make it observable

Finally, decide what the site will actually see. The battery management system holds cell-level voltage, temperature and balancing data that is invaluable for diagnosing a developing problem — but only if it reaches the energy management system or SCADA in a usable form, with an alarm strategy someone has agreed to act on. A storage asset that reports only state of charge is an asset whose degradation you will learn about from its performance rather than from its data.

If you are evaluating storage for a commercial or industrial site and want a second opinion on sizing, architecture or specification before committing, write to us at info@lmelectro.com. We respond within 24 hours.