Nameplate

What it is (precise)

The nameplate rating is the as-built rating stamped on the equipment label and stated in the datasheet. Build it from the bottom up: cell capacity in ampere-hours times nominal voltage — a 314 Ah LFP cell at 3.2 V stores roughly 1.0 kWh — aggregated through modules, racks, and containers by the series and parallel count.

A modern 20-foot LFP container typically carries a ~5 MWh DC nameplate (older air-cooled units 1-3.5 MWh; high-density designs 5-6+ MWh) on a bus at or below 1,500 VDC, and hundreds of them make up the hundreds of MWh in today's interconnection queues. The rating is fixed at a reference condition: typically 25 C, a stated C-rate, and the full 0-100% state-of-charge range.

Nameplate is a DC, gross, beginning-of-life number, and the three nameplates on a project are not one number in three costumes. The energy nameplate is MWh on the DC battery. The power nameplate on the conversion equipment is Apparent power (kVA or MVA), because a converter or transformer is limited by current and voltage, not by the phase angle between them; Real power in MW exists only at a stated boundary and Power factor, P = S × cos φ.

This is where the AC / DC distinction bites: the DC label does not subtract the energy locked out by the operating SOC window, the one-way conversion loss through the PCS and transformer on the way out, station auxiliary loads, or the capacity cells lose as they age. Those four subtractions turn nameplate into Usable energy and then deliverable energy at the point of interconnection (POI).

Separate nominal energy from BOL energy. Nominal energy is an arithmetic construct — ampere-hour capacity times nominal voltage times cell count — a paper number no meter ever reads, because a real discharge traverses the whole voltage curve rather than sitting at the nominal point. BOL energy is what the commissioning capacity test measures at the stated reference condition.

Nothing forces the two to coincide: a commissioning test can come in a few percent over the label, and whether that overage belongs to the owner or becomes the supplier's free degradation headroom depends on which baseline the capacity warranty names — BOL nameplate or as-measured commissioning capacity. They are different promises.

Why it matters in a real grid-scale project

Contracts and revenue are never written against nameplate; they are written against guaranteed energy at the POI over the life of the asset. A tolling or capacity agreement specifies, say, 100 MW for 4 hours delivered to the grid at end of year 10, 15, or 20.

Because LFP cells fade and auxiliaries draw power, the engineer oversizes the DC nameplate so the worn, derated AC system still clears that contract at end of life. The distance between nameplate and contract energy is the overbuild-and-augmentation margin, and it drives capex, container count, land area, and collection-system design — the gap where most project cost decisions are made.

Confusing nameplate with deliverable capacity is a common and expensive error. A 100 MWh DC nameplate might deliver roughly 85-92 MWh usable AC at BOL and materially less at end of life without augmentation. Lenders size debt against the warranted, degraded number; interconnection studies and grid-code compliance are judged on AC capability at the POI, not the DC label.

Utility-scale projects therefore overbuild DC nameplate by roughly 10-25% above the day-one contract quantity, or plan periodic augmentation instead. Get this stack wrong and it surfaces late as missed performance guarantees, liquidated damages, or a project too small to honor its contract.

Typical values and standards

Translate nameplate to deliverable capacity with a derate stack worth memorizing: the usable SOC window (LFP systems commonly operate across roughly 90% of the nominal range, which defines Usable energy), one-way conversion loss (the discharge path through PCS and transformer is roughly 92-96%, and only that leg stands between nameplate and delivered energy; squared, the same chain gives the familiar 85-92% AC-to-AC round-trip efficiency, which belongs to a full charge-discharge cycle), auxiliary load (often 1-3% of rated power), and calendar-plus-cycle degradation (LFP commonly retains 70-80% of BOL capacity after ~6,000-10,000 cycles, duty dependent).

NMC offers higher energy density but generally fades faster and has a lower thermal-runaway onset temperature, which is why LFP dominates stationary storage.

The AC-side plates carry their own conditions rows. A PCS's kVA is quoted at a reference ambient (often 40-45 C) and altitude (about 1000 m), and it moves with both the AC and the DC voltage — the same hardware carries different numbers at different grid voltages — so vendors publish the rating as a matrix, not one headline figure.

Transformers go further and print several MVA ratings on a single plate, one per cooling stage (ONAN by convection, ONAF with fans, OFAF with pumped oil and fans — a large unit might read 75/100/125 MVA across the three). A rated MVA without its reference voltage, temperature, and altitude is not a rating at all.

The rating is verified by test, not taken on faith. IEC 62933-2-1 defines unit parameters and test methods for energy storage systems, including capacity and round-trip efficiency at stated conditions; factory and site acceptance tests then demonstrate the nameplate at the reference C-rate and temperature.

Nameplate energy also anchors the safety code basis: NFPA 855 sets installation limits, separation distances, and hazard thresholds by stored energy; UL 9540 is the ESS product safety listing; UL 9540A is the thermal-runaway fire-propagation test method whose data feed NFPA 855; UL 1973 covers the battery racks; and NFPA 68/69 govern deflagration venting and prevention.

Keep a few utility-scale anchors handy to sanity-check any quoted figure. Durations of 2-4 hours dominate current interconnection queues, and Duration is simply energy divided by power — a 400 MWh / 100 MW plant is a 4-hour system that cycles at 0.25C, a typical C-rate for energy-shifting nameplates. PCS blocks run about 1-5 MVA each, DC buses sit at or below 1,500 VDC, and container nameplates cluster around 5 MWh (older air-cooled units 1-3.5 MWh; high-density designs 5-6+ MWh). If a vendor's number falls outside these bands, ask why before you trust it.

How it shows up in specs, studies and contracts

On a datasheet, never read the headline number without its footnotes. Check the reference temperature, C-rate, SOC range, whether the figure is DC or AC, and whether it is BOL or a warranted later-year value; two vendors quoting "5.0 MWh" at different C-rates or temperatures are not the same product.

AC-block suppliers quote an AC nameplate at the inverter terminals while DC-block suppliers quote DC at the battery terminals, so comparing the two without converting through PCS efficiency and the required Power factor systematically flatters the DC-block number. Normalize both to net AC at the POI before comparing price per MWh.

Nameplate then takes on a regulatory life of its own. Interconnection queue positions, generator interconnection agreements, and market registrations are filed in nameplate MW (and increasingly MWh), and regulators report fleet statistics on that basis even where the plant is capped below nameplate at the POI.

In the supply stack, the battery purchase agreement warrants a year-by-year capacity-retention table as a percentage of nameplate; the performance guarantee in the tolling or O&M agreement is tested against measured capacity at defined conditions; and the commissioning capacity test is the moment the paper nameplate meets reality.

Carry a short checklist to every kickoff. Which of the three nameplates is on the table, and at which boundary is it measured? At what C-rate, temperature, and SOC window is it defined? Is the quoted figure DC or AC, gross or net of auxiliaries? What usable energy does the vendor warrant at year 1 and at end of term?

What augmentation plan keeps the contract quantity whole? And does the interconnection filing use nameplate or POI-limited values? A one-line answer to each closes the ambiguity behind most sizing disputes, and forces the vendor to state the binding constraint — warranted end-of-life energy at the POI — in writing before money moves.

Common pitfalls

The classic unit error is conflating power and energy: MW is an instantaneous rating, MWh is stored energy, and a 100 MW / 400 MWh nameplate is a 4-hour system, never a 2-hour one. A close second is treating the PCS rating as the plant's real-power nameplate. Converters are sized in Apparent power (MVA), and deliverable Real power depends on the Power factor required at the POI, so a 100 MVA station obligated to hold 0.95 power factor delivers at most 95 MW of real power. Demand more Reactive power support and that headroom shrinks further still.

The subtler failure is the missing conditions row. A PCS rating transplanted from one datasheet row into a study at a different voltage or ambient misstates the plant without anyone typing a wrong digit. If a quoted rating arrives without its temperature, voltage, and power factor attached, the number is incomplete rather than wrong — send it back for its conditions.

Nameplate is also not static on paper even though it is fixed in the hardware. Augmentation adds containers and raises the installed DC nameplate over time; retirements and derates lower the effective figure; and grid operators may cap output below nameplate at the POI. Unlike an EV pack, which is marketed on a single usable-energy figure, a grid-scale BESS spends its entire commercial life in the gap between the DC label and the net AC megawatt-hour that clears the meter — which is why every serious project number is defined relative to nameplate, not equal to it.